CN107608983A

CN107608983A - Heading message optimization method, device, equipment and system

Info

Publication number: CN107608983A
Application number: CN201610543485.8A
Authority: CN
Inventors: 吴夙慧; 陈强; 郭立超; 李传福
Original assignee: Alibaba Group Holding Ltd
Current assignee: Alibaba Group Holding Ltd
Priority date: 2016-07-11
Filing date: 2016-07-11
Publication date: 2018-01-19
Anticipated expiration: 2036-07-11
Also published as: WO2018010557A1; CN107608983B; TW201802738A

Abstract

The application provides a kind of heading message optimization method, device, equipment and system, and this method includes：Determine the target service object set belonging to business object to be optimized；Obtain user characteristics corresponding to popular lexical set corresponding to the target service object set and the business object to be optimized；According to the user characteristics, target hot topic vocabulary is determined in the popular lexical set；According to the target hot topic vocabulary, the heading message of the business object to be optimized is optimized.Recommend the accuracy of business object to user for improving.

Description

Heading message optimization method, device, equipment and system

Technical field

The application is related to Internet technical field, more particularly to a kind of heading message optimization method, device, equipment and is System.

Background technology

At present, can be that business object determines title for the ease of being made a distinction to different business object (such as commodity) Information.

In actual application, when user needs to obtain the business object in data server, user can pass through Search term scans for, to obtain the multiple business objects to match with the search term, specifically：Received in data server After the search term of user's input, data server is generally matched the heading message of search term and each business object, with Obtain the business object to match with search term, and to the business object that user recommends to match.

However, because the heading message of many business objects is inaccurate so that data server is according to business object Heading message recommends accurate business object without normal direction user.

The content of the invention

The application provides a kind of heading message optimization method, device, equipment and system, recommends business for improving to user The accuracy of object.

On the one hand, the application provides a kind of heading message optimization method, including：

Determine the target service object set belonging to business object to be optimized；

Obtain corresponding to popular lexical set corresponding to the target service object set and the business object to be optimized User characteristics；

According to the user characteristics, target hot topic vocabulary is determined in the popular lexical set；

According to the target hot topic vocabulary, the heading message of the business object to be optimized is optimized.

In a kind of possible embodiment, according to the user characteristics, target is determined in the popular lexical set Popular vocabulary, including：

Obtain the matching degree of the user characteristics and each popular keyword in the popular keyword set；

According to the matching degree of each popular keyword set in the user characteristics and the popular keyword set, described Target hot topic vocabulary is determined in popular lexical set.

In alternatively possible embodiment, the user characteristics and each hot topic in the popular keyword set are obtained The matching degree of keyword, including：

Determine the second vocabulary corresponding to the first vocabulary classification corresponding to the user characteristics and each popular keyword Classification；

According to default vocabulary categorical match table, the first vocabulary classification, each second vocabulary classification, it is determined that described The matching degree of data message and each popular keyword, wherein, the default vocabulary categorical match table includes each two Matching degree between vocabulary classification.

In alternatively possible embodiment, user characteristics corresponding to the business object to be optimized is obtained, including：

Determine user to be optimized corresponding to the business object to be optimized；

The user profile of the user to be optimized is obtained, the user profile includes the registration letter of the user to be optimized At least one of breath, the adding description information of the user to be optimized, Transaction Information corresponding to the user to be optimized；

According to the user profile, user characteristics corresponding to the business object to be optimized is determined.

It is described according to target hot topic vocabulary in alternatively possible embodiment, to the business object to be optimized Heading message optimizes, including：

The target hot topic vocabulary is added in the heading message of the business object to be optimized.

It is described that the target hot topic vocabulary is added to the business pair to be optimized in alternatively possible embodiment In the heading message of elephant, including：

Obtain at least one current key word that the heading message includes；

Obtain the weighted value of each current key word and the target hot topic vocabulary；

According to each current key word and the weighted value of the target hot topic vocabulary, determine that the target hot topic vocabulary exists Position in the heading message；

According to position of the target hot topic vocabulary in the heading message, the target hot topic vocabulary is added to institute In the heading message for stating business object to be optimized.

The target hot topic vocabulary is sent to client corresponding to the business object to be optimized, so that user is according to described Target hot topic vocabulary determines the heading message after optimization；

Receive the first confirmation heading message that the client is sent；

Described first confirmation heading message is defined as the heading message after the optimization of the business object to be optimized.

According to the heading message of the business object to be optimized and the target hot topic vocabulary, title letter to be confirmed is determined Breath；

The heading message to be confirmed is sent to client corresponding to the business object to be optimized, so that user is according to institute State heading message to be confirmed and determine the heading message after optimization.

In alternatively possible embodiment, the heading message according to the business object to be optimized and the mesh Popular vocabulary is marked, determines heading message to be confirmed, including：

Determine insertion position of the target hot topic vocabulary in the heading message of the business object to be optimized；

The target hot topic vocabulary is inserted in the insertion position of the heading message, obtains title letter to be confirmed Breath.

In alternatively possible embodiment, treated really to described in client transmission corresponding to the business object to be optimized Recognize heading message, so that user determines the heading message after optimization according to the heading message to be confirmed, including：

The heading message to be confirmed is sent to client corresponding to the business object to be optimized；

Receive the confirmation that the client is sent；

According to the confirmation, after the heading message to be confirmed to be defined as to the optimization of the business object to be optimized Heading message.

The second confirmation heading message that the client is sent is received, described second confirms that heading message is the user couple The title to be confirmed modifies what is obtained；

Described second confirmation heading message is defined as the heading message after the optimization of the business object to be optimized.

In alternatively possible embodiment, the target service object set determined belonging to business object to be optimized Close, including：

Obtain the target classification belonging to business object to be optimized；

Determine multiple business object set to be selected corresponding to the target classification；

The target service object set is determined in the multiple business object set to be selected.

In alternatively possible embodiment, the target classification obtained belonging to business object to be optimized, including：

Judge that the business object to be optimized whether there is corresponding classification；

If so, classification corresponding to the business object to be optimized is then defined as the target classification；

If it is not, the first keyword set corresponding to business object to be optimized is then obtained, and according to first keyword set Close and classification forecast model determines the target classification, wherein, the classification forecast model includes pair of keyword and classification It should be related to and the matching degree of keyword and classification.

It is described to be determined according to first keyword set and classification forecast model in alternatively possible embodiment The target classification, including：

According to the classification forecast model, obtain each keyword is corresponding in first keyword set classification, And matching degree；

Obtain the weighted value of each keyword in first keyword set；

According between the corresponding classification of the weighted value of each keyword, each keyword in first keyword set Matching degree, determine the target classification.

In alternatively possible embodiment, the target service is determined in the multiple business object set to be selected Object set, including：

Judge that the business object to be optimized whether there is corresponding business object set；

If so, business object set corresponding to the business object to be optimized is then defined as the target service object set Close；

If it is not, then obtain the characteristic information of business object to be optimized, each business object set to be selected characteristic information, And the characteristic information of the business object to be optimized is similar to the feature of the characteristic information of each business object set to be selected Degree, and the feature phase of the characteristic information and the characteristic information of each business object set to be selected according to the optimization business object Like degree, target service object set is determined in the business object set to be selected.

In alternatively possible embodiment, the characteristic information according to the optimization business object described is treated with each The characteristic similarity of the characteristic information of business object set is selected, target service object is determined in the business object set to be selected Set, including：

The business object set to be selected that characteristic similarity is more than to first predetermined threshold value is defined as the target service Object set；

Or

The N number of business object set to be selected of characteristic similarity highest is defined as the target service object set, wherein, The N is the positive integer more than or equal to 1.

In alternatively possible embodiment, the characteristic information for obtaining business object to be optimized, including：

Obtain each key in the first keyword set corresponding to business object to be optimized and first keyword set The weighted value of word, according to the weighted value of each keyword in the first keyword set, the feature of the acquisition business object to be optimized Vector, and the characteristic vector of the business object to be optimized is defined as to the characteristic information of the business object to be optimized；

The characteristic information for obtaining the business object set to be selected, including：

The characteristic vector of each business object in the business object set to be selected is obtained, according to the business object collection to be selected The characteristic vector of each business object in conjunction, the central feature vector of the business object set to be selected is determined, and will be described to be selected The central feature vector of business object set is defined as the characteristic information of the business object set to be selected.

In alternatively possible embodiment, the characteristic information for obtaining business object to be optimized and the industry to be selected The characteristic similarity of the characteristic information of business object set, including：

Obtain the business object to be optimized characteristic vector and business object set to be selected central feature vector it is remaining String angle, the characteristic similarity is determined according to the cosine angle.

In alternatively possible embodiment, according to the characteristic information of the optimization business object and each industry to be selected The characteristic similarity of the characteristic information of business object set, determines target service object set in the business object set to be selected Close, including：

Obtain respectively the characteristic vector of the business object to be optimized and the central feature of each business object set to be selected to The cosine angle of amount；

The business object set to be selected that cosine angle is less than to second predetermined threshold value is defined as the target service pair As set, or, the minimum N number of business object set to be selected of cosine angle is defined as the target service object set Close, wherein, angle is that characteristic similarity corresponding to the cosine angle of second predetermined threshold value is equal to first predetermined threshold value.

It is described to obtain the first keyword set corresponding to business object to be optimized in alternatively possible embodiment, Including：

The description information of the business object to be optimized is obtained, the description information includes the business object to be optimized The characteristic parameter of heading message and/or the business object to be optimized；

Word segmentation processing is carried out to the description information of the business object to be optimized, it is corresponding to obtain the business object to be optimized Multiple keywords and each keyword part of speech；

According to the part of speech of each keyword, Screening Treatment is carried out to the multiple keyword, closed with obtaining described first Keyword set.

In alternatively possible embodiment, the target service object set determined belonging to business object to be optimized Before, in addition to：

Obtain Transaction Information corresponding to the business object to be optimized；

It is at least one in following condition to determine that Transaction Information corresponding to the business object to be optimized meets：

The exchange hand in the first preset duration of the business object to be optimized is less than exchange hand threshold value；

The click volume in the second preset duration of the business object to be optimized is less than click volume threshold value.

In alternatively possible embodiment, described true according to first keyword set and classification forecast model Before the fixed target classification, in addition to：

Obtain multiple search strings, the quilt corresponding to each search string that user inputs in the 3rd preset duration Click-to-call service object and it is each it is described be clicked business object be clicked number；

Obtain keyword corresponding to each search string and each classification being clicked belonging to business object；

Industry is clicked according to the multiple search strings inputted in the 3rd preset duration, corresponding to each search string Business object, it is each it is described be clicked business object be clicked number, it is each belonging to keyword and each institute corresponding to search string State and be clicked classification belonging to business object, generate classification forecast model.

In alternatively possible embodiment, it is described according to the multiple search strings inputted in the 3rd preset duration, Be clicked corresponding to each search string business object, it is each it is described be clicked business object be clicked number, it is each belonging to Keyword corresponding to search string and each classification being clicked belonging to business object, generate classification forecast model, bag Include：

Be clicked according to corresponding to each search string business object mark and it is each described in be clicked business pair As the number being clicked, obtain search string and be clicked the first corresponding relation of business object, first corresponding relation Include corresponding to each search string, be clicked business object be clicked number；

According to it is each it is described be clicked business object belonging to classification and first corresponding relation, obtain searching character Second corresponding relation of string and classification, it is corresponding, classification by point that second corresponding relation includes the search string Hit number；

According to keyword corresponding to each search string and second corresponding relation, keyword and class are obtained The corresponding relation of purpose the 3rd, the 3rd corresponding relation includes corresponding to keyword, classification is clicked number；

According to corresponding to each keyword, the matching for being clicked number, generating each keyword classification corresponding with its of classification Degree；

According to the 3rd corresponding relation and the matching degree of keyword and classification, the classification forecast model is generated.

In alternatively possible embodiment, before popular vocabulary corresponding to the acquisition target service object set, Also include：

Affiliated all business objects of classification identical are divided to obtain multiple business object set, each business pair As the attributes similarity of the business object in set is more than the 3rd predetermined threshold value, the multiple business object set includes the mesh Mark business object set；

The second keyword set of each business object set is obtained respectively, obtains multiple second keyword sets；

Obtain the hot value of each keyword in each second keyword set；

According to the hot value of each keyword in each second keyword set, determine that each business object set is corresponding Popular vocabulary.

It is described that affiliated all business objects of classification identical are divided to obtain in alternatively possible embodiment Multiple business object set, including：

3rd keyword set and the 3rd keyword set corresponding to each business object of classification identical belonging to acquisition The weighted value of each keyword in conjunction；

According to the weighted value of each keyword in each 3rd keyword set, obtain the feature of each business object to Amount；

According to the cosine angle between the characteristic vector of each business object, to affiliated all business pair of classification identical As being divided, to obtain the multiple business object set.

On the other hand, the application provides a kind of heading message optimization method, including：

Receive target hot topic vocabulary corresponding to data server is sent, business object to be optimized；

Receive the confirmation that user determines according to the heading message of the business object to be optimized and the target hot topic vocabulary Heading message；

The confirmation heading message is sent to the data server, so that the data server confirms to mark according to described Information is inscribed, the heading message of the business object to be optimized is optimized.

In a kind of possible embodiment, it is described receive user according to the heading message of the business object to be optimized and The confirmation heading message that the target hot topic vocabulary determines, including：

Show the heading message of the business object to be optimized and the target hot topic vocabulary；

Receive the first confirmation heading message that the user inputs according to the heading message and the target hot topic vocabulary.

In alternatively possible embodiment, the reception data server is sent, business object to be optimized corresponds to Target hot topic vocabulary, including：

The heading message to be confirmed that the data server is sent is received, the heading message to be confirmed includes described treat Optimize the heading message of business object and the target hot topic vocabulary.

In alternatively possible embodiment, receive user determined according to the target hot topic vocabulary, it is described treat it is excellent Change and heading message is confirmed corresponding to business object, including：

Show the heading message to be confirmed；

Receive confirmation corresponding to heading message to be confirmed that user inputs, described；

According to the confirmation, the heading message to be confirmed is defined as the confirmation heading message.

Show the heading message to be confirmed；

Receive modification of the user to the heading message input to be confirmed to operate, the modification operation is used to change institute State the vocabulary that position and/or the modification to be confirmed heading message of the target hot topic vocabulary in the heading message include；

Operated according to the modification, determine the second confirmation heading message.

In alternatively possible embodiment, it is described receive target hot topic vocabulary that the data server sends it Before, in addition to：

Receive the heading message of the business object to be optimized of user's input；

The heading message of the business object to be optimized is sent to data server, so that data server is treated according to Optimize popular lexical set and the business object to be optimized corresponding to the target service object set belonging to business object Attribute information, determine target hot topic vocabulary.

Determine at least one business object set that classification includes；

Obtain the second popular vocabulary corresponding to the first popular lexical set corresponding to classification and each business object set Set；

Respectively to the second popular lexical set corresponding to client transmission corresponding to each business object set, to use Family optimizes according to the described second popular lexical set to the heading message of the business object in the business object set；

Into the classification, client corresponding to unfiled business object sends the first popular lexical set, wherein, The unfiled business object is not belonging to any one business object set in the classification, so that user is according to described first Popular lexical set optimizes to the heading message of the unfiled business object.

In a kind of possible embodiment, the second popular lexical set corresponding to the business object set is obtained, is wrapped Include：

Obtain keyword set corresponding to the business object set；

Obtain the hot value of each keyword in the keyword set；

According to the hot value of each keyword in the keyword set, determine second corresponding to each business object set Popular lexical set.

In alternatively possible embodiment, the first popular lexical set corresponding to classification is obtained, including：

Duplicate removal processing is carried out to the popular keyword in each second popular keyword set；

Each second popular keyword set after being handled according to duplicate removal, determines the described first popular lexical set.

In alternatively possible embodiment, sent respectively to client corresponding to each business object set corresponding The second popular lexical set, including：

Determine client corresponding to each business object in the business object set；

Respectively the second popular word corresponding to the business object set is sent to client corresponding to each business object Collect conjunction.

In alternatively possible embodiment, the business is sent to client corresponding to each business object respectively Second popular lexical set corresponding to object set, including：

Obtain user characteristics corresponding to each business object；

According to the user characteristics of each business object, each business pair is determined in the described second popular lexical set As corresponding target hot topic vocabulary；

Respectively to target hot topic vocabulary corresponding to each business object transmission.

In alternatively possible embodiment, into the classification, client corresponding to unfiled business object sends institute The first popular lexical set is stated, including：

Obtain the class classification business object that the classification includes；

Determine client corresponding to each unfiled business object；

Respectively the first popular lexical set is sent to client corresponding to each unfiled business object.

Receive the popular lexical set that data server is sent；

Obtain the heading message of business object corresponding to the popular lexical set；

The confirmation title that user determines according to the heading message and the popular lexical set of the business object is received to believe Breath；

Heading message corresponding to the business object is updated to the confirmation heading message.

In a kind of possible embodiment, the heading message and the heat for receiving user according to the business object The confirmation heading message that door lexical set determines, including：

Show the heading message of the business object and the popular lexical set；

Receive the confirmation heading message that the user inputs according to the heading message and the popular lexical set.

In alternatively possible embodiment, the weighted value of each popular vocabulary is also included in the popular lexical set； Accordingly, it is described according to the popular lexical set, the heading message of the business object is optimized, including：

Show that heading message and the popular lexical set, the popular lexical set of the business object are each described The weighted value of popular vocabulary；

Receive weight of the user according to the heading message, the popular lexical set and each popular vocabulary Value, the confirmation heading message of input.

It is described that the corresponding heading message is updated to the confirmation title letter in alternatively possible embodiment Breath, including：

The mark for confirming heading message and the business object is sent to the data server, so that the data The heading message of the business object is updated to the confirmation heading message by server.

On the other hand, the application provides a kind of heading message optimization device, including：

First determining module, for determining the target service object set belonging to business object to be optimized；

First acquisition module, for obtaining popular lexical set corresponding to the target service object set and described treating Optimize user characteristics corresponding to business object；

Second determining module, for according to the user characteristics, target hot topic word to be determined in the popular lexical set Converge；

Optimization module, for according to the target hot topic vocabulary, being carried out to the heading message of the business object to be optimized Optimization.

