CN107885889A

CN107885889A - Feedback method, methods of exhibiting and the device of search result

Info

Publication number: CN107885889A
Application number: CN201711330089.8A
Authority: CN
Inventors: 袁丽; 徐钊
Original assignee: Poly Polytron Technologies Inc
Current assignee: Qingdao poly cloud Technology Co., Ltd.
Priority date: 2017-12-13
Filing date: 2017-12-13
Publication date: 2018-04-06

Abstract

This application discloses a kind of feedback method of search result, methods of exhibiting and device, belong to network technique field.Methods described includes：The searching request for carrying search key and user account number mark is received, it is determined that at least two search results matched with search key；Obtain user's portrait corresponding with user account number mark；Relevance between being drawn a portrait according to the attribute information of search result and user, determine the priority of search result；According to the priority of search result, to terminal send feedback information, feedback information is used to indicate in terminal to show search result according to the order of priority from high to low.The application determines the priority of search result by the relevance between the attribute information according to search result and user's portrait, so that server feedback to terminal is the feedback information that meets the user interest preference, while the individual demand of user is met, search efficiency is improved.

Description

Feedback method, methods of exhibiting and the device of search result

Technical field

The invention relates to network technique field, more particularly to a kind of feedback method of search result, methods of exhibiting And device.

Background technology

Search result refers to the related content that server is searched for according to the searching request of user on network.

In correlation technique, the feedback method of search result includes：After server obtains the search key that terminal is sent, The search key and the similarity of each content in database are calculated, the content that similarity is more than to predetermined threshold is defined as Search result, so as to obtain at least two search results, and by least two search results according to similarity from high to low suitable Sequence is ranked up, at least two search results after being sorted, and at least two search results after sequence are fed back into terminal, So that at least two search results after sequence are shown by terminal.

But in the above-mentioned methods, because search result is ranked up according to the similarity with search key, therefore When different users inputs same search key, at least two search results and its sequence that are fed back are identicals, It frequently can lead to a part of user and can not find search result required for itself in several search results before feedback, cause to use Family experience is poor.

The content of the invention

In order to solve the problems, such as individual demand that search result shown in correlation technique can not meet different user, this Application embodiment provides a kind of feedback method of search result, methods of exhibiting and device.The technical scheme is as follows：

First aspect, there is provided a kind of feedback method of search result, methods described include：

The searching request that receiving terminal is sent, search key and user account number mark are carried in the searching request；

It is determined that at least two search results matched with the search key；

User's portrait corresponding with user account number mark is obtained, the user draws a portrait for the emerging of instruction user account number Interesting preference information；

According to the relevance between the attribute information of the search result and user portrait, the search result is determined Priority；

According to the priority of the search result, to the terminal send feedback information, the feedback information is used to indicate In the terminal search result is shown according to the order of the priority from high to low.

Optionally, user's portrait includes：

N number of each self-corresponding first weighted value of video display type, the video display type correspond to K interest dimension, the interest Dimension corresponds to M dimensional characteristics；And

The N*K*M dimensional characteristics each correspond to the second weighted value；

Wherein, described N, K and M are positive integer.

Optionally, before user's portrait corresponding to the acquisition user account number mark, in addition to：

Obtain user account number history interior at preset time intervals and play record；

Determine that the history plays the first accounting corresponding to video display type described in record；

According to N number of first accounting, first object power of the user account number in the predetermined time interval is determined Weight values；

The first object weighted value is added to obtain the first additive value with the first history weighted value；

First additive value is normalized to obtain the first accumulated weight value；

The first weighted value in user's portrait is updated according to the first accumulated weight value.

Determine that the history plays the second accounting corresponding to dimensional characteristics described in record；

According to N*K*M second accountings, second mesh of the user account number in the predetermined time interval is determined Mark weighted value；

The second target weight value is added to obtain the second additive value with the second history weighted value；

Second additive value is normalized to obtain the second accumulated weight value；

The second weighted value in user's portrait is updated according to the second accumulated weight value.

Optionally, the relevance according between the attribute information of the search result and user portrait, it is determined that The priority of the search result, including：

Obtain the attribute information of each search result in the search result；

Drawn a portrait according to the attribute information and the user, determine attribute weight corresponding to the attribute information；

According to the attribute weight, the priority of the search result is obtained using default recommended models, described preset pushes away Recommend model be used for represent obtained Behavior law trained based on historical search result.

Optionally, the attribute information of the search result includes matching degree, and the matching degree is used to indicate the search knot Fruit and the matching degree of keyword in the searching request,

It is described to be drawn a portrait according to the attribute information and the user, attribute weight corresponding to the attribute information is determined, is wrapped Include：

The matching degree is defined as the first attribute weight corresponding to the attribute information.

Optionally, the attribute information of the search result includes foundation characteristic, and the foundation characteristic is used to indicate described search The build-in attribute of hitch fruit, the foundation characteristic include score value, program request amount, video resolution and the video of the search result The sub- foundation characteristic of at least one of show time,

According to the sub- foundation characteristic and the first default corresponding relation, it is determined that belonging to the sub- foundation characteristic corresponding second Property weight, the described first default corresponding relation includes the corresponding relation of the sub- foundation characteristic and second attribute weight.

Optionally, the attribute information of the search result includes video preference profiles, the video preference profiles include with The dimensional characteristics of video display type and interest dimension corresponding to the search result, the interest dimension include video display label, the age, At least one of performer, director and area,

Determined in being drawn a portrait from the user corresponding to the first weighted value corresponding to the video display type and the dimensional characteristics Second weighted value；

First weighted value is multiplied with second weighted value, obtains the 3rd Attribute Weight corresponding to the dimensional characteristics Weight.

Optionally, the attribute information of the search result includes video preference profiles and other features, other described features Including matching degree and/or foundation characteristic,

Interest dimension corresponding to video display type and the search result of the video preference profiles including the search result The dimensional characteristics of degree, interest dimension corresponding to the search result are included in video display label, age, performer, director and area It is at least one；

The matching degree is used for the matching degree for indicating the search result and keyword in the searching request；

The foundation characteristic is used for the build-in attribute for indicating the search result, and the foundation characteristic includes the search and tied At least one of score value, program request amount, video resolution and the video show time of fruit sub- foundation characteristic.

Optionally, it is described according to the attribute weight, the priority of the search result is obtained using default recommended models, Including：

According to the attribute weight, the characteristic vector of the search result is determined；

The default recommended models are obtained, the default recommended models are trained according at least one set of historical search result group Obtain, the historical behavior data group includes：Historical search keyword, historical search result and historical feedback data, it is described Historical feedback data are used to indicate satisfaction corresponding to the historical search result；

The characteristic vector is inputted in the default recommended models, obtains the priority of the search result.

Optionally, at least one of foundation characteristic of the first attribute weight, the attribute information corresponding to the matching degree Second attribute weight corresponding to sub- foundation characteristic, and at least one of video preference profiles of attribute information dimensional characteristics Corresponding 3rd attribute weight, it is described according to the attribute weight, the characteristic vector of the search result is determined, including：

First attribute weight, second attribute weight and the 3rd attribute weight are defined as the search knot The characteristic vector of fruit.

Optionally, the priority according to the search result, to the terminal send feedback information, including：

According to the order of the priority of the search result from high to low, descending processing is carried out to the search result, obtained Search result after to descending；

Search result after the descending is defined as the first feedback information；

First feedback information is sent to the terminal.

According to the priority of the search result, the second feedback information is generated, second feedback information includes described search Hitch fruit and its corresponding priority；

Second feedback information is sent to the terminal.

Second aspect, there is provided a kind of methods of exhibiting of search result, methods described include：

When inputting search key in searched page, used according to corresponding to the search key and the searched page Family account number mark generation searching request；

The searching request is sent to server according to search instruction, the searching request is used to indicate that the server is true It is fixed to be drawn a portrait with the search result of search key matching and user corresponding with user account number mark, and according to institute State associating between the attribute information of search result and the user portrait of the interest preference information for instruction user account number Property, determine the priority of the search result and feedback information is generated according to the priority；

When receiving the feedback information, according to the priority of search result in the feedback information, searched described The search result is shown on the rope page.

Optionally, the feedback information includes at least two search result and at least two search result each Corresponding priority, it is described when receiving the feedback information, according to the priority of search result in the feedback information, The search result is shown on the searched page, including：

According to the order of the priority of at least two search result from high to low, at least two search result Descending processing is carried out, obtains the search result after descending；

The search result after the descending is shown on the searched page.

The third aspect, there is provided a kind of feedback device of search result, described device include：

Receiving module, the searching request sent for receiving terminal, carry in the searching request search key and User account number identifies；

Determining module, at least two search results for determining to match with the search key；

Acquisition module, for obtaining user's portrait corresponding with user account number mark, the user draws a portrait for referring to Show the interest preference information of user account number；

The determining module, it is additionally operable to associating between the attribute information and user portrait according to the search result Property, determine the priority of the search result；

Feedback module, for the priority according to the search result, to the terminal send feedback information, the feedback Information is used to indicate in the terminal to show the search result according to the order of the priority from high to low.

Optionally, user's portrait includes：

Wherein, described N, K and M are positive integer.

Optionally, described device, in addition to：

The acquisition module, it is additionally operable to obtain user account number history broadcasting record interior at preset time intervals；

Computing module, for determining that the history plays the first accounting corresponding to video display type described in record；

Update module, for updating the first weighted value in user's portrait according to the first accumulated weight value.

Optionally, described device, in addition to：

Computing module, for determining that the history plays the second accounting corresponding to dimensional characteristics described in record；

Update module, for updating the second weighted value in user's portrait according to the second accumulated weight value.

Optionally, the determining module, including：Acquiring unit, determining unit and computing unit；

The acquiring unit, for obtaining the attribute information of each search result in the search result；

The determining unit, for according to the attribute information and user portrait, determining that the attribute information is corresponding Attribute weight；

The computing unit, for according to the attribute weight, the search result to be obtained using default recommended models Priority, the default recommended models are used for the Behavior law that expression is trained to obtain based on historical search result.

The determining unit, it is additionally operable to the matching degree being defined as the first attribute weight corresponding to the attribute information.

The determining unit, be additionally operable to according to the sub- foundation characteristic and the first default corresponding relation, it is determined that with the son Second attribute weight corresponding to foundation characteristic, the described first default corresponding relation include the sub- foundation characteristic and the described second category The corresponding relation of property weight.

The determining unit, be additionally operable to from the user draw a portrait in determine the first weighted value corresponding to the video display type and Second weighted value corresponding to the dimensional characteristics；First weighted value is multiplied with second weighted value, obtains the dimension Spend the 3rd attribute weight corresponding to feature.

Optionally, the computing unit, be additionally operable to, according to the attribute weight, to determine the feature of the search result to Amount；The default recommended models are obtained, the default recommended models are to train to obtain according at least one set of historical search result group , the historical behavior data group includes：Historical search keyword, historical search result and historical feedback data, the history Feedback data is used to indicate satisfaction corresponding to the historical search result；The characteristic vector is inputted into the default recommendation In model, the priority of the search result is obtained.

Optionally, the first attribute weight corresponding to the matching degree, at least one of the foundation characteristic for obtaining attribute information Second attribute weight corresponding to sub- foundation characteristic, and obtain at least one of the video preference profiles of attribute information dimensional characteristics Corresponding 3rd attribute weight, the computing unit, it is additionally operable to first attribute weight, second attribute weight and institute State the characteristic vector that the 3rd attribute weight is defined as the search result.

Optionally, the feedback module, including：Sequencing unit, determining unit and the first feedback unit；

Sequencing unit, for the order according to the priority of at least two search result from high to low, to it is described extremely Few two search results carry out descending processing, obtain the search result after descending；

Determining unit, for the search result after the descending to be defined as into the first feedback information；

First feedback unit, for sending first feedback information to the terminal.

Optionally, the feedback module, including：Generation unit and the second feedback unit；

Generation unit, for the priority according to the search result, generate the second feedback information, second feedback letter Breath includes the search result and its corresponding priority；

Second feedback unit, for sending second feedback information to the terminal.

Fourth aspect, there is provided a kind of exhibiting device of search result, described device include：

Generation module, for when inputting search key in searched page, according to the search key and described searching User account number mark generation searching request corresponding to the rope page；

Sending module, for sending the searching request to server according to search instruction, the searching request is used to refer to Show the server determine the search result that is matched with the search key and with the corresponding user of user account number mark Portrait, and the attribute information according to the search result and the user of the interest preference information for instruction user account number Relevance between portrait, determine the priority of the search result and feedback information is generated according to the priority；

Display module, for when receiving the feedback information, according in the feedback information search result it is preferential Sequentially, the search result is shown on the searched page.

Optionally, the feedback information includes at least two search result and at least two search result each Corresponding priority, the display module, including：Sequencing unit and display unit；

The sequencing unit, for the order according to the priority of at least two search result from high to low, to institute State at least two search results and carry out descending processing, obtain the search result after descending；

The display unit, for showing the search result after the descending on the searched page.

5th aspect, there is provided a kind of server, the server include processor and memory, deposited in the memory Contain at least one instruction, at least one instruction loaded by the processor and performed to realize such as above-mentioned first aspect or The feedback method for the search result that the optional implementation of any one in person's first aspect is provided.

6th aspect, there is provided a kind of terminal, the terminal include processor and memory, be stored with the memory At least one instruction, at least one instruction are loaded by the processor and performed to realize such as above-mentioned second aspect or the The methods of exhibiting for the search result that any one optional implementation is provided in two aspects.

A kind of 7th aspect, there is provided computer-readable recording medium, it is characterised in that be stored with the storage medium At least one instruction, at least one instruction are loaded by the processor and performed to realize such as above-mentioned first aspect or the The feedback method for the search result that the optional implementation of any one in one side is provided.

Eighth aspect, there is provided a kind of computer-readable recording medium, it is characterised in that be stored with the storage medium At least one instruction, at least one instruction are loaded by the processor and performed to realize such as above-mentioned second aspect or the The methods of exhibiting for the search result that any one optional implementation is provided in two aspects.

The beneficial effect brought of technical scheme that the embodiment of the present application provides is：

By the searching request of receiving terminal transmission, search key and user account number mark are carried in searching request, It is determined that at least two search results matched with search key, obtain user's portrait, Yong Huhua corresponding to user account number mark Interest preference information as being used for instruction user account number, associating between being drawn a portrait according to the attribute information of search result with user Property, the priority of search result is determined, according to the priority of search result, to terminal send feedback information, feedback information is used for Indicate to show search result according to the order of priority from high to low in terminal；Because user's portrait can reflect using the use The interest preference of the user of family account number, when the user of different interest preferences is with same search Keywords matching search result, lead to The attribute information crossed with reference to user's portrait and search result determines the priority of each search result, realizes between foregoing user For the differentiation of the priority of search result so that server feeds back to terminal according to the priority of search result and meets the use The feedback information of family interest preference, improve the user and find searching required for itself in several search results before feedback The possibility of hitch fruit, while the individual demand of user is met, improve search efficiency and Consumer's Experience.

Brief description of the drawings

Fig. 1 be implement to exemplify according to exemplary partial the feedback method of search result a kind of, involved by methods of exhibiting Implementation environment schematic diagram；

Fig. 2 is the method flow diagram of the feedback method for the search result that one exemplary embodiment of the application provides；

Fig. 3 is the method flow diagram of the training process for the default recommended models that one exemplary embodiment of the application provides；

Fig. 4 is the principle schematic of the training process for the default recommended models that one exemplary embodiment of the application provides；

Fig. 5 is the method for the feedback method and methods of exhibiting of the search result of the application another exemplary embodiment offer Flow chart；

Fig. 6 is the structural representation of the feedback device for the search result that one exemplary embodiment of the application provides；

Fig. 7 is the structural representation of the exhibiting device for the search result that one exemplary embodiment of the application provides；

Fig. 8 is the structured flowchart for the terminal that one exemplary embodiment of the application provides；

Fig. 9 is the structural representation for the server that one exemplary embodiment of the application provides.

Embodiment

To make the purpose, technical scheme and advantage of the application clearer, below in conjunction with accompanying drawing to the application embodiment party Formula is described in further detail.

First, to the invention relates to some nouns explain：

User draws a portrait：It is the user model of the labeling taken out according to user behavior data.

In the embodiment of the present application, user draws a portrait the interest preference information for instruction user account number.Optionally, interest is inclined Good information includes at least one of motion preference information, tourism favor information, cuisines preference information and video display preference information.Under Face only illustrates so that interest preference information includes video display preference information as an example.

Default recommended models：It is a kind of mathematical modeling for being used to determine priority according to the data of input.

Alternatively, default recommended models include but is not limited to：Deep neural network (Deep Neural Network, DNN) Model, Recognition with Recurrent Neural Network (Recurrent Neural Networks, RNN) model, insertion (embedding) model, gradient Lift decision tree (Gradient Boosting Decision Tree, GBDT) model, logistic regression (Logistic At least one of Regression, LR) model.

DNN models are a kind of deep learning frameworks.DNN models include input layer, at least one layer of hidden layer (or intermediate layer) And output layer.Alternatively, input layer, at least one layer of hidden layer (or intermediate layer) and output layer include at least one neuron, Neuron is used to handle the data received.Alternatively, the quantity of the neuron between different layers can be with identical；Or Person, can also be different.

RNN models are a kind of neutral nets with feedback arrangement.In RNN models, the output of neuron can be under One timestamp is applied directly to itself, i.e. input of the i-th layer of neuron at the m moment, except (i-1) layer neuron this when Outside the output at quarter, in addition to its own is in the output at (m-1) moment.

Embedding models are to be based on entity and relation distribution vector representation, by the relation in each triple example Regard the translation from entity head to entity tail as.Wherein, triple example includes main body, relation, object, and triple example can be with table It is shown as (main body, relation, object)；Main body is entity head, and object is entity tail.Such as：The father of Xiao Zhang is big, then passes through three Tuple example is expressed as (Xiao Zhang, father are big to open).

GBDT models are a kind of decision Tree algorithms of iteration, and the algorithm is made up of more decision trees, and the result of all trees is tired out Add up as final result.Each node of decision tree can obtain a predicted value, and by taking the age as an example, predicted value is to belong to The average value at owner's age of node corresponding to the age.

LR models refer on the basis of linear regression, apply mechanically the model that a logical function is established.

Fig. 1 be implement to exemplify according to exemplary partial the feedback method of search result a kind of, involved by methods of exhibiting Implementation environment schematic diagram.As shown in figure 1, the implementation environment includes terminal 120 and server 140.

Terminal 120 can be mobile phone, tablet personal computer, E-book reader, MP3 player (Moving Picture Experts Group Audio Layer III, dynamic image expert's compression standard audio aspect 3), MP4 (Moving Picture Experts Group Audio Layer IV, dynamic image expert's compression standard audio aspect 4) player, knee Mo(u)ld top half pocket computer and desktop computer etc..

Optionally, the application program with search process function is installed in the terminal.Optionally, the application program also has There is display search result function.

Schematically, the application program include but be not limited to network application class application program, Entertainment class application program, Video class application program and operating system management class application program etc..Below, it is video class application program only with the application program Exemplified by illustrate.For example the video class application program is to provide video search function and/or provides answering for video playback capability Use program.

It is connected between terminal 120 and server 140 by communication network.Optionally, communication network is cable network or nothing Gauze network.

Server 140 is also referred to as search server or search engine, and server 140 includes the clothes with search process function Business device.

Optionally, server 140 is a server, or by some servers, or a virtual platform, An either cloud computing service center.

Generally, when server 140 receiving terminal transmission after carrying the searching request of search key, it is determined that with At least two search results of search key matching, and the feedback of the search result provided according to following each embodiment of the method Mode, to the send feedback information of terminal 120 so that terminal according to feedback information by least two search results according to priority from High to Low order is shown.

It should be noted that the implementation environment includes at least one terminal 120 and at least one server 140, in figure A terminal 120 and a server 140 are schematically show only, the present embodiment is not limited to this.

Optionally, above-mentioned wireless network or cable network use standard communication techniques and/or agreement.Network be usually because Special net, it may also be any network, including but not limited to LAN (Local Area Network, LAN), Metropolitan Area Network (MAN) (Metropolitan Area Network, MAN), wide area network (Wide Area Network, WAN), mobile, wired or nothing Any combinations of gauze network, dedicated network or VPN).In certain embodiments, using including hypertext markup Language (Hyper Text Mark-up Language, HTML), extensible markup language (Extensible Markup Language, XML) etc. technology and/or form represent the data by network exchange.It can additionally use such as safe Socket layer (Secure Socket Layer, SSL), Transport Layer Security (Trassport Layer Security, TLS), void Intend dedicated network (Virtual Private Network, VPN), Internet Protocol Security (Internet Protocol Security, IPsec) etc. conventional encryption techniques encrypt all or some links.In further embodiments, can also make Substitute or supplement above-mentioned data communication technology with customization and/or the exclusive data communication technology.

Fig. 2 is refer to, the side of the feedback method of the search result provided it illustrates one exemplary embodiment of the application Method flow chart.The feedback method of the search result is used in the server that Fig. 1 is provided, the feedback method bag of the search result Include：

Step 201, the searching request that receiving terminal is sent, search key and user account number mark are carried in searching request Know.

Optionally, when terminal gets the search key of user's input, searching request, the search are sent to server Request carries search key；Corresponding, server receives searching request.