In a kind of possible embodiment, second determining module includes first acquisition unit and first and determines list Member, wherein,

The first acquisition unit is used for, and obtains the user characteristics and each popular key in the popular keyword set The matching degree of word；

First determining unit is used for, according to each popular crucial in the user characteristics and the popular keyword set The matching degree of set of words, target hot topic vocabulary is determined in the popular lexical set.

In alternatively possible embodiment, the first acquisition unit is specifically used for：

In alternatively possible embodiment, first acquisition module is specifically used for：

In alternatively possible embodiment, the optimization module is specifically used for：

Obtain at least one current key word that the heading message includes；

In alternatively possible embodiment, the optimization module is specifically additionally operable to：

Receive the first confirmation heading message that the client is sent；

In alternatively possible embodiment, the optimization module includes the second determining unit and transmitting element, wherein,

Second determining unit is used for, according to the heading message of the business object to be optimized and the target hot topic word Converge, determine heading message to be confirmed；

The transmitting element is used for, and sending the title to be confirmed to client corresponding to the business object to be optimized believes Breath, so that user determines the heading message after optimization according to the heading message to be confirmed.

In alternatively possible embodiment, second determining unit is specifically used for：

In alternatively possible embodiment, the optimization module also includes receiving unit and the 3rd determining unit, its In,

The transmitting element is specifically used for, and the mark to be confirmed is sent to client corresponding to the business object to be optimized Inscribe information；

The receiving unit is used for, and receives the confirmation that the client is sent；

3rd determining unit is used for, and according to the confirmation, the heading message to be confirmed is defined as described Heading message after the optimization of business object to be optimized.

In alternatively possible embodiment, the transmitting element is additionally operable to, corresponding to the business object to be optimized Client send the heading message to be confirmed；

The receiving unit is additionally operable to, and receives the second confirmation heading message that the client is sent, described second confirms The heading message user modifies to obtain to the title to be confirmed；

3rd determining unit is additionally operable to, and the described second confirmation heading message is defined as into the business object to be optimized Optimization after heading message.

In alternatively possible embodiment, first determining module includes second acquisition unit, the 4th determination list Member and the 5th determining unit, wherein,

The second acquisition unit is used for, and obtains the target classification belonging to business object to be optimized；

4th determining unit is used for, and determines multiple business object set to be selected corresponding to the target classification；

5th determining unit is used for, and the target service object is determined in the multiple business object set to be selected Set.

In alternatively possible embodiment, the second acquisition unit is specifically used for：

Obtain the weighted value of each keyword in first keyword set；

In alternatively possible embodiment, the 4th determining unit is specifically used for：

Or

In alternatively possible embodiment, the second acquisition unit and the 4th determining unit are specifically used for：

In alternatively possible embodiment, described device also includes the second acquisition module and the 3rd determining module, its In,

Second acquisition module is used for, and the target industry belonging to business object to be optimized is determined in first determining module It is engaged in before object set, obtains Transaction Information corresponding to the business object to be optimized；

3rd determining module is used for, and determines that Transaction Information corresponding to the business object to be optimized meets following condition In it is at least one：

In alternatively possible embodiment, described device also includes the 3rd acquisition module, the 4th acquisition module and life Into module, wherein,

3rd acquisition module is used for, pre- according to first keyword set and classification in the second acquisition unit Before survey model determines the target classification, multiple search strings, each institute that user inputs in the 3rd preset duration are obtained State be clicked corresponding to search string business object and it is each it is described be clicked business object be clicked number；

4th acquisition module is used for, and obtains keyword corresponding to each search string and each described by point Hit the classification belonging to business object；

The generation module is used for, according to inputted in the 3rd preset duration multiple search strings, each search word Be clicked corresponding to symbol string business object, it is each it is described be clicked business object be clicked number, it is each belonging to search string pair The keyword and each classification being clicked belonging to business object answered, generate classification forecast model.

In alternatively possible embodiment, the generation module is specifically used for：

In alternatively possible embodiment, it is true that described device also includes division module, the 5th acquisition module and the 4th Cover half block, wherein,

The division module is used for, and popular vocabulary corresponding to target service object set is obtained in the second acquisition unit Before, affiliated all business objects of classification identical are divided to obtain multiple business object set, each business object The attributes similarity of business object in set is more than the 3rd predetermined threshold value, and the multiple business object set includes the target Business object set；

5th acquisition module is used for, and obtains the second keyword set of each business object set respectively, obtains Multiple second keyword sets, and obtain the hot value of each keyword in each second keyword set；

4th determining module is used for, according to the hot value of each keyword in each second keyword set, it is determined that Popular vocabulary corresponding to each business object set.

In alternatively possible embodiment, the division module is specifically used for：

First receiving module, for receiving target hot topic word corresponding to data server transmission, business object to be optimized Converge；

Second receiving module, for receiving user according to the heading message of the business object to be optimized and target heat The confirmation heading message that door vocabulary determines；

Sending module, for sending the confirmation heading message to the data server, so that the data server According to the confirmation heading message, the heading message of the business object to be optimized is optimized.

In a kind of possible embodiment, described device also includes display module, wherein,

The display module is used for, and shows the heading message of the business object to be optimized and the target hot topic vocabulary；

Accordingly, second receiving module is used for, and receives the user according to the heading message and target heat The first of door vocabulary input confirms heading message.

In alternatively possible embodiment, first receiving module is specifically used for：

In alternatively possible embodiment, the display module is additionally operable to, and shows the heading message to be confirmed；

Accordingly, second receiving module is specifically used for, and it is corresponding to receive heading message to be confirmed that user inputs, described Confirmation, and according to the confirmation, the heading message to be confirmed is defined as the confirmation heading message.

Accordingly, second receiving module is specifically used for, and receives the user and the heading message to be confirmed is inputted Modification operation, the modification operation is used to change position of the target hot topic vocabulary in the heading message and/or repaiies Change the vocabulary that the heading message to be confirmed includes, and operated according to the modification, determine the second confirmation heading message.

In alternatively possible embodiment, second receiving module is additionally operable to, and is connect in first receiving module Before receiving the target hot topic vocabulary that the data server is sent, the title letter of the business object to be optimized of user's input is received Breath；

Accordingly, the sending module is additionally operable to, and the title that the business object to be optimized is sent to data server is believed Breath, so that popular word finder corresponding to target service object set of the data server according to belonging to the business object to be optimized Conjunction and the attribute information of the business object to be optimized, determine target hot topic vocabulary.

Determining module, at least one business object set included for determining classification；

Acquisition module, it is corresponding for obtaining the first popular lexical set corresponding to classification and each business object set The second popular lexical set；

Sending module, it is used for, it is popular to corresponding to client transmission corresponding to each business object set second respectively Lexical set, so that title of the user according to the described second popular lexical set to the business object in the business object set Information optimizes；

The sending module is additionally operable to, and into the classification, client corresponding to unfiled business object sends described first Popular lexical set, wherein, the unfiled business object is not belonging to any one business object set in the classification, with User is set to be optimized according to the described first popular lexical set to the heading message of the unfiled business object.

In a kind of possible embodiment, the acquisition module is specifically used for：

Obtain keyword set corresponding to the business object set；

Obtain the hot value of each keyword in the keyword set；

In alternatively possible embodiment, the acquisition module is specifically used for：

In alternatively possible embodiment, the sending module includes determining unit and transmitting element, wherein,

The determining unit is used for, and determines client corresponding to each business object in the business object set；

The transmitting element is used for, and sends the business object set to client corresponding to each business object respectively Corresponding second popular lexical set.

In alternatively possible embodiment, the transmitting element is specifically used for：

Obtain user characteristics corresponding to each business object；

In alternatively possible embodiment, the sending module is specifically used for：

Determine client corresponding to each unfiled business object；

First receiving module, for receiving the popular lexical set of data server transmission；

Acquisition module, for obtaining the heading message of business object corresponding to the popular lexical set；

Second receiving module, for receiving heading message and the popular lexical set of the user according to the business object The confirmation heading message of determination；

Update module, for heading message corresponding to the business object to be updated into the confirmation heading message.

The display module is used for, and shows the heading message of the business object and the popular lexical set；

Accordingly, second receiving module is used for, and receives the user according to the heading message and the popular word Collect the confirmation heading message for closing input.

In alternatively possible embodiment, the weighted value of each popular vocabulary is also included in the popular lexical set；

Accordingly, the display module is used for, and shows the heading message of the business object and the popular word finder Close, the weighted value of popular each popular vocabulary of lexical set；

Second receiving module is used for, receive the user according to the heading message, the popular lexical set and The weighted value of each popular vocabulary, the confirmation heading message of input.

In alternatively possible embodiment, the update module is specifically used for：

On the other hand, the application provides a kind of data server, including processor and the storage for storing application program Device, the processor is used to read the application program in the memory, and performs following operation：

In a kind of possible embodiment, the processor is specifically used for：

In alternatively possible embodiment, the processor is specifically used for：

In alternatively possible embodiment, the processor is specifically used for, and the target hot topic vocabulary is added to In the heading message of the business object to be optimized.

In alternatively possible embodiment, the processor is specifically used for：

Obtain at least one current key word that the heading message includes；

In alternatively possible embodiment, the processor is specifically used for：

Receive the first confirmation heading message that the client is sent；

In alternatively possible embodiment, the data server also includes COM1, wherein, the processor It is specifically used for：

The title to be confirmed is sent by client corresponding to the COM1 to the business object to be optimized to believe Breath, so that user determines the heading message after optimization according to the heading message to be confirmed.

In alternatively possible embodiment, the processor is specifically used for：

The title to be confirmed is sent by client corresponding to the COM1 to the business object to be optimized to believe Breath；

Receive the confirmation that the client is sent；

In alternatively possible embodiment, the processor is specifically used for：

In alternatively possible embodiment, the processor is specifically used for：Judging the business object to be optimized is It is no corresponding classification to be present；

In alternatively possible embodiment, the processor is specifically used for：

Obtain the weighted value of each keyword in first keyword set；

In alternatively possible embodiment, the processor is specifically used for：

Or

In alternatively possible embodiment, the processor is specifically used for：

In alternatively possible embodiment, the processor is additionally operable to：

Before the target service object set that the processor determines belonging to business object to be optimized, treated described in acquisition excellent Change Transaction Information corresponding to business object；

The target class is determined according to first keyword set and classification forecast model in the second acquisition unit Before mesh, multiple search strings, quilt corresponding to each search string that user inputs in the 3rd preset duration are obtained Click-to-call service object and it is each it is described be clicked business object be clicked number；

In alternatively possible embodiment, the processor is specifically used for：

Before the processor obtains popular vocabulary corresponding to target service object set, to affiliated classification identical institute There is business object to be divided to obtain multiple business object set, the attribute phase of the business object in each business object set It is more than the 3rd predetermined threshold value like degree, the multiple business object set includes the target service object set；

The second keyword set of each business object set is obtained respectively, obtains multiple second keyword sets, and Obtain the hot value of each keyword in each second keyword set；

In alternatively possible embodiment, the processor is specifically used for：Each business of classification identical belonging to acquisition The weighted value of each keyword in 3rd keyword set corresponding to object and the 3rd keyword set；

On the other hand, the application provides a kind of client, including processor, COM1, input equipment and for storing The memory of application program, wherein,

The processor is used for, and data server is sent, business object pair to be optimized is received by the COM1 The target hot topic vocabulary answered；

The input equipment is used for, and receives user according to the heading message of the business object to be optimized and target heat The confirmation heading message that door vocabulary determines；

The processor is additionally operable to, and sending the confirmation title to the data server by the COM1 believes Breath, so that the data server is carried out according to the confirmation heading message to the heading message of the business object to be optimized Optimization.

In a kind of possible embodiment, the client also includes display device, wherein,

The display device is used for, and shows the heading message of the business object to be optimized and the target hot topic vocabulary；

Accordingly, the input equipment is specifically used for, and receives the user according to the heading message and target heat The first of door vocabulary input confirms heading message.

In alternatively possible embodiment, the processor is specifically used for, by described in COM1 reception The heading message to be confirmed that data server is sent, the heading message to be confirmed include the mark of the business object to be optimized Inscribe information and the target hot topic vocabulary.

In alternatively possible embodiment, the display device is additionally operable to, and shows the heading message to be confirmed；

Accordingly, the input equipment is specifically used for, and it is corresponding really to receive heading message to be confirmed that user inputs, described Recognize information；

The processor is specifically used for, and according to the confirmation, the heading message to be confirmed is defined as described true Recognize heading message.

Accordingly, the input equipment is specifically used for, and receives the user and the heading message input to be confirmed is repaiied Change operation, the modification operation is used to change position and/or modification institute of the target hot topic vocabulary in the heading message State the vocabulary that heading message to be confirmed includes；

The processor is specifically used for, and is operated according to the modification, determines the second confirmation heading message.

In alternatively possible embodiment, the processor is additionally operable to, and passes through the communication ends in the processor It is excellent by the COM1 treating of receiving that user inputs before mouthful receiving the target hot topic vocabulary that the data server is sent Change the heading message of business object；

Accordingly, the processor is specifically used for, described to be optimized to data server transmission by the COM1 The heading message of business object, so that target service object set of the data server according to belonging to the business object to be optimized The attribute information of corresponding popular lexical set and the business object to be optimized, determine target hot topic vocabulary.

On the other hand, the application provides a kind of data server, including processor, COM1 and applies journey for storing The memory of sequence, the processor is used to read the application program in the memory, and performs following operation：

Determine at least one business object set that classification includes；

By the COM1 respectively to the second heat corresponding to client transmission corresponding to each business object set Door lexical set, so that mark of the user according to the described second popular lexical set to the business object in the business object set Topic information optimizes；

By the COM1, into the classification, client corresponding to unfiled business object sends first heat Door lexical set, wherein, the unfiled business object is not belonging to any one business object set in the classification, so that User optimizes according to the described first popular lexical set to the heading message of the unfiled business object.

In a kind of possible embodiment, the processor is specifically used for：

Obtain keyword set corresponding to the business object set；

Obtain the hot value of each keyword in the keyword set；

In alternatively possible embodiment, the processor is specifically used for：

The business object set is sent to client corresponding to each business object by the COM1 respectively Corresponding second popular lexical set.

In alternatively possible embodiment, the processor is specifically used for：

Obtain user characteristics corresponding to each business object；

By the COM1 respectively to target hot topic vocabulary corresponding to each business object transmission.

In alternatively possible embodiment, the processor is specifically used for：

Determine client corresponding to each unfiled business object；

First heat is sent to client corresponding to each unfiled business object by the COM1 respectively Door lexical set.

The processor is used for, and the popular lexical set of data server transmission is received by the COM1；

The processor is additionally operable to, and obtains the heading message of business object corresponding to the popular lexical set；

The input equipment is used for, and receives heading message and the popular lexical set of the user according to the business object The confirmation heading message of determination；

The processor is additionally operable to, and heading message corresponding to the business object is updated into the confirmation heading message.

The display device is used for, and shows the heading message of the business object and the popular lexical set；

Accordingly, the input equipment is specifically used for, and receives the user according to the heading message and the popular word Collect the confirmation heading message for closing input.

Accordingly, the display device is used for, and shows the heading message of the business object and the popular word finder Close, the weighted value of popular each popular vocabulary of lexical set；

The input equipment is specifically used for, receive the user according to the heading message, the popular lexical set and The weighted value of each popular vocabulary, the confirmation heading message of input.

In alternatively possible embodiment, the processor is specifically used for：

On the other hand, the application provides a kind of heading message optimization system, including data server and client, wherein,

The data server is used for, and determines the target service object set belonging to business object to be optimized, described in acquisition User characteristics corresponding to popular lexical set corresponding to target service object set and the business object to be optimized, according to institute User characteristics is stated, target hot topic vocabulary is determined in the popular lexical set, and the target hot topic word is sent to client Converge；

The client is used for, and it is popular to receive target corresponding to the data server is sent, business object to be optimized Vocabulary, receive the confirmation title that user determines according to the heading message of the business object to be optimized and the target hot topic vocabulary Information, and the confirmation heading message is sent to the data server, so that the data server confirms to mark according to described Information is inscribed, the heading message of the business object to be optimized is optimized.

The data server is used for, and determines at least one business object set that classification includes, and it is corresponding to obtain classification The first popular lexical set and each business object set corresponding to the second popular lexical set, respectively to each industry Second popular lexical set corresponding to client transmission corresponding to business object set, and the unfiled business object into the classification Corresponding client sends the first popular lexical set；

The client is used for, and receives the popular lexical set that the data server is sent, and obtains the popular vocabulary The heading message of business object corresponding to set, receive heading message and the popular vocabulary of the user according to the business object Gather the confirmation heading message determined, heading message corresponding to the business object is updated to the confirmation heading message.

In this application, when heading message optimization device needs to optimize the heading message of business object to be optimized When, heading message optimization device determines target service object set corresponding to business object to be optimized, and according to business to be optimized User characteristics corresponding to object, target hot topic vocabulary is determined in popular lexical set corresponding to target service object set, and The heading message of business object to be optimized is optimized according to target hot topic vocabulary.In above process, business pair to be optimized As higher with the attributes similarity of the business object in target service object set so that the popular word of target service object set It is higher with the matching degree of business object to be optimized to collect conjunction, further, because target hot topic vocabulary is according to business to be optimized The popular lexical set of user characteristics corresponding to object and target service object set determines what is obtained so that target hot topic word Remittance can preferably embody the characteristic information of business object to be optimized, therefore, can be to industry to be optimized by target hot topic vocabulary The heading message of business object is accurately optimized, and then improves the accuracy that data server recommends business object to user.

Brief description of the drawings

In order to illustrate more clearly about the embodiment of the present invention or technical scheme of the prior art, below will be to embodiment or existing There is the required accompanying drawing used in technology description to be briefly described, it should be apparent that, drawings in the following description are this hairs Some bright embodiments, for those of ordinary skill in the art, without having to pay creative labor, can be with Other accompanying drawings are obtained according to these accompanying drawings.