Optionally, searching request is asked including video search, for asking to search for video content in video library.Search please At least two search results corresponding to asking include at least two video search results corresponding to video search request.

Step 202, it is determined that at least two search results matched with search key.

Server determines that what is matched with the search key in searching request at least two searches according to preset matching algorithm Hitch fruit.

For example preset matching algorithm is string matching algorithm.

Step 203, user's portrait corresponding with user account number mark is obtained, user draws a portrait for the emerging of instruction user account number Interesting preference information.

Server according to the searching request received, it is determined that user account number corresponding with the searching request identify, obtain with User's portrait corresponding to user account number mark.Wherein, user account number is identified for the unique mark user account number.

Wherein, server determines that user account number mark corresponding with the searching request includes but is not limited to following two possibility Implementation：

In a kind of possible implementation, server determines terminal iidentification corresponding to searching request, based on storage One corresponding relation determines account number mark corresponding with terminal iidentification, and the first corresponding relation includes terminal iidentification and pair of account number mark It should be related to.

Wherein, terminal iidentification is used for unique mark terminal.For example terminal iidentification is International Mobile Station Equipment Identification (International Mobile Equipment Identity, IMEI).

User account number mark is also carried in alternatively possible implementation, in searching request.Server is from reception To searching request in extract user account number mark.

The present embodiment pair determines that user account number mark corresponding with the searching request is not limited.

Optionally, user account number mark and the corresponding relation of user's portrait are stored with server, server is according to storage Corresponding relation determine that corresponding with user account number mark user draws a portrait.

Optionally, user is drawn a portrait the video interest preference information for instruction user account number, and video interest preference information is used Preference in reflection user account number to video.

Schematically, user account number includes the preference to different video display types to the preference of video, and/or right The preference profiles of the different interest dimensions of same video display type.Wherein, interest dimension includes video display label, age, performer, led At least one of drill with area.

Wherein, user's portrait includes multiple user tag groups, every group of user tag group include multiple user tag types with Weighted value corresponding to every kind of user tag type.The building process of user's portrait refers to the associated description in example below, Do not introduce first herein.

Step 204, according to the relevance between the attribute information of search result and user's portrait, the excellent of search result is determined First level.

For each search result at least two search results, server determines that user corresponding to the search result marks Type is signed, is drawn a portrait based on user and determines weighted value corresponding with the user tag type.

Server weighted value according to corresponding to the search result, the preferential of the search result is obtained using default recommended models Level.The calculating process of the training process of default recommended models and the priority of search result refers to the phase in example below Description is closed, is not introduced first herein.

The priority of search result is used for the estimated satisfaction for indicating search result.Optionally, search result is preferential Level with it is expected that satisfaction into positive correlation.That is, the estimated satisfaction of the higher expression search result of priority is higher, preferentially The estimated satisfaction of the lower expression search result of level is lower.

Step 205, according to the priority of search result, to terminal send feedback information, feedback information was used to indicate at end On end search result is shown according to the order of priority from high to low.

Server is according to the priority of search result, to terminal send feedback information, it is including but not limited to following two can The implementation of energy：

In a kind of possible implementation, order of the server according to the priority of search result from high to low, to extremely Few two search results carry out descending processing, obtain the search result after descending, the search result after descending is defined as into first Feedback information, the first feedback information is sent to terminal.

Wherein, the priority of at least two search results differs, or, exist between at least two search results Priority is identical.

Schematically, it is identical when the priority between at least two search results at least two search results be present When, the clooating sequences of the two search results is determined according to preset rules or at random.

Searched for due to a fairly large number of situation of at least two search results occurs, therefore in server at least two As a result after being ranked up, the search result for the forward predetermined number that sorts can be obtained, by the forward predetermined number that sorts Search result is defined as the first feedback information.The present embodiment is not limited to this.

Server sends the first feedback information to terminal, corresponding, the first feedback information that terminal the reception server is sent, And the search result after descending is obtained from the first feedback information, the search result after descending is shown.

Optionally, at least two search results are shown with tabular form, the order of search result in lists with it is excellent First level is more forward into positive correlation, the i.e. higher search result of priority, position in lists.

In alternatively possible implementation, server is according to the priority of the search result, the feedback of generation second Information, the second feedback information include search result and its corresponding priority, and the second feedback information is sent to terminal.

Optionally, server sends the second feedback information to terminal, corresponding, terminal the second feedback information of reception, and from At least two search results and each self-corresponding priority of at least two search results are obtained in second feedback information, according to search As a result the order of priority from high to low, descending processing is carried out to search result, obtains the search result after descending；By descending Search result afterwards is shown.At least two search results are referred to server by terminal according to the process of priority ranking will At least two search results will not be repeated here according to the process of priority ranking.

It should be noted that the sequence step of at least two search results can be completed by server, can also be by terminal To complete.The present embodiment is not limited to this.

In summary, the searching request that the present embodiment is sent by receiving terminal, it is crucial that search is carried in searching request Word and user account number mark, it is determined that at least two search results matched with search key, it is corresponding to obtain user account number mark User portrait, the interest preference information that user draws a portrait for instruction user account number, according to the attribute information and use of search result Relevance between the portrait of family, the priority of search result is determined, according to the priority of search result, feedback letter is sent to terminal Breath, feedback information are used to indicate in terminal to show search result according to the order of priority from high to low；Because user draws a portrait The interest preference of the user using the user account number can be reflected, different interest preferences user with same search keyword During with search result, the attribute information by combining user's portrait and search result determines the priority of each search result, real Between present foregoing user for search result priority differentiation so that server is anti-according to the priority of search result Terminal of feeding meets the feedback information of the user interest preference, improves the user and is searched before feedback in several search results The possibility of search result required for itself, while the individual demand of user is met, improve search efficiency and Consumer's Experience.

Below, the building process of user's portrait is introduced.

Optionally, user's portrait includes multiple user tag groups, and every group of user tag group is including user tag type and often Weighted value corresponding to kind user tag type.

The dividing mode of user tag includes but is not limited to following two possible dividing modes：

In the first possible dividing mode, divided according to coarseness, user tag is divided into N number of video display type, N Individual each self-corresponding first weighted value of video display type.

Schematically, terminal storage has the corresponding relation of N number of video display type and N number of first weighted value.At one schematically Example in, the corresponding relation is as shown in Table 1.In table one, user tag includes 4 video display types, is respectively " film ", " electricity Depending on play ", " animation " and " variety ".First weighted value corresponding to " film " is w_movie, the first weighted value corresponding to " TV play " is w_tv, the first weighted value corresponding to " animation " is w_anime, the first weighted value corresponding to " variety " is w_variety。

Table one

Video display type	First weighted value
		Film	w_movie
TV play	w_tv
		Animation	w_anime
Variety	w_variety

In second of possible dividing mode, according to partition by fine granularities, for each video display class in N number of video display type Type, is further divided into K interest dimension, and corresponding each interest dimension is further divided into M dimensional characteristics, that is, obtains N* Each self-corresponding second weighted value of K*M dimensional characteristics.Wherein, N, K and M are positive integer.

Optionally, each video display type corresponds to K interest dimension, K interest dimension include video display label, the age, performer, At least one of director and area.

It should be noted that the quantity of dimensional characteristics corresponding at least two interest dimensions in K interest dimension be present is Different, or the quantity all same of dimensional characteristics corresponding to each interest dimension.For convenience of explanation, below only with each emerging Dimensional characteristics corresponding to interesting dimension are to illustrate exemplified by M dimensional characteristics.

Each interest dimension includes M dimensional characteristics, and dimensional characteristics are used for the scope for indicating the interest dimension.

For example when interest dimension includes video display label, video display label is further divided into action, magical, love, feared The M dimensional characteristics such as fear, suspense.

Again for example, when interest dimension includes the age, M age scope will be further divided into the age, by M age model Enclose as M dimensional characteristics corresponding to interest dimension " age ".Wherein, at least two's scope, which is not present, occurs simultaneously.

Again for example, when interest dimension includes performer, performer is further divided into M performer's title, by M performer's name M dimensional characteristics referred to as corresponding to interest dimension " performer ".

Again for example, when interest dimension includes director, performer is further divided into M director's title, name is directed by M M dimensional characteristics referred to as corresponding to interest dimension " director ".

Again for example, when interest dimension includes regional, performer is further divided into M region identifier, M area is marked Knowledge is used as M dimensional characteristics corresponding to interest dimension " area ".

Schematically, terminal storage has the corresponding relation of K interest dimension and K the second weighted values.At one schematically Example in, the corresponding relation is as shown in Table 2.In table two, user tag includes 4 video display types, is respectively " film ", " electricity Depending on play ", " animation " and " variety ", the corresponding 5 interest dimensions of each video display type, be respectively " video display label ", " age ", " drill Member ", " director " and " area ".

Wherein, when video display type is " film ", the second weighted value corresponding to " video display label " is w_{movie_tag1}, " age " Corresponding second weighted value is w_{movie_pubdate1}, the second weighted value corresponding to " performer " is w_{movie_actor1}, corresponding to " director " Two weighted values are w_{movie_director1}, the second weighted value corresponding to " area " is w_{movie_country1}；

When video display type is " TV play ", the second weighted value corresponding to " video display label " is w_{movie_tag2}, " age " is right The second weighted value answered is w_{movie_pubdate2}, the second weighted value corresponding to " performer " is w_{movie_actor2}, second corresponding to " director " Weighted value is w_{movie_director2}, the second weighted value corresponding to " area " is w_{movie_country2}；

When video display type is " animation ", the second weighted value corresponding to " video display label " is w_{movie_tag3}, " age " is corresponding The second weighted value be w_{movie_pubdate3}, the second weighted value corresponding to " performer " is w_{movie_actor3}, the second power corresponding to " director " Weight values are w_{movie_director3}, the second weighted value corresponding to " area " is w_{movie_country3}；

When video display type is " variety ", the second weighted value corresponding to " video display label " is w_{movie_tag4}, " age " is corresponding The second weighted value be w_{movie_pubdate4}, the second weighted value corresponding to " performer " is w_{movie_actor4}, the second power corresponding to " director " Weight values are w_{movie_director4}, the second weighted value corresponding to " area " is w_movi_{e_country4}。

Table two

Schematically, the 4 video display types and each self-corresponding 5 interest dimensions of 4 video display types provided based on table two, Each interest dimension is further divided into 3 dimensional characteristics, that is, obtains the 4*5*3=60 dimension that user portrait includes Feature.

The corresponding relation of each dimensional characteristics and the second weighted value is stored with terminal.In a schematical example, Only so that video display type is " film " as an example, the corresponding 5 interest dimensions of the video display type, the corresponding 3 dimension spies of each interest dimension The corresponding relation of sign, each dimensional characteristics and the second weighted value is as shown in Table 3.