Fig. 1 is the application scenarios schematic diagram of heading message optimization method provided by the invention；

Fig. 2 is the flow chart one of heading message optimization method provided by the invention；

Fig. 3 is business object management framework figure provided by the invention；

Fig. 4 is the schematic flow sheet of determination target hot topic vocabulary method provided by the invention；

Fig. 5 is the flow chart one provided by the invention according to target hot topic vocabulary to heading message optimization method；

Fig. 6 is the flowchart 2 provided by the invention according to target hot topic vocabulary to heading message optimization method；

Fig. 7 is the flow chart 3 provided by the invention according to target hot topic vocabulary to heading message optimization method；

Fig. 8 is the flow chart of determination target service object set method provided by the invention；

Fig. 9 is the flow chart of acquisition target classification method provided by the invention；

Figure 10 is the flow chart of first keyword set method of acquisition provided by the invention；

Figure 11 is the flow chart provided by the invention that target classification method is determined according to classification forecast model；

Figure 12 is the flow chart of determination target service object set method provided by the invention；

Figure 13 is the flow chart of generation classification forecast model method provided by the invention；

Figure 14 is the flow chart of popular vocabulary method corresponding to generation business object set provided by the invention；

Figure 15 is the flowchart 2 of heading message optimization method provided by the invention；

Figure 16 is the flow chart one provided by the invention for obtaining and confirming heading message method；

Figure 17 is client end interface schematic diagram one provided by the invention；

Figure 18 is the flowchart 2 provided by the invention for obtaining and confirming heading message method；

Figure 19 is client end interface schematic diagram two provided by the invention；

Figure 20 is the flow chart 3 provided by the invention for obtaining and confirming heading message method；

Figure 21 is client end interface schematic diagram three provided by the invention；

Figure 22 is the flow chart 3 of heading message optimization method provided by the invention；

Figure 23 is the flow chart four of heading message optimization method provided by the invention；

Figure 24 is the flow chart five of heading message optimization method provided by the invention；

Figure 25 is the structural representation one that a kind of heading message provided by the invention optimizes device；

Figure 26 is the structural representation two that a kind of heading message provided by the invention optimizes device；

Figure 27 is the structural representation one that another heading message provided by the invention optimizes device；

Figure 28 is the structural representation two that another heading message provided by the invention optimizes device；

Figure 29 is the structural representation one that another heading message provided by the invention optimizes device；

Figure 30 is the structural representation two that another heading message provided by the invention optimizes device；

Figure 31 is the structural representation one that another heading message provided by the invention optimizes device；

Figure 32 is the structural representation two that another heading message provided by the invention optimizes device；

Figure 33 is a kind of structural representation one of data server provided by the invention；

Figure 34 is a kind of structural representation two of data server provided by the invention；

Figure 35 is a kind of structural representation one of client provided by the invention；

Figure 36 is a kind of structural representation two of client provided by the invention；

Figure 37 is the structural representation of another data server provided by the invention；

Figure 38 is the structural representation one of another client provided by the invention；

Figure 39 is the structural representation two of another client provided by the invention；

Figure 40 is the structural representation that a kind of heading message provided by the invention optimizes system；

Figure 41 is the structural representation that another heading message provided by the invention optimizes system.

Embodiment

Here exemplary embodiment will be illustrated in detail, its example is illustrated in the accompanying drawings.Following description is related to During accompanying drawing, unless otherwise indicated, the same numbers in different accompanying drawings represent same or analogous key element.Following exemplary embodiment Described in embodiment do not represent and the consistent all embodiments of the present invention.On the contrary, they be only with it is such as appended The example of the consistent apparatus and method of some aspects being described in detail in claims, of the invention.

Fig. 1 is the application scenarios schematic diagram of heading message optimization method provided by the invention, refers to Fig. 1, including client End 101 and data server 102.The client 101 can be the equipment such as computer, mobile phone.User can input in client 101 The heading message of business object to be optimized, and sent the heading message of business object to be optimized to data by client 101 Server 102.Data server 102 can optimize to the heading message of business object to be optimized, to cause industry to be optimized The heading message of business object is more accurate, and then improves the accuracy that data server recommends business object to user.Below, lead to Specific embodiment is crossed the technical scheme shown in the application is described in detail.

It should be noted that these specific embodiments can be combined with each other below, for same or analogous concept Or process may repeat no more in certain embodiments.

Fig. 2 is the flow chart one of heading message optimization method provided by the invention, refer to Fig. 2, this method can include：

S201, determine target service object set belonging to business object to be optimized；

S202, obtain and used corresponding to popular lexical set corresponding to target service object set and business object to be optimized Family feature；

S203, according to user characteristics, target hot topic vocabulary is determined in popular lexical set；

S204, according to target hot topic vocabulary, the heading message of business object to be optimized is optimized.

The executive agent of the embodiment of the present invention can be that heading message optimizes device, and heading message optimization device can lead to Cross software and/or hardware is realized.Heading message optimization device can be arranged in data server.

For the ease of the understanding to scheme shown in the embodiment of the present invention, first with reference to Fig. 3, device pair is optimized to heading message The way to manage of business object illustrates.

Fig. 3 is business object management framework figure provided by the invention, Fig. 3 is refer to, for the ease of being carried out to business object Management, can be provided with multiple classifications.Each class can include multiple business object set, each business object collection now Conjunction includes multiple business objects, and the attribute information of all business objects is same or similar in a business object set.It is optional , when user needs Added Business object, one can be selected to be matched with the business object by user in multiple classifications Classification, the classification that device selects a matching for the business object automatically can also be optimized by heading message.Certainly, if user and Heading message optimization device is not in time the classification of newly-increased business object selection matching, then the business object is not belonging to any Classification, for example, business object A, business object B, business object O, business object R shown in Fig. 3 etc..In practical application mistake Cheng Zhong, heading message optimization device periodically can enter to the business object in same classification with same or similar attribute information Row, which is sorted out, obtains the multiple business object set of same class now, certainly, when Added Business object, if heading message optimization dress Put and the business object of same class now is not sorted out in time, it is any may make it that the business object of a class now is not belonging to Business object set, for example, the business object T shown in Fig. 3.

In actual application, when heading message optimization device needs the heading message progress to business object to be optimized During optimization, heading message optimization device first in all business object set of data server, determines business object to be optimized Corresponding target service object set.The target service object set can be a business object set, or multiple Business object set.Business object collection in all business object set of data server in addition to target service object set Non-targeted collection of services is combined into, the target service object set and the attributes similarity of business object to be optimized are more than non-targeted industry Business object set and the attributes similarity of business object to be optimized.

After heading message optimization device determines to obtain target service object set corresponding to business object to be optimized, mark Inscribe Advance data quality device and obtain popular lexical set corresponding to target service object set.This is popular, and lexical set can be accurate Reflect the feature of target service object set.Optionally, the popular lexical set can according to user to target service pair Actual search as the business object in set, observe conditions determination.

Heading message optimization device also obtains user characteristics corresponding to business object to be optimized, and the user characteristics is used to represent The characteristic information of user corresponding to business object to be optimized, for example, user characteristics can be to be used corresponding to business object to be optimized The style at family, qualifications and record of service etc..Optionally, heading message optimization device can first determine use to be optimized corresponding to business object to be optimized Family, and obtain the user profile of user to be optimized, the log-on message, to be optimized of user to be optimized can be included in the user profile At least one of Transaction Information corresponding to the adding description information of user, user to be optimized, and according to user profile, it is determined that treating Optimize user characteristics corresponding to business object.

Popular lexical set corresponding to target service object set and to be optimized is acquired in heading message optimization device After user characteristics corresponding to business object, heading message optimizes device according to user characteristics, is determined in popular lexical set Target hot topic vocabulary, and according to target hot topic vocabulary, the heading message of business object to be optimized is optimized.The target is popular Vocabulary can be the popular vocabulary to match with user characteristics.

Below, by specific example, the method shown in Fig. 2 embodiments is described in detail.

Exemplary, it is assumed that the current head information of business object 1 is " spring skirt ", when heading message optimization device needs When being optimized to the heading message of business object 1, heading message optimization device determines target service corresponding to business object 1 Object set, it is assumed that target service object set is business object set 1.

Heading message optimization device obtains the popular lexical set 1 of business object set 1, it is assumed that this is popular lexical set 1 For：" 2016 ", " one-piece dress ", " Korea Spro's version ", " gentlewoman ", " America and Europe ", " individual character ".Heading message optimization device also obtains business object 1 user characteristics, it is assumed that the user characteristics of business object 1 is " individual character ".

Heading message optimization device obtains user characteristics (individual character) phase with business object 1 in popular lexical set 1 The target hot topic vocabulary matched somebody with somebody：" 2016 ", " America and Europe ", " individual character ", and by target hot topic vocabulary (" 2016 ", " America and Europe ", " individual character ") Added in current head information, the heading message after being optimized is：" 2016 American-European individual character of spring skirt ".

, optionally, can be special according to user by following feasible implementation on the basis of embodiment illustrated in fig. 2 Sign, target hot topic vocabulary (S203 in embodiment illustrated in fig. 2) is determined in popular lexical set, specifically, referring to Fig. 4 Illustrated embodiment.

Fig. 4 is the schematic flow sheet of determination target hot topic vocabulary method provided by the invention, refers to Fig. 4, this method can With including：

S401, the matching degree for obtaining user characteristics and each popular keyword in popular keyword set；

S402, the matching degree according to each popular keyword set in user characteristics and popular keyword set, in popular word Collect and target hot topic vocabulary is determined in conjunction.

In the embodiment shown in fig. 4, when heading message optimization device is needed in popular word corresponding to target service object When collecting determination target hot topic vocabulary in conjunction, heading message optimization device first obtains each in user characteristics and popular keyword set The matching degree of popular keyword set；Optionally, heading message optimization device can first determine the first word corresponding to user characteristics Second vocabulary classification corresponding to remittance classification and each popular keyword, and according to default vocabulary categorical match table, the first vocabulary Classification, each second vocabulary classification, the matching degree of data message and each popular keyword is determined, wherein, default vocabulary classification Include the matching degree between each two vocabulary classifications with table.Optionally, the default vocabulary categorical match table can be such as the institute of table 1 Show：

Table 1

Vocabulary classification	Vocabulary classification	Matching degree
			Vocabulary classification 1	Vocabulary classification 2	80%
Vocabulary classification 1	Vocabulary classification 3	68%
			Vocabulary classification 1	Vocabulary classification 4	95%
Vocabulary classification 2	Vocabulary classification 3	60%
			Vocabulary classification 2	Vocabulary classification 4	90%
……	……	……

The matching degree of user characteristics and each popular keyword in popular keyword set is obtained in heading message optimization device Afterwards, according to the matching degree of each popular keyword set in user characteristics and popular keyword set, in popular lexical set Determine target hot topic vocabulary.Optionally, heading message optimization device can will be with user characteristics matching degree highest X popular words Remittance is defined as target hot topic vocabulary, wherein, X is the positive integer more than or equal to 1, can be according to reality in actual application Border needs to set the size of the X.Certainly, heading message optimization device can also will be more than predetermined threshold value with user characteristics matching degree Popular vocabulary be defined as target hot topic vocabulary.

On the basis of any one above-mentioned embodiment, heading message optimization device determine to obtain target hot topic vocabulary it Afterwards, the heading message of business object to be optimized can be carried out by a variety of feasible implementations excellent.Below, Fig. 5-figure is passed through Embodiment shown in shown in 7 introduces three kinds of feasible implementations, wherein, in the embodiment shown in fig. 5, heading message optimization Device directly optimizes to heading message；In the embodiment shown in Fig. 6-Fig. 7, heading message optimize device by with client End interacts, and realizes and heading message is optimized；Specifically, refer to the embodiment shown in Fig. 5-Fig. 6.

Fig. 5 is the flow chart one provided by the invention according to target hot topic vocabulary to heading message optimization method, is referred to Fig. 5, this method can include：

S501, obtain at least one current key word that heading message includes；

S502, the weighted value for obtaining each current key word and target hot topic vocabulary；

S503, according to each current key word and the weighted value of target hot topic vocabulary, determine that target hot topic vocabulary is believed in title Position in breath；

S504, the position according to target hot topic vocabulary in heading message, target hot topic vocabulary is added to industry to be optimized It is engaged in the heading message of object.

In the embodiment shown in fig. 5, target hot topic vocabulary is added to business to be optimized by heading message optimization device In the heading message of object, specifically：

Heading message optimization device obtains at least one current key that the heading message of business object to be optimized includes Word, optionally, heading message optimization device word segmentation processing can be carried out to the heading message of business object to be optimized with obtain to A few current key word.

Heading message optimization device also obtains the weighted value of each current key word and target hot topic vocabulary, and according to each current The weighted value of keyword and target hot topic vocabulary, determine position of the target hot topic vocabulary in heading message.Optionally, title is believed Breath optimization device can according to the hot value of each current key word and target hot topic vocabulary, it is determined that corresponding weighted value, wherein, heat Angle value is higher, then weighted value is higher.Optionally, the weighted value of target hot topic vocabulary is higher, and target hot topic vocabulary is in heading message In position it is more forward.

Below, by specific example, the method shown in Fig. 5 embodiments is described in detail.

Exemplary, it is assumed that the heading message of business object to be optimized is " spring skirt surplus ", it is further assumed that target is popular Vocabulary is：" 2016 ", " individual character ".

Heading message optimizes device and carries out word segmentation processing to heading message " spring skirt surplus ", obtains current key word " spring ", " skirt ", " surplus ", heading message optimization device obtain the weight of each current key word and each target hot topic vocabulary Value, specifically, as shown in table 2：

Table 2

Current key word or target hot topic vocabulary	Weighted value
		Spring	7
Skirt	6
		Surplus	4
2016	8
		Individual character	5

The weight of each current key word and each target hot topic vocabulary shown in table 2 is acquired in heading message optimization device After value, heading message optimization device determines a position of the target hot topic vocabulary in heading message, and optionally, heading message is excellent The order according to weighted value from high to low is put in makeup, and the priority position of current key word and target hot topic vocabulary is arranged successively Sequence, the heading message after being optimized are " the skirt individual character surplus of 2016 spring ".

In actual application, when data server matches to the heading message of search term and business object, Data server generally according to the vertical order of each keyword in heading message, is matched with search term successively；Pass through Embodiment shown in Fig. 5, heading message optimizes device according to each current key word and the weighted value of target hot topic vocabulary, by target Popular vocabulary is added in the relevant position of heading message so that the heading message after optimization is more accurate, further to improve Data server recommends the accuracy of business object to user.

Fig. 6 is the flowchart 2 provided by the invention according to target hot topic vocabulary to heading message optimization method, is referred to Fig. 6, this method can include：

S601, to corresponding to business object to be optimized client send target hot topic vocabulary so that user according to target heat Door vocabulary determines the heading message after optimization；

S602, receive the first confirmation heading message that client is sent；

S603, by first confirmation heading message be defined as the heading message after the optimization of business object to be optimized.

In the embodiment shown in fig. 6, after heading message optimization device determines to obtain target hot topic vocabulary, title letter Client corresponding to breath optimization device to business object to be optimized sends target hot topic vocabulary, so that user is according to target hot topic word The heading message converged after determining optimization.

After user determines to obtain the first confirmation heading message in client according to target hot topic vocabulary, pass through client First is sent to heading message optimization device and confirms heading message, and heading message optimizes device and confirms that heading message determines by first For the heading message after the optimization of business object to be optimized.

In above process, heading message optimization device sends target hot topic vocabulary to client, by user in client End determines the heading message after optimization according to target hot topic vocabulary, improves the flexibility optimized to heading message.

Fig. 7 is the flow chart 3 provided by the invention according to target hot topic vocabulary to heading message optimization method, is referred to Fig. 7, this method can include：

S701, heading message and target hot topic vocabulary according to business object to be optimized, determine heading message to be confirmed；

S702, to client corresponding to business object to be optimized heading message to be confirmed is sent, so that user is according to treating really Recognize heading message and determine the heading message after optimization.

In the embodiment shown in fig. 7, after heading message optimization device determines to obtain target hot topic vocabulary, title letter Breath optimization device determines heading message to be confirmed, had first according to the heading message and target hot topic vocabulary of business object to be optimized Body, set the goal insertion position of the popular vocabulary in the heading message of business object to be optimized, in the insertion position of heading message Middle insertion target hot topic vocabulary is put, obtains heading message to be confirmed.It should be noted that heading message optimization device can basis Method shown in Fig. 5 embodiments is determined to obtain the heading message to be confirmed, and the present invention is no longer repeated this.

After heading message optimization device determines to obtain heading message to be confirmed, heading message optimizes device to be optimized Client corresponding to business object sends heading message to be confirmed, so that after user determines optimization according to heading message to be confirmed Heading message.Optionally, after client receives heading message to be confirmed, user can be carried out to heading message to be confirmed Confirm, heading message to be confirmed can also be modified, specifically：

If user approves to the heading message to be confirmed, device can be optimized to heading message by client and sent really Recognize information, so that heading message optimizes device according to the confirmation, heading message to be confirmed is defined as business pair to be optimized Heading message after the optimization of elephant.

If user does not approve to the heading message to be confirmed, user can modify to obtain to heading message to be confirmed Second confirms heading message, and the second confirmation heading message is sent to heading message and optimizes device, so that heading message optimizes Second confirmation heading message is defined as the heading message after the optimization of business object to be optimized by device.

In above process, heading message optimization device first determines heading message to be confirmed according to target hot topic vocabulary, and The heading message after optimization is determined according to the heading message to be confirmed by user, further increases and heading message is optimized Flexibility.

On the basis of above-described embodiment, determine to obtain target service object set to improve heading message optimization device Speed, target service object set (Fig. 2 corresponding to business object to be optimized can be determined by following feasible implementation S201 in illustrated embodiment), specifically, embodiment shown in Figure 8.

Fig. 8 is the flow chart of determination target service object set method provided by the invention, refers to Fig. 8, this method can With including：

Target classification belonging to S801, acquisition business object to be optimized；

S802, determine multiple business object set to be selected corresponding to target classification；

S803, target service object set is determined in multiple business object set to be selected.

In the embodiment shown in fig. 8, heading message optimizes device in order to corresponding to quick obtaining business object to be optimized Target service object set, heading message optimization device can first obtain the target classification belonging to business object to be optimized, and obtain Take multiple business object set to be selected corresponding to target classification.Optionally, classification can be set corresponding with business object set Relation, so that heading message optimization device multiple according to corresponding to target classification and the corresponding relation determine target classification can be treated Select business object set.After heading message optimization device determines to obtain multiple business object set to be selected, heading message is excellent The determination target service object set in multiple business object set to be selected is put in makeup, specifically, heading message optimizes device root According to the attributes similarity of each business object set to be selected in multiple business object set to be selected and target service object set, In multiple business object set to be selected, by the one or more business object collection higher with target service object properties similarity Conjunction is defined as target service object set, optionally, can be by the attributes similarity highest N (N with business object set to be selected For the positive integer more than or equal to 1) individual business object set is defined as target service object set, or, will be with business to be selected The business object set that the attributes similarity of object set is more than the first predetermined threshold value is defined as target service object set.