In table three, interest dimension " video display label " includes 3 dimensional characteristics, respectively " action ", " magical " and " love ", Second weighted value corresponding to " action " is w_{movie_tag11}, the second weighted value corresponding to " magical " is w_{movie_tag12}, " love " is corresponding The second weighted value be w_{movie_tag13}；

Interest dimension " director " includes 3 dimensional characteristics, is respectively " Lee one ", " money five " and " poplar six ", " correspondence of Lee one " The second weighted value be w_{movie_director11}, " the second weighted value corresponding to money five " is w_{movie_director12}, " corresponding to poplar six " Two weighted values are w_{movie_director13}；

Interest dimension " area " includes 3 dimensional characteristics, is respectively " China ", " U.S. " and " Britain ", and " China " is corresponding The second weighted value be w_{movie_country11}, the second weighted value corresponding to " U.S. " is w_{movie_country12}, corresponding to " Britain " Two weighted values are w_{movie_country13}。

Table three

Optionally, the building process of user's portrait includes but is not limited to following steps：

1st, obtain user account number history interior at preset time intervals and play record.

Server obtains user's history interior at preset time intervals and plays record.Optionally, predetermined time interval 12 Individual hour or 24 hours.

Optionally, it is the broadcasting record that accumulative playing duration exceedes scheduled duration that the history, which plays record,.Optionally, make a reservation for Shi Changwei 60 seconds.

2nd, determine that history plays the first accounting corresponding to each video display type in record, and the corresponding to each dimensional characteristics Two accountings.

Server is after getting multiple history and playing record, it is determined that each history plays the shadow of video corresponding to record Depending on type and dimensional characteristics, so that it is determined that the first accounting corresponding to each video display type, and second corresponding to each dimensional characteristics Accounting.

For example user has played 3 films, 2 TV plays and 1 variety by user account number in one day, then services Device obtains the user account number and plays record in intraday 6 history, it is determined that each history plays the shadow of video corresponding to record Depending on type and dimensional characteristics, so that it is determined that the first accounting corresponding to video display type " film " is 3/6=0.5, video display type " TV First accounting corresponding to play " is 2/6=0.3, and the first accounting corresponding to video display type " animation " is 0, and video display type " variety " is right The first accounting answered is 1/6=0.2.

Calculate the second accounting corresponding to each dimensional characteristics can analogy first accounted for reference to corresponding to calculating each video display type The process of ratio, will not be repeated here.

3rd, according to N number of first accounting and N*K*M the second accountings, user account number target interior at preset time intervals is determined Weighted value.

Server determines that user account number is interior at preset time intervals according to N number of first accounting and N*K*M the second accountings Target weight value.Wherein, target weight value includes N number of first accounting and N*K*M the second accountings.

4th, target weight value is added with history weighted value, obtains additive value.

Target weight value is added by server with history weighted value, obtains additive value.Wherein, history weighted value is included in meter Calculate multiple weighted values corresponding to the user's portrait determined before obtaining target weight value.

It should be noted that when describing calculating process, target weight value is used to indicate N number of first accounting and N*K*M A value in second accounting, history weighted value are used to indicate a weight determined in history corresponding with target weight value Value.

For example as shown in Table 4, target weight value includes the first accounting " 0.5 ", video display corresponding to video display type " film " First accounting " 0.3 " corresponding to type " TV play ", the first accounting " 0 ", video display type are " comprehensive corresponding to video display type " animation " First accounting " 0.2 " corresponding to skill ", history weighted value include history weighted value " 0.6 ", video display corresponding to video display type " film " History weighted value " 0.2 " corresponding to type " TV play ", history weighted value " 0.4 ", video display class corresponding to video display type " animation " History weighted value " 0.5 " corresponding to type " variety ", then server by 4 target weight values respectively with each self-corresponding history weight Value is added, and obtains 4 additive values, additive value " 1.1 " respectively corresponding to video display type " film ", and video display type " TV play " is right The additive value " 0.5 " answered, additive value " 0.4 " corresponding to video display type " animation ", additive value corresponding to video display type " variety " “0.7”。

Table four

Video display type	Target weight value	History weighted value	Additive value
				Film	0.5	0.6	1.1
TV play	0.3	0.2	0.5
				Animation	0	0.4	0.4
Variety	0.2	0.5	0.7

5th, value is will add up to be normalized to obtain accumulated weight value.

Server will add up value and be normalized to obtain accumulated weight value.The accumulated weight obtained after normalized The interval range of value is [0,1].

6th, user's portrait is updated according to accumulated weight value.

Server updates the weighted value in user's portrait according to accumulated weight value, including：Obtained by above-mentioned computational methods Multiple accumulated weight values, multiple accumulated weight values are defined as to the included multiple weighted values of user's portrait after renewal, so as to User's portrait after being updated.

Such as 4 additive values provided based on above-mentioned table four, the 4 accumulated weight values obtained after being normalized, This 4 accumulated weight values are defined as included each self-corresponding first weighted value of 4 video display types of user portrait, respectively For：First weighted value " 0.9 " corresponding to video display type " film ", the first weighted value " 0.4 " corresponding to video display type " TV play ", First weighted value " 0.3 " corresponding to video display type " animation ", the first weighted value " 0.6 " corresponding to video display type " variety ".

Optionally, server determines user account number first object power interior at preset time intervals according to N number of first accounting Weight values, first object weighted value is added to obtain the first additive value with the first history weighted value, the first additive value is subjected to normalizing Change handles to obtain the first accumulated weight value, updates the first weighted value in user's portrait according to the first accumulated weight value.

Optionally, server determines the second interior at preset time intervals mesh of user account number according to N*K*M the second accountings Weighted value is marked, the second target weight value is added to obtain the second additive value with the second history weighted value, the second additive value is carried out Normalized obtains the second accumulated weight value, updates the second weighted value in user's portrait according to the second accumulated weight value.

Below, the training process for presetting recommended models is introduced.

The training process of default recommended models includes：Server obtains training sample set, and training sample set includes at least one Group historical search result group；At least one set of historical search result group is calculated using sequence study (Learning to Rank, LTR) Method is trained, and obtains default recommended models.

Optionally, Ranking Algorithm is lambdaMART algorithms.

Wherein, every group of historical behavior data group includes：Historical search keyword, historical search result and historical feedback number According to.Wherein, historical search keyword includes the search key carried in historical search request, and historical search result includes being based on Returned results for video list is asked in the historical search, and historical feedback data are used for instruction user to the historical search result Feedback behavior, that is, it is used to indicate satisfaction corresponding to historical search result.

Server extracts the attribute information of search result, according to attribute information according at least one set of historical search result group Determine attribute weight.Wherein, the attribute information of search result includes video preference profiles and other features, and other features include With degree and/or foundation characteristic.Below, video preference profiles, matching degree and basis spy are only included with the attribute information of search result Levy and illustrate exemplified by this three category feature.

The dimensional characteristics of interest dimension corresponding to video display type and search result of the video preference profiles including search result, Interest dimension corresponding to search result includes at least one of video display label, age, performer, director and area.

Matching degree is used for the matching degree for indicating search result and keyword in searching request.

Foundation characteristic is used for the build-in attribute for indicating search result, and foundation characteristic includes the score value of search result, program request At least one of amount, video resolution and video show time foundation characteristic.

After server gets this three category feature of video preference profiles, matching degree and foundation characteristic, this three class spy is determined Each self-corresponding attribute weight is levied, by three generic attribute weight combination output characteristic values, is trained, obtained using Ranking Algorithm To default recommended models.

Wherein, after server gets video preference profiles, matching degree and foundation characteristic this three category feature, this three class is determined Each self-corresponding attribute weight of feature includes：Server determines the first attribute weight corresponding to matching degree, determines foundation characteristic pair The second attribute weight answered, determine the 3rd attribute weight corresponding to video preference profiles.It is it should be noted that special to above-mentioned three class The calculation for levying each self-corresponding attribute weight refers to associated description in example below, wouldn't introduce herein.

Optionally, the first attribute weight includes weight w corresponding to matching degree_sim, it is corresponding that the second attribute weight includes score value Weight w_rate, weight w corresponding to program request amount_vv, weight w corresponding to " video resolution "_4k, weighed corresponding to " video show time " Weight w_newAt least one of, the 3rd attribute weight includes weight w corresponding to video display type_category, weight corresponding to video display label w_tag, weight w corresponding to the age_pubdate, weight w corresponding to performer_actor, weight w corresponding to director_directorCorresponding to area Weight w_countryAt least one of.

Wherein, three generic attribute weights are defined as input feature value by server；Such as input feature value x= [w_sim, w_rate, w_vv, w_4k, w_new, w_category, w_tag, w_pubdate, w_actor, w_director, w_country]。

Optionally, server determines output characteristic value according to historical feedback data.Schematically, such as following formula institute Show, output characteristic value is represented with y, when historical feedback data are used to indicate " user does not click on and do not watch video ", determines y's Value is " 0 "；When historical feedback data are used to indicate " user clicks on and browses video details, but does not watch video ", determine y's Value is " 1 "；When historical feedback data are used to indicate that " user clicks on and browses video details, and browses viewing video and be less than 60s " When, the value for determining y is " 2 "；When historical feedback data are used to indicate that " user clicks on and browses video details, and browses viewing and regard When frequency is more than 60s ", the value for determining y is " 3 ".

Fig. 3 is refer to, the specific training process of default recommended models includes：Step 301, obtain training sample set, training Sample set includes at least one set of historical search result group；Step 302, according at least one set of historical search result group, extraction search As a result attribute information；Step 303, according to historical feedback data determine output characteristic value；Step 304, determine that matching degree is corresponding The first attribute weight；Step 305, determine the second attribute weight corresponding to foundation characteristic；Step 306, determine video preference spy 3rd attribute weight corresponding to sign；Step 307, by three generic attribute weight combination output characteristic values, entered using Ranking Algorithm Row training, obtain default recommended models.

It should be noted that above-mentioned steps 302 and step 303 can perform side by side, step 304,305 and 306 can also Perform side by side, the present embodiment is not limited to this.

In a schematical example, Fig. 4 is refer to, server obtains training sample set, and the training sample set includes This 20 groups of historical search result groups of historical search result group A1 to historical search result group A20, this 20 groups of historical search knots Every group of historical behavior data group in fruit group includes：Historical search keyword, historical search result and historical feedback data.To go through The training process of default recommended models is introduced exemplified by history set of search results A1.It is corresponding that server extracts historical search result group A1 Historical search result " WW time " attribute information S1, attribute information S1 includes matching degree, foundation characteristic and video preference This three category feature of feature, matching degree are " 60% ", and foundation characteristic includes score value " 8.0 ", program request amount " 4,000,000 times ", video point Resolution " 4K resolution ratio " and video show time " 2010/10/12 ", video preference profiles include video display type " film ", interest The dimensional characteristics " action " of dimension " video display label ", dimensional characteristics " Lee one " and the interest dimension " area " of interest dimension " director " Dimensional characteristics " China ".Server matching degree, foundation characteristic and video preference according to corresponding to historical search result group A1 This three category feature of feature, and the first attribute weight " 0.6 " corresponding to matching degree is calculated, the second attribute corresponding to foundation characteristic Weight " 0.5 ", the 3rd attribute weight " 0.4 " corresponding to video preference profiles, determine input feature value for (0.6,0.5, 0.4).When server determines that historical feedback data corresponding to historical search result group A1 are used to indicate that " user clicks on to browse and regarded Frequency details, and viewing video is browsed when being more than 60s ", the value for determining output characteristic value is " 3 ".Server is according to input feature vector Vector is (0.6,0.5,0.4) and corresponding output characteristic value " 3 ", and default recommended models are trained using Ranking Algorithm.When When historical search result group is any one group in historical search result group A2 to historical search result group A20, training process can Analogy refers to above-mentioned historical search result group A1 training process, will not be repeated here.

It should be noted that when server gets new historical search result group, by new historical search result group Added to training sample set, the training sample set after being updated, recommended models are preset according to the training sample set pair after renewal It is trained, the default recommended models after being updated.

The default recommended models that user's portrait and training based on above-mentioned structure obtain, refer to Fig. 4, it illustrates this Shen Please another exemplary embodiment provide the feedback method of search result and the method flow diagram of methods of exhibiting.The search result Feedback method and methods of exhibiting include：

Step 501, when inputting search key in searched page, terminal is corresponding according to search key and searched page User account number mark generation searching request.

Step 502, terminal sends searching request according to the search instruction to server.

When terminal receives search instruction, searching request is sent to server according to the search instruction, corresponding, service The searching request that device receiving terminal is sent.Wherein, search key and user account number mark are carried in searching request.

Step 503, the searching request that server receiving terminal is sent.

Step 504, server determines at least two search results matched with search key.

Server is according to string matching algorithm, it is determined that match with the search key in searching request at least two Search result.

Step 505, server obtains user's portrait, user corresponding to user account number mark and drawn a portrait for instruction user account number Interest preference information.

Server obtains user's portrait of user account number, and user portrait includes：N number of video display type each self-corresponding first Weighted value, and, each self-corresponding second weighted value of N*K*M dimensional characteristics.

Step 506, server obtains the attribute information of each search result at least two search results.

For each search result at least two search results, server obtains the attribute information of search result.

Step 507, server is drawn a portrait according to attribute information and user, determines attribute weight corresponding to attribute information.

Server is drawn a portrait according to attribute information and user, determines attribute weight corresponding to attribute information, is included but is not limited to Three kinds of possible implementations below：

In the first possible implementation, the attribute information of search result includes matching degree, and matching degree is used to indicate The matching degree of search result and keyword in searching request.Matching degree is defined as the first category corresponding to attribute information by server Property weight.

Optionally, matching degree Score (Q, d) is calculated by equation below in server, and by matching degree Score (Q, d) is defined as the first attribute weight w corresponding to attribute information_sim。

Wherein, Q is search statement corresponding to searching request, and d is the search result of matching, q_iIt is each in search statement Search key, f_iIt is search key q_iThe word frequency occurred in search result d, dl are search result d length, avgdl It is the average length of the search result matched, IDF (q_i) it is search key q_iInverse document frequency, k₁, b be parameter preset.

In second of possible implementation, the attribute information of search result includes foundation characteristic, and foundation characteristic is used for The build-in attribute of search result is indicated, foundation characteristic is included on score value, program request amount, video resolution and the video of search result Reflect at least one of time sub- foundation characteristic.

It is pre- according to sub- foundation characteristic and first for every kind of sub- foundation characteristic at least one sub- foundation characteristic, server If corresponding relation, it is determined that the second attribute weight corresponding with sub- foundation characteristic, the first default corresponding relation includes sub- foundation characteristic With the corresponding relation of the second attribute weight.

For example scoring scope and the second attribute weight w are previously stored with server_rateThe first default corresponding relation. Schematically, the first default corresponding relation is as shown in Table 5.When the score value of search result is " 7.5 ", server determines The score value of the search result scoring scope [6,8) in, according to the first preset relation determine corresponding with the score value second Attribute weight w_rateFor " 0.6 ".

Table five

Score scope	w_rate
		[0,2)	0.1
[2,4)	0.2
		[4,6)	0.4
[6,8)	0.6
		[8,10]	0.8

Again for example, default resolution ratio is 4K resolution ratio, i.e., 4096 × 2160 pixel resolution.When point of the search result When resolution is 4K resolution ratio, w is determined_4kValue be " 1 "；When the resolution ratio of the search result is not 4K resolution ratio, w is determined_4k Value be " 0 ".

It should be noted that when group foundation characteristic is program request amount or video show time, can analogy refer to above-mentioned determination The determination mode of second attribute weight, will not be repeated here.

In the third possible implementation, the attribute information of search result includes video preference profiles, video preference Feature is used for the preference profiles for indicating multiple dimensions corresponding to search result, including video display type and other video preference profiles, Other video preference profiles include at least one of video display label, age, performer, director and area dimensional characteristics.

Drawn a portrait according to user, determine the first weighted value corresponding to video display type；For every at least one dimensional characteristics Kind dimensional characteristics, draw a portrait according to user, determine the second weighted value corresponding to dimensional characteristics；By the first weighted value and the second weighted value It is multiplied, obtains the 3rd attribute weight corresponding to dimensional characteristics.

Optionally, for same interest dimension, when dimensional characteristics corresponding to the interest dimension include at least two dimensions During feature, each self-corresponding second weighted value of at least two dimensional characteristics is determined, by the second weight corresponding to each dimensional characteristics Value is multiplied to obtain multiplied value with the first weighted value, then multiplied value corresponding to each dimensional characteristics is added to obtain the dimensional characteristics pair The 3rd attribute weight answered.

For example only illustrated so that dimensional characteristics are dimensional characteristics corresponding to interest dimension " video display label " as an example, when searching When rope keyword is " action class film ", three search results of search key matching, respectively search result 1 are determined " XX is chivalrous ", search result 2 " YY chants the spring " and search result 3 " WW time ".For search result 1, the video display of acquisition search result 1 Label is label 1 " action " and label 2 " magical ".Server obtains user corresponding to the user account number and drawn a portrait, so that it is determined that shadow The first weighted value corresponding to type " film " is regarded as 0.3, the second weighted value corresponding to label 1 is 0.5 in video display label, label 1 Corresponding second weighted value is 0.2, passes through equation below ∑ w_movie·w_{movie_tag}Calculated, i.e. 0.3*0.5+0.3*0.2= 0.21, so that it is determined that the 3rd attribute weight w corresponding to the dimensional characteristics_tagFor 0.21.

It should be noted that when video type is other video types can analogy refer to above-mentioned calculation, work as dimension Degree be characterized as during other dimensional characteristics also can analogy refer to above-mentioned calculation, do not repeating herein.

Step 508, server attribute weight according to corresponding to attribute information, search result is obtained using default recommended models Priority.

Wherein, recommended models are preset and are used for the Behavior law that expression is trained to obtain based on historical search result.

Attribute weight includes at least one of foundation characteristic of the first attribute weight, attribute information subbase corresponding to matching degree Second attribute weight corresponding to plinth feature, and corresponding at least one of video preference profiles of attribute information dimensional characteristics Three attribute weights.

Optionally, server attribute weight according to corresponding to attribute information, search result is obtained using default recommended models Priority include：The attribute weight according to corresponding to attribute information, the characteristic vector of search result is determined, obtain default recommendation mould Type, characteristic vector is inputted in default recommended models, obtains the priority of search result.

Wherein, server attribute weight according to corresponding to attribute information, determining the characteristic vector of search result includes：By One attribute weight, at least one second attribute weight and at least one 3rd attribute weight be defined as the feature of search result to Amount.

Such as the first attribute weight w_simFor " 0.3 ", the second attribute weight includes w_rate“0.2”、w_vv“0.6”、w_4k“1”、 w_new" 0 ", the 3rd attribute weight include：w_category“0.5”、w_tag“0.4”、w_pubdate“0.2”、w_actor“0.1”、w_director " 0.2 " and w_country" 0.3 ", server is by the first above-mentioned attribute weight, 4 the second attribute weights and 6 the 3rd Attribute Weights Be defined as the characteristic vector of search result again, i.e. characteristic vector x=[0.3,0.2,0.6,1,0,0.5,0.4,0.2,0.1,0.2, 0.3].Server inputs characteristic vector x in default recommended models, and it is " 2 " to obtain output characteristic value, by output characteristic value " 2 " It is defined as the priority of search result 1.

Again for example, the first attribute weight w_simFor " 0.3 ", the second attribute weight includes w_rate“0.2”、w_vv“0.6”、w_4k “1”、w_new" 0 ", the 3rd attribute weight include：w_category“0.5”、w_tag“0.4”、w_pubdate“0.2”、w_actor“0.1”、 w_director" 0.2 " and w_country" 0.3 ", server is by 2 in the first above-mentioned attribute weight, 4 the second attribute weights Two attribute weight (i.e. w_vv" 0.6 " and w_4k" 1 ") and 6 the 3rd attribute weights be defined as the characteristic vector of search result, i.e. feature Vector x=[0.3,0.6,1,0.5,0.4,0.2,0.1,0.2,0.3].Characteristic vector x is inputted default recommended models by server In, it is " 1 " to obtain output characteristic value, and output characteristic value " 1 " is defined as to the priority of search result 1.

Step 509, server is according to the priority of search result, to terminal send feedback information.

Optionally, server generates feedback information, the feedback information includes according to the priority of at least two search results At least two search results and each self-corresponding priority of at least two search results, server send the feedback letter to terminal Breath；Corresponding, terminal receives the feedback information.

For example server analyzes the priority of six search results by the above method, as shown in Table 6, six search As a result priority includes：Search result S1 priority is 1, and search result S2 priority is 3, and search result S3's is preferential Level is 2, and search result S4 priority is 0, and search result S5 priority is 1, and search result S6 priority is 2.

Table six

Search result	Priority
		S1	1
S2	3
		S3	2
S4	0
		S5	1
S6	2

Step 510, order of the terminal according to the priority of search result from high to low, search result is carried out at descending Reason, obtains the search result after descending.

When terminal receives feedback information, according to the priority of search result in feedback information, on searched page Before showing search result, in addition to：Terminal extracts at least two search results and at least two search knots from feedback information Each self-corresponding priority of fruit, at least two search results are subjected to descending processing according to the order of priority from high to low, obtained Search result after to descending.