In above process, the target classification belonging to business object to be optimized is first obtained, and more corresponding to target classification Target service object set is determined in individual business object set to be selected, reduces heading message optimization device selection target business pair As the scope of set, and then improve the speed that heading message optimization device determines target service object set.

On the basis of embodiment illustrated in fig. 8, optionally, heading message optimization device can pass through following feasible realization Mode obtains the target classification (S801 in embodiment illustrated in fig. 8) belonging to business object to be optimized, specifically, refer to Fig. 9 Shown embodiment.

Fig. 9 is the flow chart of acquisition target classification method provided by the invention, refer to Fig. 9, this method can include：

S901, judge that business object to be optimized whether there is corresponding classification；

If so, then perform S902；

If it is not, then perform S903-S904；

S902, classification corresponding to business object to be optimized is defined as target classification；

S903, obtain the first keyword set corresponding to business object to be optimized；

S904, target classification determined according to the first keyword set and classification forecast model, wherein, in classification forecast model The matching degree of corresponding relation and keyword and classification including keyword and classification.

It can be seen from the management framework figure of business object shown in Fig. 3, partial service object has been classified to the class belonging to it Now, and partial service object is not classified to the class belonging to it now, therefore, when it needs to be determined that belonging to business object to be optimized Target classification when, heading message optimization device first determine whether that business object to be optimized whether there is corresponding classification：

If in the presence of illustrating that business object to be optimized has been classified to the classification belonging to it, then heading message optimization device can So that classification corresponding to business object to be optimized directly is defined as into target classification.

If being not present, heading message optimization device needs to obtain the first keyword set corresponding to business object to be optimized With default classification forecast model, include the corresponding relation and keyword of keyword and classification in the classification forecast model With the matching degree of classification, keyword and the matching degree of classification refer to that the corresponding business object of the keyword is divided to such now Probability, a keyword can correspond to one or more classifications, and a classification can also correspond to one or more keywords, can Choosing, classification forecast model can be as shown in table 3：

Table 3

After heading message optimization device acquires the first keyword set and classification forecast model, closed according to first Keyword set and classification forecast model determine target classification corresponding to business object to be optimized.

In order to which embodiment illustrated in fig. 9 is further elaborated, below, respectively by Figure 10-Figure 11 shown in reality Example is applied to obtaining the first keyword set corresponding to business object to be optimized and being predicted according to the first keyword set and classification Model determines that target class purpose process is described in detail corresponding to business object to be optimized.

Figure 10 is the flow chart of first keyword set method of acquisition provided by the invention, refer to Figure 10, this method can With including：

S1001, the description information for obtaining business object to be optimized, the title that description information includes business object to be optimized are believed The characteristic parameter of breath and/or business object to be optimized；

S1002, the description information to business object to be optimized carry out word segmentation processing, obtain corresponding to business object to be optimized The part of speech of multiple keywords and each keyword；

S1003, the part of speech according to each keyword, Screening Treatment is carried out to multiple keywords, to obtain the first keyword set Close.

In the embodiment shown in fig. 10, when heading message optimization device needs to obtain corresponding to business object to be optimized the During one keyword set, heading message optimization device obtains the business object description information of business object to be optimized, the business pair Include the heading message of business object to be optimized and/or the characteristic parameter of business object to be optimized as description information, wherein, treat Optimize business object characteristic parameter be business object to be optimized attribute information, such as business object to be optimized brand, treat it is excellent The production firm of change business object, place of production of business object to be optimized etc., this feature parameter can be by business object pair to be optimized The user answered provides.

After heading message optimization device acquires the description information of business object to be optimized, heading message optimization dress Put and word segmentation processing carried out to the description information of business object to be optimized, obtain multiple keywords corresponding to business object to be optimized, And the part of speech of each keyword, the part of speech include noun, adjective, adverbial word, auxiliary word etc., then, heading message optimization device root Screening Treatment is carried out to the plurality of keyword according to the part of speech of each keyword, the first keyword set is obtained, optionally, to multiple When keyword is screened, it is noun or adjectival keyword that can retain part of speech, deletes the pass that part of speech is auxiliary word or adverbial word Keyword.

Exemplary, it is assumed that the description information of business object to be optimized is " spring very comfortable one-piece dress ", when title is believed When breath optimization device needs to obtain the first keyword set of the business object to be optimized, heading message optimization device is first retouched to this State information and carry out word segmentation processing, obtain the part of speech of following multiple keywords and each keyword：It is " spring (noun) ", " very (secondary Word) ", " comfortable (adjective) ", " (auxiliary word) ", " one-piece dress (noun) ", it is noun that heading message optimization device, which retains part of speech, Or adjectival keyword and keyword of the part of speech for adverbial word or auxiliary word is deleted, obtain the first keyword set：" spring ", " relax It is suitable ", " one-piece dress ".

Figure 11 is the flow chart provided by the invention that target classification method is determined according to classification forecast model, refer to figure 11, this method can include：

S1101, according to classification forecast model, obtain each keyword is corresponding in the first keyword set classification and Matching degree；

S1102, the weighted value for obtaining each keyword in the first keyword set；

S1103, according between the corresponding classification of the weighted value of each keyword, each keyword in the first keyword set Matching degree, determine target classification.

In the embodiment shown in fig. 11, obtain business object to be optimized first is needed to close in heading message optimization device After keyword set and classification forecast model, when heading message optimization device is needed according to the first keyword set and classification When forecast model determines the target classification of business object to be optimized, heading message optimizes device according to classification forecast model, obtains The corresponding classification of each keyword and the matching degree between classification in first keyword set, then obtain first and close The weighted value of each keyword in keyword set, optionally, it can be obtained according to TF-IDF statistical methods in the first keyword set The weighted value of each keyword, it is, of course, also possible to the weighted value of each keyword in the first keyword set is obtained by other means, And according to the matching degree between the corresponding classification of the weighted value of each keyword, each keyword in the first keyword set, really Set the goal classification, optionally, weighted value can be multiplied by into matching degree, and the maximum classification of the product of weighted value and matching degree is determined For target classification.

Exemplary, it is assumed that first keyword set is combined into corresponding to business object to be optimized：" spring ", " comfortable ", " even clothing Skirt ", weighted value corresponding to each keyword and classification forecast model are as shown in table 4：

Table 4

Heading message optimizes weighted value and classification forecast model corresponding to each keyword of the device according to table 4, really The product of weighted value and matching degree corresponding to fixed each classification is as shown in table 5：

Table 5

Classification	The product of weighted value and matching degree
		Overcoat	0.4*0.9=0.36
Skirt	0.40.8+0.91=1.22
		Shoes	0.4*0.9=0.36
Trousers	0.4*0.8=0.32

Heading message optimizes the product of weighted value and matching degree corresponding to each classification of the device according to table 5, and product is most Skirt corresponding to (1.22) is defined as target classification greatly.

It should be noted that table 5 simply illustrates the corresponding relation of keyword and classification in exemplary fashion, it is not to closing The restriction of keyword and classification corresponding relation.

On the basis of Fig. 8-Figure 11 any embodiments, optionally, heading message optimization device can be by following feasible Implementation determine target service object set (in embodiment illustrated in fig. 8 in multiple business object set to be selected S803), specifically, refer to embodiment illustrated in fig. 12.

Figure 12 is the flow chart of determination target service object set method provided by the invention, refers to Figure 12, this method It can include：

S1201, judge that business object to be optimized whether there is corresponding business object set；

If so, then perform S1202；

If it is not, then perform S1203-S1204；

S1202, business object set corresponding to business object to be optimized is defined as target service object set；

S1203, the attribute information for obtaining business object to be optimized, each business object set to be selected attribute information and The attributes similarity of the attribute information of business object to be optimized and the attribute information of each business object set to be selected；

S1204, the attribute according to the attribute information of attribute information and each business object set to be selected of optimization business object Similarity, target service object set is determined in business object set to be selected.

It can be seen from the management framework figure of business object shown in Fig. 3, partial service object has been classified to the industry belonging to it It is engaged in object set, and partial service object is not classified into the business object set belonging to it, therefore, when it needs to be determined that treating When optimizing the target service object set belonging to business object, heading message optimization device first determines whether that business object to be optimized is It is no corresponding business object set to be present：

If in the presence of illustrating that business object to be optimized has been classified into the business object set belonging to it, then heading message Business object set corresponding to business object to be optimized directly can be defined as target service object set by optimization device.

If being not present, heading message optimization device needs to obtain the attribute information of business object to be optimized, each industry to be selected The attribute information of business object set and the attribute information of business object to be optimized and the attribute of each business object set to be selected are believed The attributes similarity of breath, and the category of the attribute information and the attribute information of each business object set to be selected according to optimization business object Property similarity, in business object set to be selected determine target service object set.

Optionally, the attribute information of business object to be optimized can be represented by characteristic vector, accordingly, can be passed through Following feasible implementation obtains the attribute information of business object to be optimized：Obtain the first pass corresponding to business object to be optimized The weighted value of each keyword in keyword set and the first keyword set, according to each keyword in the first keyword set Weighted value, the characteristic vector of business object to be optimized is obtained, and the characteristic vector of business object to be optimized is defined as to be optimized The attribute information of business object.

For example, it is assumed that the keyword and the weighted value of each keyword that the first keyword set includes are as shown in table 6：

Table 6

Keyword	Weighted value
		Spring	0.4
Comfortably	0.4
		One-piece dress	0.9

Heading message optimize device according to table 6, it may be determined that the characteristic vector of business object to be optimized for (0.4,0.4, 0.9)。

Optionally, the attribute information of business object set to be selected, accordingly, Ke Yitong can be represented by characteristic vector Cross the attribute information that following feasible implementation obtains business object set to be selected：Obtain each industry in business object set to be selected The characteristic vector of business object, according to the characteristic vector of each business object in business object set to be selected, determines business object to be selected The central feature vector of set, and the central feature vector of business object set to be selected is defined as business object set to be selected Attribute information, optionally, heading message optimization device can carry out alignment addition to the characteristic vector of each business object and be averaged Value, to obtain the central feature of business object set to be selected vector.

For example, it is assumed that including 3 business objects in business object set to be selected, business object 1- business pair is designated as respectively As 3, it is assumed that the characteristic vector of business object 1- business objects 3 is as shown in table 7：

Table 7

The mark of business object	Characteristic vector
		Business object 1	(0.5,0.45,0.7)
Business object 2	(0.3,0.5,0.8)
		Business object 3	(0.4,0.3,0.8)

Heading message optimizes the characteristic vector of each business object of the device according to table 7, determines business object to be selected The central feature vector of set isThat is central feature vector is (0.4,0.42,0.77).

In above process, when the characteristic vector by business object to be optimized represents that the attribute of business object to be optimized is believed Breath, by the attribute information of the central feature vector representation business object set to be selected of business object set to be selected when, accordingly, Can be according to the characteristic vector of business object to be optimized and the cosine angle of the central feature vector of business object set to be selected, really Fixed business object to be optimized and the attributes similarity of business object set to be selected, specifically, cosine angle and attributes similarity into Inverse ratio, cosine angle is smaller, and attributes similarity is bigger.

Certainly, when representing the attribute information of business object to be optimized by the characteristic vector of business object to be optimized, pass through , can also direct basis during the attribute information of the central feature vector representation business object set to be selected of business object set to be selected The characteristic vector of business object to be optimized and the cosine angle of the central feature vector of each business object set to be selected, multiple Target service object set is determined in business object set to be selected, specifically：Heading message optimization device obtains to be optimized respectively The characteristic vector of business object and the cosine angle of the central feature vector of each business object set to be selected, cosine angle is less than The business object set to be selected of second predetermined threshold value is defined as target service object set, or, by the N number of of cosine angle minimum Business object set to be selected is defined as target service object set, wherein, angle is that the cosine angle of the second predetermined threshold value is corresponding Attributes similarity be equal to the first predetermined threshold value.

On the basis of any one above-mentioned embodiment, heading message optimization device can periodically perform such scheme The heading message of business object is optimized with realizing, certainly, in actual application, in order to save heading message optimization The resource of device, the heading message that heading message optimization device can be only relatively low to the degree of accuracy optimize, optionally, title letter Breath optimization device can judge the degree of accuracy of the heading message of the business object according to the Transaction Information of business object, specifically： Before heading message optimizes device to business object progress heading message optimization, the Transaction Information of business object is first obtained, is sentenced Disconnected Transaction Information meets at least one in following condition：The exchange hand in the first preset duration of business object, which is less than, to strike a bargain Threshold value is measured, the click volume in the second preset duration of business object is less than click volume threshold value, if the Transaction Information of business object Meet any one in above-mentioned condition, illustrate that the heading message degree of accuracy of the business object is poor, then pass through the above method again Technical scheme in embodiment optimizes to the heading message of the business object.

In above process, heading message optimization device business object information only relatively low to the degree of accuracy optimizes, and subtracts Lack the workload of heading message optimization device, and then improve the operating efficiency of heading message optimization device.

On the basis of any one above-mentioned embodiment, heading message optimization device can periodically generate classification prediction Model, or, maintenance periodically is updated to the classification forecast model generated, below, pass through the implementation shown in Figure 13 Example, the process for generating classification forecast model is described in detail.

Figure 13 is the flow chart of generation classification forecast model method provided by the invention, refer to Figure 13, and this method can be with Including：

S1301, obtain multiple search strings that user inputs in the 3rd preset duration, corresponding to each search string It is clicked business object and be respectively clicked business object is clicked number；

S1302, obtain keyword corresponding to each search string and be respectively clicked classification belonging to business object；

S1303, it is clicked according to the multiple search strings inputted in the 3rd preset duration, corresponding to each search string Business object, be respectively clicked business object be clicked number, it is each belonging to keyword corresponding to search string and respectively by point The classification belonging to business object is hit, generates classification forecast model.

In the embodiment shown in fig. 13, when heading message optimization device needs to generate classification forecast model, heading message Optimization device obtains multiple search strings that user inputs in the 3rd preset duration, is clicked corresponding to each search string Business object and the number that is clicked for being respectively clicked business object, then, heading message optimization device obtain each search word Accord with keyword corresponding to string and be respectively clicked the classification belonging to business object；Wherein, search string is user in searching service The search information inputted during object, such as search string can be " spring one-piece dress " etc., and search string is inputted in user Afterwards, data server recommends multiple business object information to user, and user opens in multiple business object information to be checked in detail The business object of feelings information is that number is clicked corresponding to search string, for example, user is in input search string " spring After one-piece dress ", data server recommends business object 1- business objects 100 to user, if user is to business object 2 and business The detail information of object 3 is checked that then business object 2 and business object 3 are corresponding for search string " spring one-piece dress " Be clicked business object, the search string and search string pair that heading message optimization device inputs to all users That answers is clicked business object, can be respectively clicked the number that business object is clicked.

After heading message optimization device acquires above- mentioned information, heading message optimizes device according to each searching character The mark of business object is clicked corresponding to string and is respectively clicked the number that business object is clicked, obtains search string With the first corresponding relation for being clicked business object, the first corresponding relation includes corresponding to each search string, is clicked industry Business object is clicked number.

Exemplary, first corresponding relation can be as shown in table 8：

Table 8

After heading message optimization device acquires the first corresponding relation, heading message optimizes device according to each by point The classification and the first corresponding relation belonging to business object are hit, obtains the second corresponding relation of search string and classification, second Corresponding relation includes corresponding to search string, classification is clicked number, specifically：Most of business object has its institute The classification of category, after heading message optimization device acquires the first corresponding relation, heading message optimization device obtains first Classification in corresponding relation belonging to each business object, and according to it is individual be clicked business object belonging to classification and first corresponding close System, obtains the second corresponding relation.

Exemplary, on the basis of table 8, it is assumed that classification corresponding to each business object is as shown in table 9：

Table 9

Heading message optimizes the corresponding relation that is clicked business object and classification of the device according to table 9, can obtain The second corresponding relation shown in table 10：

Table 10

After heading message optimization device acquires the second corresponding relation, heading message optimizes device according to each search Keyword corresponding to character string and the second corresponding relation, obtain the 3rd corresponding relation of keyword and classification, the 3rd corresponding pass System includes corresponding to keyword, classification is clicked number, and optionally, heading message optimization device can be implemented by Figure 10 Mode shown in example obtains keyword corresponding to search string.

Exemplary, on the basis of table 10, it is assumed that it is as shown in table 11 that search string corresponds to keyword：

Table 11

Heading message optimizes the corresponding relation of search string and keyword of the device according to table 11, can obtain table The 3rd corresponding relation shown in 12：

Table 12

After heading message optimization device acquires the 3rd corresponding relation, heading message optimizes device according to each key Word is corresponding, the matching degree for being clicked number, generating each keyword classification corresponding with its of classification, and according to the 3rd corresponding pass System and the matching degree of keyword and classification, classification forecast model is generated, wherein, the matching degree of keyword and classification refers to close Business object corresponding to keyword belongs to such purpose probability.

It is exemplary, on the basis of table 12, the corresponding class of each keyword that heading message optimization device obtains Purpose is matched to as shown in table 13：

Table 13

Heading message optimizes device according to table 13, it is determined that obtained classification forecast model is as shown in table 14：

Table 14

In above process, heading message optimization device generates classification forecast model according to the practical operation of user so that Heading message optimizes device according to corresponding to classification forecast model can obtain business object to be optimized, accurate classification.

It should be noted that heading message optimization device generates the process of above-mentioned classification forecast model, it is excellent with heading message It is two independent processes that the process optimized to the heading message of business object to be optimized is put in makeup.Generate classification prediction mould The process of type can be before being optimized to heading message any time perform, the present invention to heading message optimization device life It is not especially limited at the time of into above-mentioned classification forecast model.

On the basis of any one above-mentioned embodiment, heading message optimization device can periodically generate electric business platform In popular vocabulary corresponding to each business object set, or, periodically to popular word corresponding to each business object set Remittance is updated maintenance, below, by the embodiment shown in Figure 14, each business object collection is generated to heading message optimization device The process of popular vocabulary is described in detail corresponding to conjunction.

Figure 14 is the flow chart of popular vocabulary method corresponding to generation business object set provided by the invention, refer to figure 14, this method can include：

S1401, affiliated all business objects of classification identical are divided to obtain multiple business object set, each business The attributes similarity of business object in object set is more than the 3rd predetermined threshold value, and the plurality of business object set includes target industry Business object set；

S1402, the second keyword set for obtaining each business object set respectively, obtain multiple second keyword sets；

S1403, the keyword in each second keyword set is counted respectively, to obtain each second keyword set In each keyword hot value；

S1404, the hot value according to each keyword in each second keyword set, are determined corresponding to each business object set Popular vocabulary.

In the embodiment shown in fig. 14, when heading message optimization device needs to obtain corresponding to each business object set During popular vocabulary, to a class, all business objects divide heading message optimization device now first, obtain multiple business Object set, specifically, heading message optimization device can be obtained first the 3rd corresponding to each business object of affiliated classification identical The weighted value of each keyword in keyword set and the 3rd keyword set, according to each key in each 3rd keyword set The weighted value of word, the characteristic vector of each business object is obtained, it is then remaining between the characteristic vector of acquisition each two business object String angle, and according to the cosine angle between the characteristic vector of each two business object, to affiliated all business of classification identical Object is divided, optionally, can be by the multiple industry of cosine angle within a preset range to obtain multiple business object set Business object is divided into a business object set, wherein, obtain shown in mode and Figure 12 of the characteristic vector of each business object It is identical that the mode of the characteristic vector of business object to be optimized is obtained in embodiment, is no longer repeated herein.

After heading message optimization device acquires multiple business object set, heading message optimization device obtains each Second keyword set corresponding to individual business object set, to obtain multiple second keyword sets, for each the second pass Keyword set, heading message optimize device and carry out TF-IDF statistics to the keyword in the second keyword set, to obtain each the The hot value of each keyword in two keyword sets, and according to the hot value of each keyword in the second keyword set, it is determined that Popular vocabulary corresponding to each business object set, wherein, importance of the higher keyword of hot value to business object set It is higher, it more can accurately reflect the feature of business object set, optionally, heading message optimization device can be by hot value M (M is the positive integer more than or equal to 1) the individual keyword being most somebody's turn to do is defined as the popular vocabulary of business object set, or, title Hot value can be defined as the popular vocabulary of business object set by Advance data quality device more than the keyword of preset heat threshold value.

In above process, heading message optimization device passes through the pass in the second keyword set to business object set Keyword carries out TF-IDF statistics, and the popular vocabulary of each business object set acquired is that business object set is relatively attached most importance to It is wanting, the vocabulary of a business object set feature can be reflected, and then ensure that the popular vocabulary of each business object set Accuracy.

It should be noted that heading message optimization device obtains the popular vocabulary process of each business object set, with mark The process that topic Advance data quality device optimizes to the heading message of business object to be optimized is two independent processes.Obtain each The popular vocabulary process of individual business object set can be before being optimized to heading message any time perform, the present invention It is not especially limited at the time of obtaining the popular vocabulary of each business object set to heading message optimization device.

Below, using client as executive agent, heading message optimization method is described in detail.

Figure 15 is the flowchart 2 of heading message optimization method provided by the invention, refers to Figure 15, this method can wrap Include：

S1501, receive target hot topic vocabulary corresponding to data server is sent, business object to be optimized；

S1502, receive the confirmation mark that user determines according to the heading message and target hot topic vocabulary of business object to be optimized Inscribe information；

S1503, confirmation heading message is sent to data server, so that data server is according to confirmation heading message, it is right The heading message of business object to be optimized optimizes.

In the embodiment shown in fig. 5, when user needs to optimize the heading message of business object to be optimized, use Family can input the heading message of business object to be optimized in the client, client is sent industry to be optimized to data server The heading message of business object.Data server can be according to corresponding to the target service object set belonging to business object to be optimized The attribute information of popular lexical set and business object to be optimized, target hot topic vocabulary is determined, and target is sent to client Popular vocabulary.

After client receives target hot topic vocabulary corresponding to data server transmission, business object to be optimized, The target hot topic vocabulary that client receives to user's displaying so that user can determine to confirm title according to target hot topic vocabulary Information.After user determines to be confirmed heading message, user can input validation heading message, and by visitor in the client Family end to data server send confirm heading message so that data server according to confirm heading message, to business to be optimized The heading message of object optimizes.

Heading message optimization method provided in an embodiment of the present invention, client receive data server transmission, treat After optimizing target hot topic vocabulary corresponding to business object, client receives heading message of the user according to business object to be optimized The confirmation heading message determined with target hot topic vocabulary, and sent to data server and confirm heading message, so that data, services Device optimizes according to heading message is confirmed to the heading message of business object to be optimized.In above process, target hot topic word Converge be data server according to belonging to business object to be optimized target service object set corresponding to popular lexical set and The attribute information of business object to be optimized determines what is obtained so that target hot topic vocabulary can accurately embody business pair to be optimized The characteristic information of elephant, therefore, the heading message of business object to be optimized can be carried out by target hot topic vocabulary accurately excellent Change, and then improve the accuracy that data server recommends business object to user.

On the basis of embodiment illustrated in fig. 15, optionally, client can be obtained by a variety of feasible implementations The confirmation heading message that user determines according to the heading message and target hot topic vocabulary of business object to be optimized is (in Figure 15 embodiments Shown S1502), below, three kinds of feasible implementations are introduced by the embodiment shown in Figure 16-Figure 21.

Figure 16 is the flow chart one provided by the invention for obtaining and confirming heading message method, refers to Figure 16, this method can With including：

S1601, display business object to be optimized heading message and target hot topic vocabulary；

S1602, receive the first confirmation heading message that user inputs according to heading message and target hot topic vocabulary.

In the embodiment shown in Figure 16, after the target hot topic vocabulary that client receives data server transmission, Client obtains the heading message of business object to be optimized, and shows the heading message and target hot topic word of business object to be optimized Converge.

User can determine the first confirmation according to the heading message and target hot topic vocabulary of the business object to be optimized of display Heading message.Client can also show heading message input frame, so that user can be defeated in heading message by input equipment Enter input first in frame and confirm heading message, so that client acquires the first confirmation heading message of user's input.

Below, the client end interface schematic diagram with reference to shown in Figure 17, by specific example, the method shown in Figure 16 is carried out Describe in detail.

Figure 17 is client end interface schematic diagram one provided by the invention, refers to Figure 17, including interface 1701-1702.

Assuming that business object to be optimized is business object 1, the heading message of the business object 1 is：Spring skirt surplus, then Assuming that the target hot topic vocabulary that data server determines to obtain business object 1 is：" 2016 ", " individual character ", " America and Europe ".

After data server determines to obtain the target hot topic vocabulary of business object 1, data server is sent out to client Send the target hot topic vocabulary of business object 1.After the target hot topic vocabulary of business object 1 is received, client shows client Show the heading message and target hot topic vocabulary of business object 1, specifically, referring to interface 1701.

In interface 1701, including the heading message (spring skirt surplus) of business object 1, target hot topic vocabulary (" 2016 ", " individual character ", " America and Europe ") and optimization title entry field M, user can according to the heading message of business object 1 and Target hot topic vocabulary, the input validation heading message in title entry field M is optimized.

In interface 1702, it is assumed that user determines what is obtained according to the heading message and target hot topic vocabulary of business object 1 Heading message is confirmed for " 2016 spring skirt surplus individual character is American-European ", then user can input in title entry field M is optimized " 2016 spring skirt surplus individual character is American-European ", and to " it is determined that " button progress clicking operation, to cause client will " 2016 spring Skirt surplus individual character is American-European " it is defined as confirming heading message.

In above process, data server sends target hot topic vocabulary to client, by user client according to Target hot topic vocabulary determines the heading message after optimization, improves the flexibility optimized to heading message.

Figure 18 is the flowchart 2 provided by the invention for obtaining and confirming heading message method, refers to Figure 18, this method can With including：

S1801, the heading message to be confirmed that data server is sent is received, heading message to be confirmed includes industry to be optimized The heading message and target hot topic vocabulary of business object；

S1802, display heading message to be confirmed；

S1803, receive confirmation corresponding to user's input, heading message to be confirmed；

S1804, according to confirmation, heading message to be confirmed is defined as to confirm heading message.

In the embodiment shown in Figure 18, target hot topic vocabulary is added to the mark of business object to be optimized by data server Inscribe in information, obtain heading message to be confirmed, and the heading message to be confirmed is sent to client.Receive and treated really in client Recognize after heading message, client shows heading message to be confirmed.Optionally, client can also show business object to be optimized Heading message.

Optionally, client can also show " confirmation " button and " modification " button.If customer acceptance title letter to be confirmed Breath, clicking operation can be carried out to " confirmation " button with user, to realize input validation information in the client, client root The confirmation inputted according to user, heading message to be confirmed is defined as to confirm heading message.

Below, the client end interface schematic diagram with reference to shown in Figure 19, by specific example, the method shown in Figure 18 is carried out Describe in detail.

Figure 19 is client end interface schematic diagram two provided by the invention, refers to Figure 19, including interface 1901.

Assuming that business object to be optimized is business object 1, the heading message of the business object 1 is：Spring skirt surplus, number The target hot topic vocabulary for determining to obtain business object 1 according to server is：" 2016 ", " individual character ", " America and Europe ", data server according to The heading message to be confirmed that the heading message and target hot topic vocabulary of business object 1 determine to obtain is " 2016 spring skirt surplus Individual character is American-European ".

After data server determines to obtain the heading message to be confirmed of business object 1, data server is to client Send heading message to be confirmed.For client after heading message to be confirmed is received, client shows the title of business object 1 Information and heading message to be confirmed, specifically, referring to interface 1901.

In interface 1901, include heading message (spring skirt surplus), the heading message to be confirmed of business object 1 (2016 spring skirt surplus individual character is American-European).If customer acceptance heading message to be confirmed, user can enter to " confirmation " button Row clicking operation, to realize input validation information in the client, so that client is according to the confirmation, by title to be confirmed Information (2016 spring skirt surplus individual character is American-European) is defined as confirming heading message.

Figure 20 is the flow chart 3 provided by the invention for obtaining and confirming heading message method, refers to Figure 20, this method can With including：

S2001, the heading message to be confirmed that data server is sent is received, heading message to be confirmed includes industry to be optimized The heading message and target hot topic vocabulary of business object；

S2002, display heading message to be confirmed；

S2003, the modification operation that user inputs to heading message to be confirmed is received, modification operation is used to change target hot topic The vocabulary that position and/or modification to be confirmed heading message of the vocabulary in heading message include；

S2004, according to modification operate, determine the second confirmation heading message.

It should be noted that S2001-S2002 is identical with S1801-S1802, no longer repeated herein.

After client shows heading message to be confirmed, client can also show that " confirmation " button and " modification " are pressed Button.If user does not approve heading message to be confirmed, clicking operation can be carried out to " modification " button with user, and to target heat The vocabulary that position and/or modification to be confirmed heading message of the door vocabulary in heading message include is modified.Client root Operated according to the modification of user's input, determine the second confirmation heading message.

Below, the client end interface schematic diagram with reference to shown in Figure 21, by specific example, the method shown in Figure 20 is carried out Describe in detail.

Figure 21 is client end interface schematic diagram three provided by the invention, refers to Figure 21, including interface 2101-2105.

After data server determines to obtain the heading message to be confirmed of business object 1, data server is to client Send heading message to be confirmed.For client after heading message to be confirmed is received, client shows the title of business object 1 Information and heading message to be confirmed, specifically, referring to interface 2101.

In interface 2101, include heading message (spring skirt surplus), the heading message to be confirmed of business object 1 (2016 spring skirt surplus individual character is American-European).If user does not approve heading message to be confirmed, user can be to " modification " button Clicking operation is carried out, and causes current interface to jump to interface 2102.

In interface 2102, respectively to each vocabulary (target hot topic vocabulary and the industry to be optimized in heading message to be confirmed The vocabulary being engaged in the heading message of object) shown, and show and button P and newly-increased button Q is deleted corresponding to each vocabulary.When When user needs to delete a vocabulary in heading message to be confirmed, user can be carried out a little to deleting button corresponding to the vocabulary Hit operation.When user needs to increase a vocabulary newly in title to be confirmed, for clicking operation can be carried out to newly-increased button, Client shows input frame, so that user inputs newly-increased vocabulary in input frame.Assuming that user needs to believe in title to be confirmed " surplus " vocabulary is deleted in breath, then user can carry out clicking operation to deleting button corresponding to " surplus ", and to work as prezone Face jumps to interface 2103.

In interface 2103, " surplus " vocabulary is deleted in heading message to be confirmed.User can also be to each vocabulary Drag operation is carried out, to realize the position for changing each vocabulary.Assuming that user is needed " individual character " and " America and Europe " two vocabulary Position when being interchangeable, user can choose " America and Europe " vocabulary, and " America and Europe " vocabulary is drawn to the position of " individual character " vocabulary Before.Certainly, user can also choose " individual character " vocabulary, and " individual character " vocabulary is drawn to after the position of " America and Europe " vocabulary, So that obtaining current interface jumps to interface 2104.

In interface 2104, the position of " America and Europe " and " individual character " has been exchanged.When user has been changed heading message to be confirmed Into afterwards, user can to " it is determined that " button carries out clicking operation, show each vocabulary to cancel in interface, and cause current Interface jumps to interface 2105.

In interface 2105, the confirmation heading message of display is " 2016 spring skirt America and Europe individual character ".

In the embodiment shown in Figure 18-Figure 21, data server first determines title to be confirmed according to target hot topic vocabulary Information, and the heading message after optimization is determined according to the heading message to be confirmed by user, further increase to heading message The flexibility optimized.

Below, by taking the interaction of data server and client as an example, the method shown in above-described embodiment is carried out detailed Describe in detail bright.

Figure 22 is the flow chart 3 of heading message optimization method provided by the invention, refers to Figure 22, this method can wrap Include：

S2201, user input the heading message of business object to be optimized in the client；

S2202, client send the heading message of business object to be optimized to data server；

S2203, data server determine the target service object set belonging to business object to be optimized；

S2204, data server obtain popular lexical set and business pair to be optimized corresponding to target service object set As corresponding user characteristics；

S2205, data server determine target hot topic vocabulary according to user characteristics in popular lexical set；

Client corresponding to S2206, data server to business object to be optimized sends target hot topic vocabulary；

S2207, client receive what user determined according to the heading message and target hot topic vocabulary of business object to be optimized Confirm heading message；

S2208, client are sent to data server confirms heading message；

S2209, data server optimize according to heading message is confirmed to the heading message of business object to be optimized.

It should be noted that S2201-S2210 has been carried out being described in detail in the above-described embodiments, the embodiment of the present invention is not Repeated again.

In above process, by data server and client (user) jointly to the heading message of business object to be optimized Optimize.Wherein, target hot topic vocabulary is user characteristics and target service object set according to corresponding to business object to be optimized The popular lexical set of conjunction determines what is obtained so that target hot topic vocabulary can preferably embody the feature of business object to be optimized Information, and then cause user determines to obtain according to target keyword accurately to confirm heading message, and then improve data server Recommend the accuracy of business object to user.

Figure 23 is the flow chart four of heading message optimization method provided by the invention, refers to Figure 23, this method can wrap Include：

S2301, determine at least one business object set that classification includes；

S2302, obtain the second popular word corresponding to the first popular lexical set corresponding to classification and each business object set Collect conjunction；

S2303, respectively to corresponding to each business object set client send corresponding to the second popular lexical set so that User optimizes according to the second popular lexical set to the heading message of the business object in business object set；

S2304, into classification, client corresponding to unfiled business object sends the first popular lexical set, wherein, not Classification business object is not belonging to any one business object set in classification, so that user is according to the first popular lexical set pair The heading message of unfiled business object optimizes.

It should be noted that the concept of classification, business object set and popular vocabulary shown in the embodiment of the present invention with Concept shown in above-mentioned Fig. 1-Figure 22 embodiments is identical, is no longer repeated herein.

In actual application, heading message optimization device determines at least one business object collection that classification includes Close, and obtain the second popular lexical set corresponding to the first popular lexical set corresponding to classification and each business object set.

After heading message optimization device acquires the second popular lexical set corresponding to each business object set, mark Advance data quality device is inscribed respectively to the second popular lexical set corresponding to client transmission corresponding to each business object set, so that User optimizes according to the second popular lexical set to the heading message of the business object in business object set.Optionally, Heading message optimization device can first determine client corresponding to each business object in business object set, and respectively to each business Client corresponding to object sends the second popular lexical set corresponding to business object set.

After heading message optimization device acquires the first popular lexical set corresponding to classification, heading message optimization Device client corresponding to unfiled business object into classification sends the first popular lexical set, wherein, unfiled business pair As any one the business object set being not belonging in classification so that user according to the first popular lexical set to unfiled business The heading message of object optimizes.Optionally, heading message optimization device can obtain the class classification business that classification includes Object, client corresponding to each unfiled business object is determined, and sent out respectively to client corresponding to each unfiled business object Send the first popular lexical set.

Below, by specific example, the method shown in Figure 23 embodiments is described in detail.

Exemplary, it is assumed that classification 1 includes 3 business object set and 10 unfiled business objects, 3 industry Business object set is designated as business object set 1- business objects set 3 respectively, and 10 unfiled business objects are designated as industry respectively Business object 1- business objects 10.Assuming that popular lexical set corresponding to each business object set of classification 1, level is as shown in Table 15：

Table 15

Business object set 1	Popular lexical set 1
		Business object set 2	Popular lexical set 2
Business object set 3	Popular lexical set 3
		Classification 1	Popular lexical set 4

After heading message optimization device determines to obtain the popular lexical set shown in table 15, heading message optimization device Client corresponding to business object into business object set 1 sends popular lexical set 1, into business object set 2 Client corresponding to business object sends popular lexical set 2, client corresponding to the business object into business object set 3 Send popular lexical set 3, send popular lexical set 4 to client corresponding to business object 1- business objects 10, so that industry Client optimizes according to corresponding popular lexical set to its heading message corresponding to business object.

In above process, client corresponding to business object of the heading message optimization device into different business object set End sends popular lexical set corresponding to the business object set, and classification pair is sent to client corresponding to unfiled business object The popular lexical set answered so that the popular lexical set that client corresponding to each business object receives is corresponding Popular lexical set, and then allow the popular lexical set according to corresponding to it of client corresponding to each business object right Its heading message is accurately optimized.