For example the priority of six search results provided based on above-mentioned table six, terminal are excellent according to six search results The order of first level from high to low, this six search results are ranked up, the search result after obtained descending is：Search result S2, search result S3, search result S6, search result S1, search result S5 and search result S4.

Step 511, terminal shows the search result after descending on searched page.

Optionally, terminal pre-sets the quantity for the search result for needing to show.

For example terminal is shown on searched page to forward 3 search results of sorting, the search result of displaying according to It is secondary to be：Search result S2, search result S3, search result S6.

In summary, the present embodiment is also by according to corresponding to four big video display type dimensions and each video display type dimension Multiple dimensional characteristics determine that user draws a portrait so that user's portrait can more accurately reflect the interest preference letter of the user Breath, and then improve the accuracy rate drawn a portrait according to user to search result corresponding to user's recommendation.

Attribute of the present embodiment also by for each search result at least two search results, obtaining search result Information, drawn a portrait according to attribute information and/or user, attribute weight corresponding to attribute information is determined, according to corresponding to attribute information Attribute weight, the priority of search result is obtained using default recommended models；Because default recommended models are used to represent to be based on going through History search result trains obtained Behavior law, so as to ensure that presetting the priority that recommended models obtain according to this is relatively defined Really.

In a schematical example, terminal 1 is sent to server carries search key " action class film " Searching request 1, three search results that server matches according to the search key is determined, respectively search result 1 " XX is chivalrous ", Search is calculated according to the user of the terminal 1 portrait 1 in search result 2 " YY chants the spring " and search result 3 " WW time ", server As a result priority corresponding to 1 is 1, and priority corresponding to search result 2 is 0, and priority corresponding to search result 3 is 2, to terminal 1 send feedback information 1, the feedback information 1 include these three search results and each self-corresponding priority, and terminal is according to receiving Feedback information 1, these three search results are shown according to the order of priority from high to low, i.e., shown in terminal 1 Search result be followed successively by：Search result 3 " WW time ", search result 1 " XX is chivalrous " and search result 2 " YY chants the spring ".Terminal 2 to Server sends the searching request 2 for carrying same search key " action class film ", and server is according to the determination search Keywords matching goes out three same search results, i.e. search result 1 " XX is chivalrous ", search result 2 " YY chants the spring " and search result 3 " WW time ".Server priority according to corresponding to being calculated search result 1 in the user of the terminal 2 portrait 2 is 1, search knot Priority corresponding to fruit 2 is 3, and priority corresponding to search result 3 is 0, to the send feedback information 2 of terminal 2, the feedback information 2 Including these three search results and each self-corresponding priority, terminal ties these three search according to the feedback information 2 received Fruit is shown according to the order of priority from high to low, i.e., the search result shown in terminal 2 is followed successively by：Search result 2 " YY chants the spring ", search result 1 " XX is chivalrous " and search result 3 " WW time ".It can be seen that because user's portrait of different user account number is logical It is often different so that when different terminals send the searching request for carrying same search key " action class film ", The feedback information of server feedback is different, so cause the clooating sequence of search result that terminal is shown be it is different, Improve follow-up search efficiency.

Following is the application device embodiment, can be used for performing the application embodiment of the method.It is real for the application device The details not disclosed in example is applied, refer to the application embodiment of the method.

Fig. 6 is refer to, the knot of the feedback device of the search result provided it illustrates one exemplary embodiment of the application Structure schematic diagram.The feedback device of the search result can by special hardware circuit, or, software and hardware be implemented in combination with turn into figure In implementation environment in 1 server all or part of, the device includes：Receiving module 610, determining module 620, obtain mould Block 630 and feedback module 640.

Receiving module 610, for the searching request of receiving terminal transmission, search key and use are carried in searching request Family account number mark；

Determining module 620, at least two search results for determining to match with search key；

Acquisition module 630, for obtaining user's portrait corresponding with user account number mark, user draws a portrait for instruction user The interest preference information of account number；

Determining module 620, it is additionally operable to according to the relevance between the attribute information of search result and user's portrait, it is determined that searching The priority of hitch fruit；

Feedback module 640, for the priority according to search result, to terminal send feedback information, feedback information is used for Indicate to show search result according to the order of priority from high to low in terminal.

Optionally, user's portrait includes：

N number of each self-corresponding first weighted value of video display type, video display type correspond to K interest dimension, and interest dimension corresponds to M Individual dimensional characteristics；And

N*K*M dimensional characteristics each correspond to the second weighted value；

Wherein, N, K and M are positive integer.

Acquisition module 630, it is additionally operable to obtain user account number history broadcasting record interior at preset time intervals；

Computing module, for determining that history plays the first accounting corresponding to video display type in record；

According to N number of first accounting, user account number first object weighted value interior at preset time intervals is determined；

First object weighted value is added to obtain the first additive value with the first history weighted value；

Optionally, the device, in addition to：

Computing module, for determining that history plays the second accounting corresponding to dimensional characteristics in record；

According to N*K*M the second accountings, the second interior at preset time intervals target weight value of user account number is determined；

Second target weight value is added to obtain the second additive value with the second history weighted value；

Optionally, determining module 620, including：Acquiring unit, determining unit and computing unit；

Acquiring unit, for obtaining the attribute information of each search result in search result；

Determining unit, for according to attribute information and user's portrait, determining attribute weight corresponding to attribute information；

Computing unit, for according to attribute weight, obtaining the priority of search result using default recommended models, presetting and push away Recommend model be used for represent obtained Behavior law trained based on historical search result.

Optionally, the attribute information of search result includes matching degree, and matching degree is used to indicate search result and searching request The matching degree of middle keyword,

Determining unit, it is additionally operable to matching degree being defined as the first attribute weight corresponding to attribute information.

Optionally, the attribute information of search result includes foundation characteristic, and foundation characteristic is used to indicate the intrinsic of search result Attribute, foundation characteristic include at least one of score value, program request amount, video resolution and video show time of search result Sub- foundation characteristic,

Determining unit, it is additionally operable to according to sub- foundation characteristic and the first default corresponding relation, it is determined that corresponding with sub- foundation characteristic The second attribute weight, the first default corresponding relation includes the corresponding relation of sub- foundation characteristic and the second attribute weight.

Optionally, the attribute information of search result includes video preference profiles, and video preference profiles include and search result The dimensional characteristics of corresponding video display type and interest dimension, interest dimension include video display label, age, performer, director and area At least one of,

Determining unit, it is additionally operable to determine that the first weighted value corresponding to video display type and dimensional characteristics are corresponding in drawing a portrait from user The second weighted value；First weighted value is multiplied with the second weighted value, obtains the 3rd attribute weight corresponding to dimensional characteristics.

Optionally, the attribute information of search result includes video preference profiles and other features, and other features include matching Degree and/or foundation characteristic,

The dimensional characteristics of interest dimension corresponding to video display type and search result of the video preference profiles including search result, Interest dimension corresponding to search result includes at least one of video display label, age, performer, director and area；

Matching degree is used for the matching degree for indicating search result and keyword in searching request；

Optionally, computing unit, it is additionally operable to according to attribute weight, determines the characteristic vector of search result；Obtain default push away Model is recommended, default recommended models train to obtain according at least one set of historical search result group, and historical behavior data group includes： Historical search keyword, historical search result and historical feedback data, historical feedback data are used to indicate historical search result pair The satisfaction answered；Characteristic vector is inputted in default recommended models, obtains the priority of search result.

Optionally, at least one of foundation characteristic of the first attribute weight, attribute information subbase plinth corresponding to matching degree is special Second attribute weight corresponding to sign, and the 3rd category corresponding at least one of video preference profiles of attribute information dimensional characteristics Property weight, computing unit, is additionally operable to the first attribute weight, the second attribute weight and the 3rd attribute weight being defined as search result Characteristic vector.

Optionally, feedback module 640, including：Sequencing unit, determining unit and the first feedback unit；

Sequencing unit, for the order according to the priority of search result from high to low, at least two search results are entered The processing of row descending, obtains the search result after descending；

Determining unit, for the search result after descending to be defined as into the first feedback information；

First feedback unit, for sending the first feedback information to terminal.

Optionally, feedback module, including：Generation unit and the second feedback unit；

Generation unit, for the priority according to search result, the second feedback information is generated, the second feedback information includes searching Hitch fruit and its corresponding priority；

Second feedback unit, for sending the second feedback information to terminal.

Correlative detail can be with reference to shown in referring to figs. 2 to Fig. 5 embodiment of the method.Wherein, receiving module 610 is additionally operable to realize Other in above method embodiment arbitrarily imply or the disclosed function related to receiving step；Determining module 620 is additionally operable to reality Other in existing above method embodiment arbitrarily imply or the disclosed function related to determining step；Acquisition module 630 is additionally operable to Realize any implicit or disclosed function related to obtaining step of other in above method embodiment；Feedback module 640 is also used In realizing that other in above method embodiment are arbitrarily implicit or the disclosed function related to feedback step.

Fig. 7 is refer to, the knot of the exhibiting device of the search result provided it illustrates one exemplary embodiment of the application Structure schematic diagram.The exhibiting device of the search result can by special hardware circuit, or, software and hardware be implemented in combination with turn into figure In implementation environment in 1 terminal all or part of, the device includes：Generation module 710, sending module 720 and displaying mould Block 730.

Generation module 710, for when searched page input search key when, according to search key and searched page Corresponding user account number mark generation searching request；

Sending module 720, for sending searching request to server according to search instruction, searching request is used to indicate to service Device determines to draw a portrait with the search result of search key matching and user corresponding with user account number mark, and is tied according to search The attribute information of fruit and for instruction user account number interest preference information user portrait between relevance, it is determined that search knot The priority of fruit and according to priority generate feedback information；

Display module 730, for upon receiving the feedback information, according to the priority of search result in feedback information, Search result is shown on searched page.

Optionally, feedback information includes at least two search results and at least two search results are each self-corresponding preferential Level, display module 730, including：Sequencing unit and display unit；

Sequencing unit, for the order according to the priority of at least two search results from high to low, searched at least two Hitch fruit carries out descending processing, obtains the search result after descending；

Display unit, for the search result after the displaying descending on searched page.

Correlative detail can be with reference to shown in referring to figs. 2 to Fig. 5 embodiment of the method.Wherein, generation module 710 is additionally operable to realize Other in above method embodiment arbitrarily imply or the disclosed function related to generation step；Sending module 720 is additionally operable to reality Any implicit or disclosed function related to forwarding step of other in existing above method embodiment；Display module 730 is additionally operable to Realize that other in above method embodiment are arbitrarily implicit or the disclosed function related with showing step.

It should be noted that the device that above-described embodiment provides, when realizing its function, only with above-mentioned each functional module Division for example, in practical application, can be completed as needed and by above-mentioned function distribution by different functional modules, The internal structure of equipment is divided into different functional modules, to complete all or part of function described above.In addition, The apparatus and method embodiment that above-described embodiment provides belongs to same design, and its specific implementation process refers to embodiment of the method, this In repeat no more.

Fig. 8 shows the structured flowchart for the terminal 800 that an illustrative embodiment of the invention provides.The terminal 800 can be with It is：Smart mobile phone, tablet personal computer, MP3 player (Moving Picture Experts Group Audio Layer III, Dynamic image expert's compression standard audio aspect 3), MP4 (Moving Picture Experts Group Audio Layer IV, dynamic image expert's compression standard audio aspect 4) player, notebook computer or desktop computer.Terminal 800 be also possible to by Referred to as other titles such as user equipment, portable terminal, laptop terminal, terminal console.

Generally, terminal 800 includes：Processor 801 and memory 802.

Processor 801 can include one or more processing cores, such as 4 core processors, 8 core processors etc..Place Reason device 801 can use DSP (Digital Signal Processing, Digital Signal Processing), FPGA (Field- Programmable Gate Array, field programmable gate array), PLA (Programmable Logic Array, may be programmed Logic array) at least one of example, in hardware realize.Processor 801 can also include primary processor and coprocessor, main Processor is the processor for being handled data in the awake state, also referred to as CPU (Central Processing Unit, central processing unit)；Coprocessor is the low power processor for being handled data in the standby state. In some embodiments, processor 801 can be integrated with GPU (Graphics Processing Unit, image processor), GPU is used for rendering and drawing for the content of display required for being responsible for display screen.In some embodiments, processor 801 can also wrap AI (Artificial Intelligence, artificial intelligence) processor is included, the AI processors are used to handle relevant machine learning Calculate operation.

Memory 802 can include one or more computer-readable recording mediums, and the computer-readable recording medium can To be non-transient.Memory 802 may also include high-speed random access memory, and nonvolatile memory, such as one Or multiple disk storage equipments, flash memory device.In certain embodiments, the non-transient computer in memory 802 can Read storage medium to be used to store at least one instruction, at least one instruct is used for performed by processor 801 to realize this Shen Please in embodiment of the method provide search result methods of exhibiting.

In certain embodiments, terminal 800 is also optional includes：Peripheral interface 803 and at least one ancillary equipment. It can be connected between processor 801, memory 802 and peripheral interface 803 by bus or signal wire.Each ancillary equipment It can be connected by bus, signal wire or circuit board with peripheral interface 803.Specifically, ancillary equipment includes：Radio circuit 804th, at least one of touch display screen 805, camera 806, voicefrequency circuit 807, positioning component 808 and power supply 809.

Peripheral interface 803 can be used at least one outer of I/O (Input/Output, input/output) correlations Peripheral equipment is connected to processor 801 and memory 802.In certain embodiments, processor 801, memory 802 and ancillary equipment Interface 803 is integrated on same chip or circuit board；In some other embodiments, processor 801, memory 802 and outer Any one or two in peripheral equipment interface 803 can realize on single chip or circuit board, the present embodiment to this not It is limited.

Radio circuit 804 is used to receive and launch RF (Radio Frequency, radio frequency) signal, also referred to as electromagnetic signal.Penetrate Frequency circuit 804 is communicated by electromagnetic signal with communication network and other communication equipments.Radio circuit 804 turns electric signal Electromagnetic signal is changed to be transmitted, or, the electromagnetic signal received is converted into electric signal.Alternatively, radio circuit 804 wraps Include：Antenna system, RF transceivers, one or more amplifiers, tuner, oscillator, digital signal processor, codec chip Group, user identity module card etc..Radio circuit 804 can be carried out by least one wireless communication protocol with other terminals Communication.The wireless communication protocol includes but is not limited to：WWW, Metropolitan Area Network (MAN), Intranet, each third generation mobile communication network (2G, 3G, 4G and 5G), WLAN and/or WiFi (Wireless Fidelity, Wireless Fidelity) network.In certain embodiments, penetrate Frequency circuit 804 can also include the relevant circuits of NFC (Near Field Communication, wireless near field communication), this Application is not limited to this.

Display screen 805 is used to show UI (User Interface, user interface).The UI can include figure, text, figure Mark, video and its their any combination.When display screen 805 is touch display screen, display screen 805 also there is collection to show The surface of screen 805 or the ability of the touch signal of surface.The touch signal can be inputted to processor as control signal 801 are handled.Now, display screen 805 can be also used for providing virtual push button and/or dummy keyboard, also referred to as soft key and/or Soft keyboard.In certain embodiments, display screen 805 can be one, set the front panel of terminal 800；In other embodiments In, display screen 805 can be at least two, be separately positioned on the different surfaces of terminal 800 or in foldover design；In still other reality Apply in example, display screen 805 can be flexible display screen, be arranged on the curved surface of terminal 800 or on fold plane.Even, show Display screen 805 can also be arranged to non-rectangle irregular figure, namely abnormity screen.Display screen 805 can use LCD (Liquid Crystal Display, LCDs), OLED (Organic Light-Emitting Diode, Organic Light Emitting Diode) Prepared etc. material.

CCD camera assembly 806 is used to gather image or video.Alternatively, CCD camera assembly 806 include front camera and Rear camera.Generally, front camera is arranged on the front panel of terminal, and rear camera is arranged on the back side of terminal.One Rear camera at least two, it is main camera, depth of field camera, wide-angle camera, focal length shooting respectively in a little embodiments Head in any one, with realize main camera and the depth of field camera fusion realize background blurring function, main camera and wide-angle Camera fusion realizes that pan-shot and VR (Virtual Reality, virtual reality) shoot functions or other fusions are clapped Camera shooting function.In certain embodiments, CCD camera assembly 806 can also include flash lamp.Flash lamp can be monochromatic warm flash lamp, It can also be double-colored temperature flash lamp.Double-colored temperature flash lamp refers to the combination of warm light flash lamp and cold light flash lamp, can be used for not With the light compensation under colour temperature.

Voicefrequency circuit 807 can include microphone and loudspeaker.Microphone is used for the sound wave for gathering user and environment, and will Sound wave, which is converted to electric signal and inputted to processor 801, to be handled, or input to radio circuit 804 to realize voice communication. For stereo collection or the purpose of noise reduction, microphone can be multiple, be separately positioned on the different parts of terminal 800.Mike Wind can also be array microphone or omnidirectional's collection type microphone.Loudspeaker is then used to that processor 801 or radio circuit will to be come from 804 electric signal is converted to sound wave.Loudspeaker can be traditional wafer speaker or piezoelectric ceramic loudspeaker.When When loudspeaker is piezoelectric ceramic loudspeaker, the audible sound wave of the mankind can be not only converted electrical signals to, can also be by telecommunications Sound wave that the mankind do not hear number is converted to carry out the purposes such as ranging.In certain embodiments, voicefrequency circuit 807 can also include Earphone jack.

Positioning component 808 is used for the current geographic position of positioning terminal 800, to realize navigation or LBS (Location Based Service, location Based service).Positioning component 808 can be the GPS (Global based on the U.S. Positioning System, global positioning system), China dipper system or Russia Galileo system positioning group Part.

Power supply 809 is used to be powered for each component in terminal 800.Power supply 809 can be alternating current, direct current, Disposable battery or rechargeable battery.When power supply 809 includes rechargeable battery, the rechargeable battery can be wired charging electricity Pond or wireless charging battery.Wired rechargeable battery is the battery to be charged by Wireline, and wireless charging battery is by wireless The battery of coil charges.The rechargeable battery can be also used for supporting fast charge technology.

In certain embodiments, terminal 800 also includes one or more sensors 810.One or more sensors 810 include but is not limited to：Acceleration transducer 811, gyro sensor 812, pressure sensor 813, fingerprint sensor 814, Optical sensor 815 and proximity transducer 816.

The acceleration that acceleration transducer 811 can be detected in three reference axis of the coordinate system established with terminal 800 is big It is small.For example acceleration transducer 811 can be used for detecting component of the acceleration of gravity in three reference axis.Processor 801 can With the acceleration of gravity signal gathered according to acceleration transducer 811, control touch display screen 805 is regarded with transverse views or longitudinal direction Figure carries out the display of user interface.Acceleration transducer 811 can be also used for game or the collection of the exercise data of user.

Gyro sensor 812 can be with the body direction of detection terminal 800 and rotational angle, and gyro sensor 812 can To cooperate with collection user to act the 3D of terminal 800 with acceleration transducer 811.Processor 801 is according to gyro sensor 812 The data of collection, it is possible to achieve following function：When action induction (for example changing UI according to the tilt operation of user), shooting Image stabilization, game control and inertial navigation.

Pressure sensor 813 can be arranged on the side frame of terminal 800 and/or the lower floor of touch display screen 805.Work as pressure When sensor 813 is arranged on the side frame of terminal 800, gripping signal of the user to terminal 800 can be detected, by processor 801 The gripping signal gathered according to pressure sensor 813 carries out right-hand man's identification or prompt operation.When pressure sensor 813 is arranged on During the lower floor of touch display screen 805, the pressure operation by processor 801 according to user to touch display screen 805, realize to UI circle Operability control on face is controlled.Operability control includes button control, scroll bar control, icon control, menu At least one of control.

Fingerprint sensor 814 is used for the fingerprint for gathering user, is collected by processor 801 according to fingerprint sensor 814 The identity of fingerprint recognition user, or, the identity by fingerprint sensor 814 according to the fingerprint recognition user collected.Identifying When the identity for going out user is trusted identity, the user is authorized to perform related sensitive operation, the sensitive operation bag by processor 801 Solution lock screen is included, encryption information is checked, downloads software, payment and change setting etc..Terminal can be set in fingerprint sensor 814 800 front, the back side or side.When being provided with physical button or manufacturer Logo in terminal 800, fingerprint sensor 814 can be with Integrated with physical button or manufacturer Logo.

Optical sensor 815 is used to gather ambient light intensity.In one embodiment, processor 801 can be according to optics The ambient light intensity that sensor 815 gathers, control the display brightness of touch display screen 805.Specifically, when ambient light intensity is higher When, heighten the display brightness of touch display screen 805；When ambient light intensity is relatively low, the display for turning down touch display screen 805 is bright Degree.In another embodiment, the ambient light intensity that processor 801 can also gather according to optical sensor 815, dynamic adjust The acquisition parameters of CCD camera assembly 806.

Proximity transducer 816, also referred to as range sensor, it is generally arranged at the front panel of terminal 800.Proximity transducer 816 The distance between front for gathering user and terminal 800.In one embodiment, when proximity transducer 816 detects use When the distance between family and the front of terminal 800 taper into, touch display screen 805 is controlled from bright screen state by processor 801 Switch to breath screen state；When proximity transducer 816 detects that the distance between front of user and terminal 800 becomes larger, Touch display screen 805 is controlled to switch to bright screen state from breath screen state by processor 801.