On the basis of embodiment illustrated in fig. 23, optionally, heading message optimization device can be according to following feasible reality Existing mode obtains the second popular lexical set corresponding to business object set：Obtain keyword set corresponding to business object set Close, obtain the hot value of each keyword in keyword set, according to the hot value of each keyword in keyword set, determine each industry Second popular lexical set corresponding to business object set.It should be noted that the second popular vocabulary is determined in the embodiment of the present invention The process of set, the process of popular vocabulary corresponding with determining each business object set shown in Figure 14 embodiments is similar, herein not Repeated again.

Accordingly, heading message optimization device can according to corresponding to following feasible implementation obtains classification the first heat Door keyword set：Duplicate removal processing is carried out to the popular keyword in each second popular keyword set, after being handled according to duplicate removal The each second popular keyword set, determine the first popular lexical set.

, can in order to improve the accuracy that popular vocabulary is sent to each business object on the basis of embodiment illustrated in fig. 23 Choosing, heading message optimization device can also obtain user characteristics corresponding to each business object, according to the user of each business object Feature, target hot topic vocabulary corresponding to each business object is determined in the second popular lexical set, and respectively to each business object Target hot topic vocabulary corresponding to transmission.

Figure 24 is the flow chart five of heading message optimization method provided by the invention, refers to Figure 24, this method can wrap Include：

S2401, receive the popular lexical set that data server is sent；

S2402, the heading message for obtaining business object corresponding to popular lexical set；

S2403, receive the confirmation heading message that user determines according to the heading message of business object and popular lexical set；

S2404, heading message corresponding to business object is updated to confirmation heading message.

In the method shown in Figure 24 embodiments, client receive data server transmission popular lexical set it Afterwards, client obtains the heading message of business object corresponding to popular lexical set, and shows the title of business object to be optimized Information and popular lexical set.

The heading message for the business object that user shows according to client and popular lexical set determine to confirm title letter Breath, and heading message corresponding to business object is updated to confirmation heading message.Optionally, can be taken by client to data The mark for device transmission confirmation heading message and the business object of being engaged in, so that the heading message of business object is updated to by data server Confirm heading message.

In above process, the popular lexical set that client corresponding to each business object receives is corresponding Popular lexical set, and then allow the popular lexical set according to corresponding to it of client corresponding to each business object right Its heading message is accurately optimized.

On the basis of embodiment illustrated in fig. 24, optionally, each popular vocabulary can also be included in popular lexical set Weighted value；Accordingly, client can show that the heading message of business object and popular lexical set, popular lexical set are each The weighted value of popular vocabulary, so that weighted value of the user according to heading message, popular lexical set and each popular vocabulary, input Confirmation heading message.

Figure 25 is the structural representation one that a kind of heading message provided by the invention optimizes device, refers to Figure 25, the dress Putting to include：

First determining module 11, for determining the target service object set belonging to business object to be optimized；

First acquisition module 12, for obtaining popular lexical set corresponding to the target service object set and described User characteristics corresponding to business object to be optimized；

Second determining module 13, for according to the user characteristics, determining that target is popular in the popular lexical set Vocabulary；

Optimization module 14, for according to the target hot topic vocabulary, entering to the heading message of the business object to be optimized Row optimization.

Heading message optimization device provided in an embodiment of the present invention can perform the technical side shown in above method embodiment Case, its realization principle and beneficial effect are similar, are no longer repeated herein.

Figure 26 is the structural representation two that a kind of heading message provided by the invention optimizes device, in embodiment illustrated in fig. 25 On the basis of, Figure 26 is referred to, second determining module 13 includes first acquisition unit 13-1 and the first determining unit 13-2, Wherein,

The first acquisition unit 13-1 is used for, and obtains the user characteristics and each hot topic in the popular keyword set The matching degree of keyword；

The first determining unit 13-2 is used for, according to each hot topic in the user characteristics and the popular keyword set The matching degree of keyword set, target hot topic vocabulary is determined in the popular lexical set.

In a kind of possible embodiment, the first acquisition unit 13-1 is specifically used for：

In alternatively possible embodiment, the first acquisition module 13-1 is specifically used for：

In alternatively possible embodiment, the optimization module 14 is specifically used for：

Obtain at least one current key word that the heading message includes；

In alternatively possible embodiment, the optimization module 14 is specifically additionally operable to：

Receive the first confirmation heading message that the client is sent；

In alternatively possible embodiment, the optimization module 14 includes the second determining unit 14-1 and transmitting element 14-2, wherein,

The second determining unit 14-1 is used for, according to the heading message of the business object to be optimized and target heat Door vocabulary, determines heading message to be confirmed；

The transmitting element 14-2 is used for, and the mark to be confirmed is sent to client corresponding to the business object to be optimized Information is inscribed, so that user determines the heading message after optimization according to the heading message to be confirmed.

In alternatively possible embodiment, the second determining unit 14-1 is specifically used for：

In alternatively possible embodiment, the optimization module also includes receiving unit 14-3 and the 3rd determining unit 14-4, wherein,

The transmitting element 14-2 is specifically used for, and is treated really to described in client transmission corresponding to the business object to be optimized Recognize heading message；

The receiving unit 14-3 is used for, and receives the confirmation that the client is sent；

The 3rd determining unit 14-4 is used for, and according to the confirmation, the heading message to be confirmed is defined as Heading message after the optimization of the business object to be optimized.

In alternatively possible embodiment, the transmitting element 14-2 is additionally operable to, to the business object to be optimized Corresponding client sends the heading message to be confirmed；

The receiving unit 14-3 is additionally operable to, and receive that the client is sent second confirms heading message, and described second Confirm that the heading message user modifies to obtain to the title to be confirmed；

The 3rd determining unit 14-4 is additionally operable to, and the described second confirmation heading message is defined as into the business to be optimized Heading message after the optimization of object.

In alternatively possible embodiment, first determining module 11 includes second acquisition unit 11-1, the 4th Determining unit 11-2 and the 5th determining unit 11-3, wherein,

The second acquisition unit 11-1 is used for, and obtains the target classification belonging to business object to be optimized；

The 4th determining unit 11-2 is used for, and determines multiple business object set to be selected corresponding to the target classification；

The 5th determining unit 11-3 is used for, and the target service is determined in the multiple business object set to be selected Object set.

In alternatively possible embodiment, the second acquisition unit 11-1 is specifically used for：

Obtain the weighted value of each keyword in first keyword set；

In alternatively possible embodiment, the 4th determining unit 11-2 is specifically used for：

Or

In alternatively possible embodiment, the second acquisition unit 11-1 and the 4th determining unit 11-2 tools Body is used for：

In alternatively possible embodiment, described device also includes the second acquisition module 15 and the 3rd determining module 16, wherein,

Second acquisition module 15 is used for, and the mesh belonging to business object to be optimized is determined in first determining module 11 Before marking business object set, Transaction Information corresponding to the business object to be optimized is obtained；

3rd determining module 16 is used for, and determines that Transaction Information corresponding to the business object to be optimized meets following bar It is at least one in part：

In alternatively possible embodiment, described device also include the 3rd acquisition module 17, the 4th acquisition module 18, And generation module 19, wherein,

3rd acquisition module 17 is used for, the second acquisition unit 11-1 according to first keyword set and Before classification forecast model determines the target classification, multiple searching characters that user inputs in the 3rd preset duration are obtained String, be clicked corresponding to each search string business object and it is each it is described be clicked business object be clicked number；

4th acquisition module 18 is used for, and obtains keyword and each quilt corresponding to each search string Classification belonging to click-to-call service object；

The generation module 19 is used for, according to inputted in the 3rd preset duration multiple search strings, each search Be clicked corresponding to character string business object, it is each it is described be clicked business object be clicked number, it is each belonging to search string Corresponding keyword and each classification being clicked belonging to business object, generate classification forecast model.

In alternatively possible embodiment, the generation module 19 is specifically used for：

In alternatively possible embodiment, described device also include division module 110, the 5th acquisition module 111 and 4th determining module 112, wherein,

The division module 110 is used for, and is obtained in the second acquisition unit 11-1 corresponding to target service object set Before popular vocabulary, affiliated all business objects of classification identical are divided to obtain multiple business object set, it is each described The attributes similarity of business object in business object set is more than the 3rd predetermined threshold value, and the multiple business object set includes The target service object set；

5th acquisition module 111 is used for, and obtains the second keyword set of each business object set respectively, obtains To multiple second keyword sets, and obtain the hot value of each keyword in each second keyword set；

4th determining module 112 is used for, according to the hot value of each keyword in each second keyword set, really Popular vocabulary corresponding to fixed each business object set.

In alternatively possible embodiment, the division module 110 is specifically used for：

Figure 27 is the structural representation one that another heading message provided by the invention optimizes device, refers to Figure 27, should Device includes：

First receiving module 21 is popular for receiving target corresponding to data server transmission, business object to be optimized Vocabulary；

Second receiving module 22, for receiving heading message and the target of the user according to the business object to be optimized The confirmation heading message that popular vocabulary determines；

Sending module 23, for sending the confirmation heading message to the data server, so that the data, services Device optimizes according to the confirmation heading message to the heading message of the business object to be optimized.

Figure 28 is the structural representation two that another heading message provided by the invention optimizes device, in implementation shown in Figure 27 On the basis of example, Figure 28 is referred to, described device also includes display module 24, wherein,

The display module 24 is used for, and shows the heading message of the business object to be optimized and the target hot topic word Converge；

Accordingly, second receiving module 22 is used for, and receives the user according to the heading message and the target The first of popular vocabulary input confirms heading message.

In a kind of possible embodiment, first receiving module 21 is specifically used for：

In alternatively possible embodiment, the display module 24 is additionally operable to, and shows the heading message to be confirmed；

Accordingly, second receiving module 22 is specifically used for, and receives heading message pair to be confirmed that user inputs, described The confirmation answered, and according to the confirmation, the heading message to be confirmed is defined as the confirmation heading message.

Accordingly, second receiving module 22 is specifically used for, and it is defeated to the heading message to be confirmed to receive the user The modification operation entered, the modification operation be used to changing position of the target hot topic vocabulary in the heading message and/or The vocabulary that the heading message to be confirmed includes is changed, and is operated according to the modification, determines the second confirmation heading message.

In alternatively possible embodiment, second receiving module 22 is additionally operable to, in first receiving module Before receiving the target hot topic vocabulary that the data server is sent, the title letter of the business object to be optimized of user's input is received Breath；

Accordingly, the sending module 23 is additionally operable to, and the title of the business object to be optimized is sent to data server Information, so that popular vocabulary corresponding to target service object set of the data server according to belonging to the business object to be optimized Set and the attribute information of the business object to be optimized, determine target hot topic vocabulary.

Figure 29 is the structural representation one that another heading message provided by the invention optimizes device, refers to Figure 29, should Device includes：

Determining module 31, at least one business object set included for determining classification；

Acquisition module 32, for obtaining the first popular lexical set corresponding to classification and each business object set pair The the second popular lexical set answered；

Sending module 33, for popular to corresponding to client transmission corresponding to each business object set second respectively Lexical set, so that title of the user according to the described second popular lexical set to the business object in the business object set Information optimizes；

The sending module 33 is additionally operable to, and into the classification, client corresponding to unfiled business object sends described the One popular lexical set, wherein, the unfiled business object is not belonging to any one business object set in the classification, So that user optimizes according to the described first popular lexical set to the heading message of the unfiled business object.

In a kind of possible embodiment, the acquisition module 32 is specifically used for：

Obtain keyword set corresponding to the business object set；

Obtain the hot value of each keyword in the keyword set；

In alternatively possible embodiment, the acquisition module 32 is specifically used for：It is popular to each described second crucial Popular keyword in set of words carries out duplicate removal processing；

Figure 30 is the structural representation two that another heading message provided by the invention optimizes device, in implementation shown in Figure 29 On the basis of example, Figure 30 is referred to, the sending module 33 includes determining unit 33-1 and transmitting element 33-2, wherein,

The determining unit 33-1 is used for, and determines client corresponding to each business object in the business object set；

The transmitting element 33-2 is used for, and sends the business object to client corresponding to each business object respectively Second popular lexical set corresponding to set.

In alternatively possible embodiment, the transmitting element 33-2 is specifically used for：

Obtain user characteristics corresponding to each business object；

In alternatively possible embodiment, the sending module 33 is specifically used for：

Determine client corresponding to each unfiled business object；

Figure 31 is the structural representation one that another heading message provided by the invention optimizes device, refers to Figure 31, should Device includes：

First receiving module 41, for receiving the popular lexical set of data server transmission；

Acquisition module 42, for obtaining the heading message of business object corresponding to the popular lexical set；

Second receiving module 43, for receiving heading message and the popular word finder of the user according to the business object Close the confirmation heading message determined；

Update module 44, for heading message corresponding to the business object to be updated into the confirmation heading message.

Figure 32 is the structural representation two that another heading message provided by the invention optimizes device, in implementation shown in Figure 31 On the basis of example, Figure 32 is referred to, described device also includes display module 45, wherein,

The display module 45 is used for, and shows the heading message of the business object and the popular lexical set；

Accordingly, second receiving module 43 is used for, and receives the user according to the heading message and the hot topic The confirmation heading message of lexical set input.

In a kind of possible embodiment, the weighted value of each popular vocabulary is also included in the popular lexical set；

Accordingly, the display module 45 is used for, and shows the heading message of the business object and the popular word finder Close, the weighted value of popular each popular vocabulary of lexical set；

Second receiving module 43 is used for, receive the user according to the heading message, the popular lexical set, And the weighted value of each popular vocabulary, the confirmation heading message of input.

In alternatively possible embodiment, the update module 44 is specifically used for：

Figure 33 is a kind of structural representation one of data server provided by the invention, refers to Figure 33, the data, services Device includes processor 51, the memory 52 for storing application program and the communication for realizing the communication connection between element Bus 53, the processor 51 is used to read the application program in the memory 52, and performs following operation：

Data server provided in an embodiment of the present invention can perform the technical scheme shown in above method embodiment, in fact Existing principle and beneficial effect are similar, are no longer repeated herein.

In a kind of possible embodiment, the processor 51 is specifically used for：

In alternatively possible embodiment, the processor 51 is specifically used for：

In alternatively possible embodiment, the processor 51 is specifically used for, and the target hot topic vocabulary is added Into the heading message of the business object to be optimized.

Obtain at least one current key word that the heading message includes；

Receive the first confirmation heading message that the client is sent；

Figure 34 is a kind of structural representation two of data server provided by the invention, on the basis of Figure 33 illustrated embodiments On, Figure 34 is referred to, the data server also includes COM1 54, wherein, the processor 51 is specifically used for：

The title to be confirmed is sent by client corresponding to the COM1 54 to the business object to be optimized Information, so that user determines the heading message after optimization according to the heading message to be confirmed.

The title to be confirmed is sent by client corresponding to the COM1 54 to the business object to be optimized Information；

The confirmation of the client transmission is received by the COM1 54；

Receive the client transmission by the COM1 54 second confirms heading message, and described second confirms mark The topic information user modifies to obtain to the title to be confirmed；

In alternatively possible embodiment, the processor 51 is specifically used for：Judge the business object to be optimized With the presence or absence of corresponding classification；

In alternatively possible embodiment, the processor 51 is specifically used for：According to the classification forecast model, obtain Take the classification and matching degree that each keyword is corresponding in first keyword set；

Obtain the weighted value of each keyword in first keyword set；

In alternatively possible embodiment, the processor 51 is specifically used for：Judge the business object to be optimized With the presence or absence of corresponding business object set；

In alternatively possible embodiment, the processor 51 is specifically used for：Characteristic similarity is more than described the The business object set to be selected of one predetermined threshold value is defined as the target service object set；

Or

In alternatively possible embodiment, the processor 51 is specifically used for：Obtain the business object to be optimized Characteristic vector and business object set to be selected central feature vector cosine angle, according to determining the cosine angle Characteristic similarity.

In alternatively possible embodiment, the processor 51 is specifically used for：The business to be optimized is obtained respectively The characteristic vector of object and the cosine angle of the central feature vector of each business object set to be selected；

In alternatively possible embodiment, the processor 51 is additionally operable to, and is determined in the processor 51 to be optimized Before target service object set belonging to business object, Transaction Information corresponding to the business object to be optimized is obtained；

In alternatively possible embodiment, the processor 51 is additionally operable to：

The processor 51 according to first keyword set and classification forecast model determine the target classification it Before, obtain multiple search strings that user inputs in the 3rd preset duration, be clicked corresponding to each search string Business object and it is each it is described be clicked business object be clicked number；

In alternatively possible embodiment, the processor 51 is specifically used for：According to each search string pair The mark for being clicked business object answered and it is each it is described be clicked the number that business object is clicked, obtain search string With the first corresponding relation for being clicked business object, first corresponding relation includes corresponding to each search string, Be clicked business object is clicked number；

Before the processor 51 obtains popular vocabulary corresponding to target service object set, to affiliated classification identical All business objects are divided to obtain multiple business object set, the attribute of the business object in each business object set Similarity is more than the 3rd predetermined threshold value, and the multiple business object set includes the target service object set；

In alternatively possible embodiment, the processor 51 is specifically used for：Each industry of classification identical belonging to acquisition The weighted value of each keyword in 3rd keyword set corresponding to business object and the 3rd keyword set；

Figure 35 is a kind of structural representation one of client provided by the invention, and the client includes processor 61, communication Port 62, input equipment 63, the memory 64 for storing application program and for realizing the logical of the communication connection between element Believe bus 65, wherein,

The processor 61 is used for, and data server is sent, business pair to be optimized is received by the COM1 62 As corresponding target hot topic vocabulary；

The input equipment 63 is used for, and receives heading message and the target of the user according to the business object to be optimized The confirmation heading message that popular vocabulary determines；

The processor 61 is additionally operable to, and sending the confirmation title to the data server by the COM1 believes Breath, so that the data server is carried out according to the confirmation heading message to the heading message of the business object to be optimized Optimization.

Client provided in an embodiment of the present invention can perform the technical scheme shown in above method embodiment, and it realizes former Reason and beneficial effect are similar, are no longer repeated herein.

Figure 36 is a kind of structural representation two of client provided by the invention, on the basis of Figure 35 illustrated embodiments, Figure 36 is referred to, the client also includes display device 66, wherein,

The display device 66 is used for, and shows the heading message of the business object to be optimized and the target hot topic word Converge；

Accordingly, the input equipment 63 is specifically used for, and receives the user according to the heading message and the target The first of popular vocabulary input confirms heading message.

In a kind of possible embodiment, the processor 61 is specifically used for, by described in COM1 reception The heading message to be confirmed that data server is sent, the heading message to be confirmed include the mark of the business object to be optimized Inscribe information and the target hot topic vocabulary.

In alternatively possible embodiment, the display device 66 is additionally operable to, and shows the heading message to be confirmed；

Accordingly, the input equipment 63 is specifically used for, and receives corresponding to heading message to be confirmed that user inputs, described Confirmation；

The processor 61 is specifically used for, and according to the confirmation, the heading message to be confirmed is defined as described Confirm heading message.

Accordingly, the input equipment 63 is specifically used for, and receives the user to the heading message input to be confirmed Modification operation, the modification operation are used to change position and/or modification of the target hot topic vocabulary in the heading message The vocabulary that the heading message to be confirmed includes；

The processor 61 is specifically used for, and is operated according to the modification, determines the second confirmation heading message.

In alternatively possible embodiment, the processor 61 is additionally operable to, and passes through the communication in the processor Before port receives the target hot topic vocabulary that the data server is sent, waiting for user's input is received by the COM1 Optimize the heading message of business object；

Accordingly, the processor 61 is specifically used for, excellent by being treated described in the COM1 to data server transmission Change the heading message of business object, so that target service object set of the data server according to belonging to the business object to be optimized The attribute information of popular lexical set corresponding to conjunction and the business object to be optimized, determine target hot topic vocabulary.

Figure 37 is the structural representation of another data server provided by the invention, and referring to Figure 37 includes processor 71st, COM1 72, the memory 73 for storing application program and for realizing that the communication of the communication connection between element is total Line 74, the processor 71 is used to read the application program in the memory 73, and performs following operation：

Determine at least one business object set that classification includes；

By the COM1 72 respectively to second corresponding to client transmission corresponding to each business object set Popular lexical set so that user according to the described second popular lexical set to the business object in the business object set Heading message optimizes；

By the COM1 72, into the classification, client corresponding to unfiled business object sends described first Popular lexical set, wherein, the unfiled business object is not belonging to any one business object set in the classification, with User is set to be optimized according to the described first popular lexical set to the heading message of the unfiled business object.

In a kind of possible embodiment, the processor 71 is specifically used for：

Obtain keyword set corresponding to the business object set；

Obtain the hot value of each keyword in the keyword set；

In alternatively possible embodiment, the processor 71 is specifically used for：

Obtain user characteristics corresponding to each business object；

Determine client corresponding to each unfiled business object；

Figure 38 is the structural representation one of another client provided by the invention, refers to Figure 38, including processor 81, COM1 82, input equipment 83, the memory 84 for storing application program and for realizing the communication connection between element Communication bus 85, wherein,

The processor 81 is used for, and the popular lexical set of data server transmission is received by the COM1 82；

The processor 81 is additionally operable to, and obtains the heading message of business object corresponding to the popular lexical set；

The input equipment 83 is used for, and receives heading message and the popular word finder of the user according to the business object Close the confirmation heading message determined；

The processor 81 is additionally operable to, and heading message corresponding to the business object is updated into the confirmation title letter Breath.

Figure 39 is the structural representation two of another client provided by the invention, on the basis of Figure 38 illustrated embodiments On, Figure 39 is referred to, the client also includes display device 86, wherein,

The display device 86 is used for, and shows the heading message of the business object and the popular lexical set；

Accordingly, the input equipment 83 is specifically used for, and receives the user according to the heading message and the hot topic The confirmation heading message of lexical set input.

Accordingly, the display device 86 is used for, and shows the heading message of the business object and the popular word finder Close, the weighted value of popular each popular vocabulary of lexical set；

The input equipment 83 is specifically used for, receive the user according to the heading message, the popular lexical set, And the weighted value of each popular vocabulary, the confirmation heading message of input.

In alternatively possible embodiment, the processor 81 is specifically used for：

Figure 40 is the structural representation that a kind of heading message provided by the invention optimizes system, refers to Figure 40, including number According to server 901 and client 902, wherein,

The data server 901 is used for, and determines the target service object set belonging to business object to be optimized, obtains institute User characteristics corresponding to popular lexical set corresponding to target service object set and the business object to be optimized is stated, according to The user characteristics, target hot topic vocabulary is determined in the popular lexical set, and the target hot topic is sent to client Vocabulary；

The client 902 is used for, and receives target heat corresponding to the data server is sent, business object to be optimized Door vocabulary, receive the confirmation mark that user determines according to the heading message of the business object to be optimized and the target hot topic vocabulary Information is inscribed, and the confirmation heading message is sent to the data server, so that the data server is according to the confirmation Heading message, the heading message of the business object to be optimized is optimized.

Heading message optimization system provided in an embodiment of the present invention can perform the technical side shown in above method embodiment Case, its realization principle and beneficial effect are similar, are no longer repeated herein.

Figure 41 is the structural representation that another heading message provided by the invention optimizes system, refers to Figure 41, including Data server 1001 and client 1002, wherein,

The data server 1001 is used for, and determines at least one business object set that classification includes, and obtains classification Second popular lexical set corresponding to corresponding first popular lexical set and each business object set, respectively to each institute State the second popular lexical set corresponding to client transmission corresponding to business object set, and the unfiled business into the classification Client corresponding to object sends the first popular lexical set；

The client 1002 is used for, and receives the popular lexical set that the data server is sent, obtains the hot topic The heading message of business object corresponding to lexical set, receive heading message and the hot topic of the user according to the business object The confirmation heading message that lexical set determines, heading message corresponding to the business object is updated to the confirmation title letter Breath.

Finally it should be noted that：Various embodiments above is only to illustrate the technical scheme of the application, rather than its limitations；To the greatest extent The application is described in detail with reference to foregoing embodiments for pipe, it will be understood by those within the art that：Its according to The technical scheme described in foregoing embodiments can so be modified, either which part or all technical characteristic are entered Row equivalent substitution；And these modifications or replacement, the essence of appropriate technical solution is departed from each embodiment technology of the application The scope of scheme.

Claims

A kind of 1. heading message optimization method, it is characterised in that including：

Determine the target service object set belonging to business object to be optimized；

Obtain user corresponding to popular lexical set corresponding to the target service object set and the business object to be optimized Feature；

According to the user characteristics, target hot topic vocabulary is determined in the popular lexical set；

According to the target hot topic vocabulary, the heading message of the business object to be optimized is optimized.
2. according to the method for claim 1, it is characterised in that it is described according to the user characteristics, in the popular vocabulary Target hot topic vocabulary is determined in set, including：

Obtain the matching degree of the user characteristics and each popular keyword in the popular keyword set；

According to the matching degree of each popular keyword set in the user characteristics and the popular keyword set, in the hot topic Target hot topic vocabulary is determined in lexical set.
3. according to the method for claim 2, it is characterised in that described to obtain the user characteristics and the popular keyword The matching degree of each popular keyword in set, including：

Determine the second vocabulary class corresponding to the first vocabulary classification corresponding to the user characteristics and each popular keyword Not；

According to default vocabulary categorical match table, the first vocabulary classification, each second vocabulary classification, the data are determined Information and the matching degree of each popular keyword, wherein, the default vocabulary categorical match table includes each two vocabulary Matching degree between classification.
4. according to the method described in claim any one of 1-3, it is characterised in that described to obtain the business object pair to be optimized The user characteristics answered, including：

Determine user to be optimized corresponding to the business object to be optimized；

Obtain the user profile of the user to be optimized, the user profile include the user to be optimized log-on message, At least one of Transaction Information corresponding to the adding description information of the user to be optimized, the user to be optimized；

According to the user profile, user characteristics corresponding to the business object to be optimized is determined.
5. according to the method described in claim any one of 1-3, it is characterised in that it is described according to target hot topic vocabulary, to described The heading message of business object to be optimized optimizes, including：

The target hot topic vocabulary is added in the heading message of the business object to be optimized.
6. according to the method for claim 5, it is characterised in that it is described by the target hot topic vocabulary added to it is described treat it is excellent In the heading message for changing business object, including：

Obtain at least one current key word that the heading message includes；

Obtain the weighted value of each current key word and the target hot topic vocabulary；

According to each current key word and the weighted value of the target hot topic vocabulary, determine the target hot topic vocabulary described Position in heading message；

According to position of the target hot topic vocabulary in the heading message, the target hot topic vocabulary is treated added to described In the heading message for optimizing business object.
7. according to the method described in claim any one of 1-3, it is characterised in that it is described according to target hot topic vocabulary, to described The heading message of business object to be optimized optimizes, including：

The target hot topic vocabulary is sent to client corresponding to the business object to be optimized, so that user is according to the target Popular vocabulary determines the heading message after optimization；

Receive the first confirmation heading message that the client is sent；

Described first confirmation heading message is defined as the heading message after the optimization of the business object to be optimized.
8. according to the method described in claim any one of 1-3, it is characterised in that it is described according to target hot topic vocabulary, to described The heading message of business object to be optimized optimizes, including：

According to the heading message of the business object to be optimized and the target hot topic vocabulary, heading message to be confirmed is determined；

The heading message to be confirmed is sent to client corresponding to the business object to be optimized, so that user treats according to described Confirm that heading message determines the heading message after optimization.
9. according to the method for claim 8, it is characterised in that the heading message according to the business object to be optimized With the target hot topic vocabulary, heading message to be confirmed is determined, including：

Determine insertion position of the target hot topic vocabulary in the heading message of the business object to be optimized；

The target hot topic vocabulary is inserted in the insertion position of the heading message, obtains heading message to be confirmed.
10. according to the method for claim 8, it is characterised in that described to client corresponding to the business object to be optimized End sends the heading message to be confirmed, so that user determines the heading message after optimization according to the heading message to be confirmed, Including：

The heading message to be confirmed is sent to client corresponding to the business object to be optimized；

Receive the confirmation that the client is sent；

According to the confirmation, the mark heading message to be confirmed being defined as after the optimization of the business object to be optimized Inscribe information.
11. according to the method for claim 8, it is characterised in that described to client corresponding to the business object to be optimized End sends the heading message to be confirmed, so that user determines the heading message after optimization according to the heading message to be confirmed, Including：

The heading message to be confirmed is sent to client corresponding to the business object to be optimized；

The second confirmation heading message that the client is sent is received, the second confirmation heading message is the user to described Title to be confirmed modifies what is obtained；

Described second confirmation heading message is defined as the heading message after the optimization of the business object to be optimized.
12. according to the method described in claim any one of 1-3, it is characterised in that described to determine belonging to business object to be optimized Target service object set, including：

Obtain the target classification belonging to business object to be optimized；

Determine multiple business object set to be selected corresponding to the target classification；

The target service object set is determined in the multiple business object set to be selected.
13. according to the method for claim 12, it is characterised in that the target class obtained belonging to business object to be optimized Mesh, including：

Judge that the business object to be optimized whether there is corresponding classification；

If so, classification corresponding to the business object to be optimized is then defined as the target classification；

If it is not, then obtain the first keyword set corresponding to business object to be optimized, and according to first keyword set and Classification forecast model determines the target classification, wherein, the classification forecast model includes the corresponding pass of keyword and classification System and the matching degree of keyword and classification.
14. according to the method for claim 13, it is characterised in that described pre- according to first keyword set and classification Survey model and determine the target classification, including：

According to the classification forecast model, obtain each keyword is corresponding in first keyword set classification and With degree；

Obtain the weighted value of each keyword in first keyword set；

According to the matching between the corresponding classification of the weighted value of each keyword, each keyword in first keyword set Degree, determines the target classification.
15. according to the method for claim 12, it is characterised in that described true in the multiple business object set to be selected The fixed target service object set, including：

Judge that the business object to be optimized whether there is corresponding business object set；

If so, business object set corresponding to the business object to be optimized is then defined as the target service object set；

If it is not, then obtain the characteristic information of business object to be optimized, each business object set to be selected characteristic information and The characteristic similarity of the characteristic information of the business object to be optimized and the characteristic information of each business object set to be selected, and According to the characteristic similarity of the characteristic information of characteristic information and each business object set to be selected of the optimization business object, Target service object set is determined in the business object set to be selected.
16. according to the method for claim 15, it is characterised in that the characteristic information according to the optimization business object With the characteristic similarity of the characteristic information of each business object set to be selected, mesh is determined in the business object set to be selected Business object set is marked, including：

The business object set to be selected that characteristic similarity is more than to first predetermined threshold value is defined as the target service object Set；

Or

The N number of business object set to be selected of characteristic similarity highest is defined as the target service object set, wherein, it is described N is the positive integer more than or equal to 1.
17. the method according to claim 15 or 16, it is characterised in that

The characteristic information for obtaining business object to be optimized, including：

Obtain each keyword in the first keyword set corresponding to business object to be optimized and first keyword set Weighted value, according to the weighted value of each keyword in the first keyword set, the characteristic vector of the business object to be optimized is obtained, And the characteristic vector of the business object to be optimized is defined as to the characteristic information of the business object to be optimized；

The characteristic information for obtaining the business object set to be selected, including：

The characteristic vector of each business object in the business object set to be selected is obtained, according in the business object set to be selected The characteristic vector of each business object, the central feature vector of the business object set to be selected is determined, and by the business to be selected The central feature vector of object set is defined as the characteristic information of the business object set to be selected.
18. according to the method for claim 17, it is characterised in that the characteristic information for obtaining business object to be optimized with The characteristic similarity of the characteristic information of the business object set to be selected, including：

The cosine for obtaining the characteristic vector of the business object to be optimized and the central feature vector of business object set to be selected presss from both sides Angle, the characteristic similarity is determined according to the cosine angle.
19. according to the method for claim 17, it is characterised in that according to the characteristic information of the optimization business object and respectively The characteristic similarity of the characteristic information of the business object set to be selected, target industry is determined in the business object set to be selected Business object set, including：

The characteristic vector of the business object to be optimized and the central feature vector of each business object set to be selected are obtained respectively Cosine angle；

The business object set to be selected that cosine angle is less than to second predetermined threshold value is defined as the target service object set Close, or, the minimum N number of business object set to be selected of cosine angle is defined as the target service object set, its In, angle is that characteristic similarity corresponding to the cosine angle of second predetermined threshold value is equal to first predetermined threshold value.
20. the method according to claim 13 or 17, it is characterised in that described to obtain corresponding to business object to be optimized the One keyword set, including：

The description information of the business object to be optimized is obtained, the description information includes the title of the business object to be optimized The characteristic parameter of information and/or the business object to be optimized；

Word segmentation processing is carried out to the description information of the business object to be optimized, obtained more corresponding to the business object to be optimized The part of speech of individual keyword and each keyword；

According to the part of speech of each keyword, Screening Treatment is carried out to the multiple keyword, to obtain first keyword Set.
21. according to the method described in claim any one of 13-18, it is characterised in that described to determine business object institute to be optimized Before the target service object set of category, in addition to：

Obtain Transaction Information corresponding to the business object to be optimized；

It is at least one in following condition to determine that Transaction Information corresponding to the business object to be optimized meets：

The exchange hand in the first preset duration of the business object to be optimized is less than exchange hand threshold value；

The click volume in the second preset duration of the business object to be optimized is less than click volume threshold value.
22. according to the method described in claim any one of 13-18, it is characterised in that described according to first keyword Before set and classification forecast model determine the target classification, in addition to：

Obtain multiple search strings that user inputs in the 3rd preset duration, be clicked corresponding to each search string Business object and it is each it is described be clicked business object be clicked number；

Obtain keyword corresponding to each search string and each classification being clicked belonging to business object；

Business pair is clicked according to the multiple search strings inputted in the 3rd preset duration, corresponding to each search string As, it is each described in be clicked business object be clicked number, it is each belonging to keyword and each quilt corresponding to search string Classification belonging to click-to-call service object, generate classification forecast model.
23. according to the method for claim 22, it is characterised in that described multiple to be searched according to what is inputted in the 3rd preset duration Business object is clicked corresponding to rope character string, each search string, each described is clicked being clicked time for business object Keyword corresponding to search string belonging to several, each and each classification being clicked belonging to business object, generation classification are pre- Model is surveyed, including：

Be clicked according to corresponding to each search string business object mark and it is each described in be clicked business object quilt The number of click, obtain search string and be clicked the first corresponding relation of business object, wrapped in first corresponding relation Include corresponding to each search string, be clicked business object be clicked number；

According to it is each it is described be clicked business object belonging to classification and first corresponding relation, obtain search string and Second corresponding relation of classification, second corresponding relation is included corresponding to the search string, classification is clicked time Number；

According to keyword corresponding to each search string and second corresponding relation, keyword and classification are obtained 3rd corresponding relation, the 3rd corresponding relation includes corresponding to keyword, classification is clicked number；

According to corresponding to each keyword, the matching degree for being clicked number, generating each keyword classification corresponding with its of classification；

According to the 3rd corresponding relation and the matching degree of keyword and classification, the classification forecast model is generated.
24. according to the method described in claim any one of 13-18, it is characterised in that the acquisition target service object set Before corresponding popular vocabulary, in addition to：

Affiliated all business objects of classification identical are divided to obtain multiple business object set, each business object collection The attributes similarity of business object in conjunction is more than the 3rd predetermined threshold value, and the multiple business object set includes the target industry Business object set；

The second keyword set of each business object set is obtained respectively, obtains multiple second keyword sets；

Obtain the hot value of each keyword in each second keyword set；

According to the hot value of each keyword in each second keyword set, determine hot corresponding to each business object set Door vocabulary.
25. according to the method for claim 24, it is characterised in that described that affiliated all business objects of classification identical are entered Row division obtains multiple business object set, including：

In 3rd keyword set corresponding to each business object of classification identical belonging to acquisition and the 3rd keyword set The weighted value of each keyword；

According to the weighted value of each keyword in each 3rd keyword set, the characteristic vector of each business object of acquisition；

According to the cosine angle between the characteristic vector of each business object, affiliated all business objects of classification identical are entered Row division, to obtain the multiple business object set.
A kind of 26. heading message optimization method, it is characterised in that including：

Receive target hot topic vocabulary corresponding to data server is sent, business object to be optimized；

Receive the confirmation title that user determines according to the heading message of the business object to be optimized and the target hot topic vocabulary Information；

The confirmation heading message is sent to the data server, so that the data server is believed according to the confirmation title Breath, is optimized to the heading message of the business object to be optimized.
27. according to the method for claim 26, it is characterised in that the reception user is according to the business object to be optimized Heading message and the target hot topic vocabulary determine confirmation heading message, including：

Show the heading message of the business object to be optimized and the target hot topic vocabulary；

Receive the first confirmation heading message that the user inputs according to the heading message and the target hot topic vocabulary.
28. according to the method for claim 26, it is characterised in that described to receive data server is sent, industry to be optimized Target hot topic vocabulary corresponding to business object, including：

The heading message to be confirmed that the data server is sent is received, the heading message to be confirmed includes described to be optimized The heading message of business object and the target hot topic vocabulary.
29. according to the method for claim 28, it is characterised in that receive user and determined according to the target hot topic vocabulary , confirm heading message corresponding to the business object to be optimized, including：

Show the heading message to be confirmed；

Receive confirmation corresponding to heading message to be confirmed that user inputs, described；

According to the confirmation, the heading message to be confirmed is defined as the confirmation heading message.
30. according to the method for claim 28, it is characterised in that receive user and determined according to the target hot topic vocabulary , confirm heading message corresponding to the business object to be optimized, including：

Show the heading message to be confirmed；

Receive modification of the user to the heading message input to be confirmed to operate, the modification operation is used to change the mesh Mark the vocabulary that position and/or the modification to be confirmed heading message of the popular vocabulary in the heading message include；

Operated according to the modification, determine the second confirmation heading message.
31. according to the method described in claim any one of 26-30, it is characterised in that described to receive the data server hair Before the target hot topic vocabulary sent, in addition to：

Receive the heading message of the business object to be optimized of user's input；

The heading message of the business object to be optimized is sent to data server, so that data server is according to described to be optimized Popular lexical set corresponding to target service object set belonging to business object and the attribute of the business object to be optimized Information, determine target hot topic vocabulary.
A kind of 32. heading message optimization method, it is characterised in that including：

Determine at least one business object set that classification includes；

Obtain the second popular word finder corresponding to the first popular lexical set corresponding to classification and each business object set Close；

Respectively to the second popular lexical set corresponding to client transmission corresponding to each business object set, so that user's root The heading message of the business object in the business object set is optimized according to the described second popular lexical set；

Into the classification, client corresponding to unfiled business object sends the first popular lexical set, wherein, it is described Unfiled business object is not belonging to any one business object set in the classification, so that user is popular according to described first Lexical set optimizes to the heading message of the unfiled business object.
33. according to the method for claim 32, it is characterised in that obtain the second hot topic corresponding to the business object set Lexical set, including：

Obtain keyword set corresponding to the business object set；

Obtain the hot value of each keyword in the keyword set；

According to the hot value of each keyword in the keyword set, the second hot topic corresponding to each business object set is determined Lexical set.
34. according to the method for claim 32, it is characterised in that obtain the first popular lexical set corresponding to classification, bag Include：

Duplicate removal processing is carried out to the popular keyword in each second popular keyword set；

Each second popular keyword set after being handled according to duplicate removal, determines the described first popular lexical set.
35. according to the method described in claim any one of 32-34, it is characterised in that respectively to each business object set Second popular lexical set corresponding to corresponding client transmission, including：

Determine client corresponding to each business object in the business object set；

Respectively the second popular word finder corresponding to the business object set is sent to client corresponding to each business object Close.
36. according to the method for claim 35, it is characterised in that sent out respectively to client corresponding to each business object The second popular lexical set corresponding to the business object set is sent, including：

Obtain user characteristics corresponding to each business object；

According to the user characteristics of each business object, each business object pair is determined in the described second popular lexical set The target hot topic vocabulary answered；

Respectively to target hot topic vocabulary corresponding to each business object transmission.
37. according to the method described in claim any one of 32-36, it is characterised in that the unfiled business pair into the classification As the corresponding client transmission first popular lexical set, including：

Obtain the class classification business object that the classification includes；

Determine client corresponding to each unfiled business object；

Respectively the first popular lexical set is sent to client corresponding to each unfiled business object.
A kind of 38. heading message optimization method, it is characterised in that including：

Receive the popular lexical set that data server is sent；

Obtain the heading message of business object corresponding to the popular lexical set；

Receive the confirmation heading message that user determines according to the heading message and the popular lexical set of the business object；

Heading message corresponding to the business object is updated to the confirmation heading message.
39. according to the method for claim 38, it is characterised in that described to receive title of the user according to the business object The confirmation heading message that information and the popular lexical set determine, including：

Show the heading message of the business object and the popular lexical set；

Receive the confirmation heading message that the user inputs according to the heading message and the popular lexical set.
40. the method according to claim 38 or 39, it is characterised in that also include each hot topic in the popular lexical set The weighted value of vocabulary；Accordingly, described according to the popular lexical set, the heading message progress to the business object is excellent Change, including：

Show the heading message and the popular lexical set, popular each hot topic of lexical set of the business object The weighted value of vocabulary；

Weighted value of the user according to the heading message, the popular lexical set and each popular vocabulary is received, it is defeated The confirmation heading message entered.
41. the method according to claim 38 or 39, it is characterised in that described to be updated to the corresponding heading message The confirmation heading message, including：

The mark for confirming heading message and the business object is sent to the data server, so that the data, services The heading message of the business object is updated to the confirmation heading message by device.
42. a kind of heading message optimizes device, it is characterised in that including：

First determining module, for determining the target service object set belonging to business object to be optimized；

First acquisition module, for obtaining popular lexical set corresponding to the target service object set and described to be optimized User characteristics corresponding to business object；

Second determining module, for according to the user characteristics, target hot topic vocabulary to be determined in the popular lexical set；

Optimization module, for according to the target hot topic vocabulary, being optimized to the heading message of the business object to be optimized.
43. device according to claim 42, it is characterised in that second determining module include first acquisition unit and First determining unit, wherein,

The first acquisition unit is used for, and obtains the user characteristics and each popular keyword in the popular keyword set Matching degree；

First determining unit is used for, according to each popular keyword set in the user characteristics and the popular keyword set The matching degree of conjunction, target hot topic vocabulary is determined in the popular lexical set.
44. device according to claim 42, it is characterised in that the optimization module is specifically used for：

Obtain at least one current key word that the heading message includes；

Obtain the weighted value of each current key word and the target hot topic vocabulary；

According to each current key word and the weighted value of the target hot topic vocabulary, determine the target hot topic vocabulary described Position in heading message；

According to position of the target hot topic vocabulary in the heading message, the target hot topic vocabulary is treated added to described In the heading message for optimizing business object.
45. device according to claim 42, it is characterised in that the optimization module is specifically used for：

The target hot topic vocabulary is sent to client corresponding to the business object to be optimized, so that user is according to the target Popular vocabulary determines the heading message after optimization；

Receive the first confirmation heading message that the client is sent；

Described first confirmation heading message is defined as the heading message after the optimization of the business object to be optimized.
46. according to the device described in claim any one of 42-45, it is characterised in that first determining module includes second Acquiring unit, the 4th determining unit and the 5th determining unit, wherein,

The second acquisition unit is used for, and obtains the target classification belonging to business object to be optimized；

4th determining unit is used for, and determines multiple business object set to be selected corresponding to the target classification；

5th determining unit is used for, and the target service object set is determined in the multiple business object set to be selected Close.
47. device according to claim 46, it is characterised in that the second acquisition unit is specifically used for：

Judge that the business object to be optimized whether there is corresponding classification；

If so, classification corresponding to the business object to be optimized is then defined as the target classification；

If it is not, then obtain the first keyword set corresponding to business object to be optimized, and according to first keyword set and Classification forecast model determines the target classification, wherein, the classification forecast model includes the corresponding pass of keyword and classification System and the matching degree of keyword and classification.
48. device according to claim 46, it is characterised in that the 4th determining unit is specifically used for：

Judge that the business object to be optimized whether there is corresponding business object set；

If so, business object set corresponding to the business object to be optimized is then defined as the target service object set；

If it is not, then obtain the characteristic information of business object to be optimized, each business object set to be selected characteristic information and The characteristic similarity of the characteristic information of the business object to be optimized and the characteristic information of each business object set to be selected, and According to the characteristic similarity of the characteristic information of characteristic information and each business object set to be selected of the optimization business object, Target service object set is determined in the business object set to be selected.
49. device according to claim 46, it is characterised in that the 4th determining unit is specifically used for：

Obtain each keyword in the first keyword set corresponding to business object to be optimized and first keyword set Weighted value, according to the weighted value of each keyword in the first keyword set, the characteristic vector of the business object to be optimized is obtained, And the characteristic vector of the business object to be optimized is defined as to the characteristic information of the business object to be optimized；

The characteristic vector of each business object in the business object set to be selected is obtained, according in the business object set to be selected The characteristic vector of each business object, the central feature vector of the business object set to be selected is determined, and by the business to be selected The central feature vector of object set is defined as the characteristic information of the business object set to be selected.
50. according to the device described in claim any one of 42-45, it is characterised in that described device also includes second and obtains mould Block and the 3rd determining module, wherein,

Second acquisition module is used for, and the target service pair belonging to business object to be optimized is determined in first determining module Before set, Transaction Information corresponding to the business object to be optimized is obtained；

3rd determining module is used for, and determines that Transaction Information corresponding to the business object to be optimized meets in following condition It is at least one：

The exchange hand in the first preset duration of the business object to be optimized is less than exchange hand threshold value；

The click volume in the second preset duration of the business object to be optimized is less than click volume threshold value.
51. device according to claim 46, it is characterised in that described device also obtains including the 3rd acquisition module, the 4th Modulus block and generation module, wherein,

3rd acquisition module is used for, and mould is predicted according to first keyword set and classification in the second acquisition unit Before type determines the target classification, obtain multiple search strings that user inputs in the 3rd preset duration, each described search Be clicked corresponding to rope character string business object and it is each it is described be clicked business object be clicked number；

4th acquisition module is used for, and obtains keyword corresponding to each search string and each described is clicked industry The classification being engaged in belonging to object；

The generation module is used for, according to multiple search strings, each search string inputted in the 3rd preset duration It is corresponding be clicked business object, it is each it is described be clicked business object be clicked number, it is each belonging to corresponding to search string Keyword and each classification being clicked belonging to business object, generate classification forecast model.
52. device according to claim 46, it is characterised in that described device also includes division module, the 5th acquisition mould Block and the 4th determining module, wherein,

The division module is used for, the second acquisition unit obtain target service object set corresponding to popular vocabulary it Before, affiliated all business objects of classification identical are divided to obtain multiple business object set, each business object collection The attributes similarity of business object in conjunction is more than the 3rd predetermined threshold value, and the multiple business object set includes the target industry Business object set；

5th acquisition module is used for, and obtains the second keyword set of each business object set respectively, obtains multiple Second keyword set, and obtain the hot value of each keyword in each second keyword set；

4th determining module is used for, and according to the hot value of each keyword in each second keyword set, determines each institute State popular vocabulary corresponding to business object set.
53. a kind of heading message optimizes device, it is characterised in that including：

First receiving module, for receiving target hot topic vocabulary corresponding to data server transmission, business object to be optimized；

Second receiving module, for receiving heading message and the target hot topic word of the user according to the business object to be optimized The confirmation heading message that remittance determines；

Sending module, for sending the confirmation heading message to the data server so that the data server according to The confirmation heading message, the heading message of the business object to be optimized is optimized.
54. device according to claim 53, it is characterised in that described device also includes display module, wherein,

The display module is used for, and shows the heading message of the business object to be optimized and the target hot topic vocabulary；

Accordingly, second receiving module is used for, and receives the user according to the heading message and the target hot topic word First inputted that converge confirms heading message.
55. device according to claim 53, it is characterised in that

First receiving module is specifically used for, and receives the heading message to be confirmed that the data server is sent, described to treat really Recognizing heading message includes the heading message of the business object to be optimized and the target hot topic vocabulary；

The display module is additionally operable to, and shows the heading message to be confirmed；

Accordingly, second receiving module is specifically used for, and it is corresponding really to receive heading message to be confirmed that user inputs, described Recognize information, and according to the confirmation, the heading message to be confirmed is defined as the confirmation heading message.
56. device according to claim 53, it is characterised in that

First receiving module is specifically used for, and receives the heading message to be confirmed that the data server is sent, described to treat really Recognizing heading message includes the heading message of the business object to be optimized and the target hot topic vocabulary；

The display module is additionally operable to, and shows the heading message to be confirmed；

Accordingly, second receiving module is specifically used for, and receives the user and the heading message input to be confirmed is repaiied Change operation, the modification operation is used to change position and/or modification institute of the target hot topic vocabulary in the heading message The vocabulary that heading message to be confirmed includes is stated, and is operated according to the modification, determines the second confirmation heading message.
57. a kind of heading message optimizes device, it is characterised in that including：

Determining module, at least one business object set included for determining classification；

Acquisition module, for obtaining corresponding to the first popular lexical set corresponding to classification and each business object set Two popular lexical sets；

Sending module, it is used for, respectively to the second popular vocabulary corresponding to client transmission corresponding to each business object set Set, so that heading message of the user according to the described second popular lexical set to the business object in the business object set Optimize；

The sending module is additionally operable to, and into the classification, it is popular to send described first for client corresponding to unfiled business object Lexical set, wherein, the unfiled business object is not belonging to any one business object set in the classification, to use Family optimizes according to the described first popular lexical set to the heading message of the unfiled business object.
58. device according to claim 57, it is characterised in that the acquisition module is specifically used for：

Obtain keyword set corresponding to the business object set；

Obtain the hot value of each keyword in the keyword set；

According to the hot value of each keyword in the keyword set, the second hot topic corresponding to each business object set is determined Lexical set.
59. device according to claim 58, it is characterised in that the acquisition module is specifically used for：To each described second Popular keyword in popular keyword set carries out duplicate removal processing；

Each second popular keyword set after being handled according to duplicate removal, determines the described first popular lexical set.
60. a kind of heading message optimizes device, it is characterised in that including：

First receiving module, for receiving the popular lexical set of data server transmission；

Acquisition module, for obtaining the heading message of business object corresponding to the popular lexical set；

Second receiving module, determined for receiving user according to the heading message and the popular lexical set of the business object Confirmation heading message；

Update module, for heading message corresponding to the business object to be updated into the confirmation heading message.
61. device according to claim 60, it is characterised in that described device also includes display module, wherein,

The display module is used for, and shows the heading message of the business object and the popular lexical set；

Accordingly, second receiving module is used for, and receives the user according to the heading message and the popular word finder Close the confirmation heading message of input.
62. device according to claim 60, it is characterised in that also include each popular vocabulary in the popular lexical set Weighted value；

Accordingly, the display module is used for, and shows heading message and the popular lexical set, the institute of the business object State the weighted value of each popular vocabulary of popular lexical set；

Second receiving module is used for, and receives the user according to the heading message, the popular lexical set and each institute State the weighted value of popular vocabulary, the confirmation heading message of input.
63. a kind of data server, it is characterised in that including processor and for storing the memory of application program, the place Reason device is used to read the application program in the memory, and performs following operation：

Determine the target service object set belonging to business object to be optimized；

Obtain user corresponding to popular lexical set corresponding to the target service object set and the business object to be optimized Feature；

According to the user characteristics, target hot topic vocabulary is determined in the popular lexical set；

According to the target hot topic vocabulary, the heading message of the business object to be optimized is optimized.
64. a kind of client, it is characterised in that including processor, COM1, input equipment and for storing application program Memory, wherein,

The processor is used for, and is received by the COM1 corresponding to data server is sent, business object to be optimized Target hot topic vocabulary；

The input equipment is used for, and receives heading message and the target hot topic word of the user according to the business object to be optimized The confirmation heading message that remittance determines；

The processor is additionally operable to, and the confirmation heading message is sent to the data server by the COM1, with The data server is set to be optimized according to the confirmation heading message to the heading message of the business object to be optimized.
65. a kind of data server, it is characterised in that including processor, COM1 and storage for storing application program Device, the processor is used to read the application program in the memory, and performs following operation：

Determine at least one business object set that classification includes；

Obtain the second popular word finder corresponding to the first popular lexical set corresponding to classification and each business object set Close；

By the COM1 respectively to the second popular word corresponding to client transmission corresponding to each business object set Collect conjunction, so that user is believed the title of the business object in the business object set according to the described second popular lexical set Breath optimizes；

By the COM1, into the classification, client corresponding to unfiled business object sends the first popular word Collect conjunction, wherein, the unfiled business object is not belonging to any one business object set in the classification, so that user The heading message of the unfiled business object is optimized according to the described first popular lexical set.
66. a kind of client, it is characterised in that including processor, COM1, input equipment and for storing application program Memory, wherein,

The processor is used for, and the popular lexical set of data server transmission is received by the COM1；

The processor is additionally operable to, and obtains the heading message of business object corresponding to the popular lexical set；

The input equipment is used for, and receives user and is determined according to the heading message and the popular lexical set of the business object Confirmation heading message；

The processor is additionally operable to, and heading message corresponding to the business object is updated into the confirmation heading message.
67. a kind of heading message optimizes system, it is characterised in that including data server and client, wherein,

The data server is used for, and determines the target service object set belonging to business object to be optimized, obtains the target User characteristics corresponding to popular lexical set corresponding to business object set and the business object to be optimized, according to the use Family feature, target hot topic vocabulary is determined in the popular lexical set, and the target hot topic vocabulary is sent to client；

The client is used for, and receives target hot topic vocabulary corresponding to the data server is sent, business object to be optimized, The confirmation heading message that user determines according to the heading message of the business object to be optimized and the target hot topic vocabulary is received, And the confirmation heading message is sent to the data server, so that the data server is believed according to the confirmation title Breath, is optimized to the heading message of the business object to be optimized.
68. a kind of heading message optimizes system, it is characterised in that including data server and client, wherein,

The data server is used for, and determines at least one business object set that classification includes, and obtains corresponding to classification the Second popular lexical set corresponding to one popular lexical set and each business object set, respectively to each business pair Second popular lexical set corresponding to the client transmission as corresponding to set, and unfiled business object is corresponding into the classification Client send the first popular lexical set；

The client is used for, and receives the popular lexical set that the data server is sent, and obtains the popular lexical set The heading message of corresponding business object, receive heading message and the popular lexical set of the user according to the business object The confirmation heading message of determination, heading message corresponding to the business object is updated to the confirmation heading message.