It will be understood by those skilled in the art that the restriction of the structure shown in Fig. 8 not structure paired terminal 800, can be wrapped Include than illustrating more or less components, either combine some components or arranged using different components.

Fig. 9 is refer to, the structural representation of the server 900 provided it illustrates one exemplary embodiment of the application. The server includes the server 140 in Fig. 1.Specifically：The server 900 include CPU (CPU) 901, System storage 904 including random access memory (RAM) 902 and read-only storage (ROM) 903, and connection system are deposited Reservoir 904 and the system bus of CPU 901 905.The server 900 also includes helping each device in computer The basic input/output (I/O systems) 906 of information is transmitted between part, and for storage program area 913, application program 914 and other program modules 915 mass-memory unit 907.

The basic input/output 906 includes for the display 908 of display information and inputs letter for user The input equipment 909 of such as mouse, keyboard etc of breath.Wherein described display 908 and input equipment 909 are all by being connected to The IOC 910 of system bus 905 is connected to CPU 901.The basic input/output 906 Can also include IOC 910 for receive and handle from keyboard, mouse or electronic touch pen etc. it is multiple its The input of his equipment.Similarly, IOC 910 also provides output to display screen, printer or other kinds of defeated Go out equipment.

The mass-memory unit 907 is by being connected to the bulk memory controller (not shown) of system bus 905 It is connected to CPU 901.The mass-memory unit 907 and its associated computer-readable medium are server 900 provide non-volatile memories.That is, the mass-memory unit 907 can include such as hard disk or CD-ROI The computer-readable medium (not shown) of driver etc.

Without loss of generality, the computer-readable medium can include computer-readable storage medium and communication media.Computer Storage medium is included for information such as storage computer-readable instruction, data structure, program module or other data The volatibility and non-volatile, removable and irremovable medium that any method or technique is realized.Computer-readable storage medium includes RAM, ROM, EPROM, EEPROM, flash memory or other solid-state storages its technologies, CD-ROM, DVD or other optical storages, tape Box, tape, disk storage or other magnetic storage apparatus.Certainly, skilled person will appreciate that the computer-readable storage medium It is not limited to above-mentioned several.Above-mentioned system storage 904 and mass-memory unit 907 may be collectively referred to as memory.

According to the various embodiments of the application, the server 900 can also be arrived by network connections such as internets Remote computer operation on network.Namely server 900 can be by the network interface that is connected on the system bus 905 Unit 911 is connected to network 912, in other words, can also be connected to using NIU 911 other kinds of network or Remote computer system (not shown).

Optionally, be stored with least one instruction in the memory, at least one instruction loaded by processor and performed with Realize the feedback method for the search result that above-mentioned each embodiment of the method provides.

Above-mentioned the embodiment of the present application sequence number is for illustration only, does not represent the quality of embodiment.

One of ordinary skill in the art will appreciate that realize feedback method and the displaying side of the search result of above-described embodiment All or part of step can be completed by hardware in method, by program the hardware of correlation can also be instructed to complete, described Program can be stored in a kind of computer-readable recording medium, storage medium mentioned above can be read-only storage, Disk or CD etc..In other words, at least one instruction is stored with the storage medium, at least one instruction is loaded by processor And perform to realize the feedback method and methods of exhibiting of the search result described in each embodiment of the method as described above.

The foregoing is only the preferred embodiment of the application, not to limit the application, it is all in spirit herein and Within principle, any modification, equivalent substitution and improvements made etc., it should be included within the protection domain of the application.

Claims

1. a kind of feedback method of search result, it is characterised in that methods described includes：

User's portrait corresponding with user account number mark is obtained, the user draws a portrait inclined for the interest of instruction user account number Good information；

According to the relevance between the attribute information of the search result and user portrait, the excellent of the search result is determined First level；

According to the priority of the search result, to the terminal send feedback information, the feedback information is used to indicate in institute State and show the search result according to the order of the priority from high to low in terminal.

2. according to the method for claim 1, it is characterised in that user's portrait includes：

N number of each self-corresponding first weighted value of video display type, the video display type correspond to K interest dimension, the interest dimension Corresponding M dimensional characteristics；And

Wherein, described N, K and M are positive integer.

3. according to the method for claim 2, it is characterised in that described to obtain user's picture corresponding to the user account number mark Before picture, in addition to：

According to N number of first accounting, first object weighted value of the user account number in the predetermined time interval is determined；

4. according to the method for claim 2, it is characterised in that described to obtain user's picture corresponding to the user account number mark Before picture, in addition to：

According to N*K*M second accountings, second target power of the user account number in the predetermined time interval is determined Weight values；

5. according to the method for claim 1, it is characterised in that the attribute information according to the search result with it is described Relevance between user's portrait, the priority of the search result is determined, including：

Obtain the attribute information of each search result in the search result；

According to the attribute weight, the priority of the search result, the default recommendation mould are obtained using default recommended models Type is used for the Behavior law that expression is trained to obtain based on historical search result.

6. according to the method for claim 5, it is characterised in that the attribute information of the search result includes matching degree, institute The matching degree that matching degree is used to indicate the search result and keyword in the searching request is stated,

It is described to be drawn a portrait according to the attribute information and the user, attribute weight corresponding to the attribute information is determined, including：

7. according to the method for claim 5, it is characterised in that the attribute information of the search result includes foundation characteristic, The foundation characteristic is used for the build-in attribute for indicating the search result, and the foundation characteristic includes the scoring of the search result The sub- foundation characteristic of at least one of value, program request amount, video resolution and video show time,

According to the sub- foundation characteristic and the first default corresponding relation, it is determined that the second Attribute Weight corresponding with the sub- foundation characteristic Weight, the described first default corresponding relation include the corresponding relation of the sub- foundation characteristic and second attribute weight.

8. according to the method for claim 5, it is characterised in that it is special that the attribute information of the search result includes video preference Sign, the video preference profiles include the dimensional characteristics of video display type corresponding with the search result and interest dimension, described Interest dimension includes at least one of video display label, age, performer, director and area,

Determined in being drawn a portrait from the user second corresponding to the first weighted value corresponding to the video display type and the dimensional characteristics Weighted value；

First weighted value is multiplied with second weighted value, obtains the 3rd attribute weight corresponding to the dimensional characteristics.

9. according to the method for claim 5, it is characterised in that it is special that the attribute information of the search result includes video preference To seek peace other features, other described features include matching degree and/or foundation characteristic,

Interest dimension corresponding to video display type and the search result of the video preference profiles including the search result Dimensional characteristics, interest dimension corresponding to the search result are included in video display label, age, performer, director and area at least It is a kind of；

The foundation characteristic is used for the build-in attribute for indicating the search result, and the foundation characteristic includes the search result At least one of score value, program request amount, video resolution and video show time foundation characteristic.

10. according to the method for claim 5, it is characterised in that it is described according to the attribute weight, using default recommendation mould Type obtains the priority of the search result, including：

The default recommended models are obtained, the default recommended models are to train to obtain according at least one set of historical search result group , the historical behavior data group includes：Historical search keyword, historical search result and historical feedback data, the history Feedback data is used to indicate satisfaction corresponding to the historical search result；

11. according to the method for claim 10, it is characterised in that the attribute weight includes the first category corresponding to matching degree At least one of foundation characteristic the second attribute weight corresponding to sub- foundation characteristic of property weight, the attribute information, and it is described 3rd attribute weight corresponding at least one of the video preference profiles of attribute information dimensional characteristics, it is described according to the Attribute Weight Weight, the characteristic vector of the search result is determined, including：

First attribute weight, second attribute weight and the 3rd attribute weight are defined as the search result Characteristic vector.

12. method according to any one of claims 1 to 11, it is characterised in that described according to the preferential of the search result Level, to the terminal send feedback information, including：

According to the order of the priority of the search result from high to low, descending is carried out to the search result and handles to obtain descending Search result afterwards；

First feedback information is sent to the terminal.

13. method according to any one of claims 1 to 11, it is characterised in that described according to the preferential of the search result Level, to the terminal send feedback information, including：

According to the priority of the search result, the second feedback information is generated, second feedback information includes the search and tied Fruit and its corresponding priority；

Second feedback information is sent to the terminal.

14. a kind of methods of exhibiting of search result, it is characterised in that methods described includes：

When inputting search key in searched page, according to user's account corresponding to the search key and the searched page Number mark generation searching request；

The searching request is sent to server according to search instruction, the searching request be used to indicating the server determine with The search result and user corresponding with user account number mark portrait of the search key matching, and searched according to described The attribute information of hitch fruit and for instruction user account number interest preference information the user portrait between relevance, really Determine the priority of the search result and feedback information is generated according to the priority；

When receiving the feedback information, according to the priority of search result in the feedback information, in the search page The search result is shown on face.

15. according to the method for claim 14, it is characterised in that the feedback information includes described at least two search and tied Fruit and each self-corresponding priority of at least two search result, it is described when receiving the feedback information, according to described The priority of search result in feedback information, the search result is shown on the searched page, including：

According to the order of the priority of at least two search result from high to low, at least two search result is carried out Descending processing, obtains the search result after descending；

The search result after the descending is shown on the searched page.

16. a kind of feedback device of search result, it is characterised in that described device includes：

Receiving module, for the searching request of receiving terminal transmission, search key and user are carried in the searching request Account number identifies；

Acquisition module, for obtaining user's portrait corresponding with user account number mark, the user draws a portrait for indicating to use The interest preference information of family account number；

The determining module, it is additionally operable to according to the relevance between the attribute information of the search result and user portrait, Determine the priority of the search result；

Feedback module, for the priority according to the search result, to the terminal send feedback information, the feedback information For indicating to show the search result according to the order of the priority from high to low in the terminal.

17. device according to claim 16, it is characterised in that user's portrait includes：

Wherein, described N, K and M are positive integer.

18. device according to claim 16, it is characterised in that the feedback module, including：Generation unit and second anti- Present unit；

Generation unit, for the priority according to the search result, generate the second feedback information, the second feedback information bag Include the search result and its corresponding priority；

Second feedback unit, for sending second feedback information to the terminal.

19. a kind of exhibiting device of search result, it is characterised in that described device includes：

Generation module, for when searched page input search key when, according to the search key and the search page User account number mark generation searching request corresponding to face；

Sending module, for sending the searching request to server according to search instruction, the searching request is used to indicate institute Server is stated to determine to draw a portrait with the search result of search key matching and user corresponding with user account number mark, And the attribute information according to the search result and the user of the interest preference information for instruction user account number draw a portrait Between relevance, determine the search result priority and according to the priority generate feedback information；

Display module, for when receiving the feedback information, according to the priority of search result in the feedback information, The search result is shown on the searched page.

20. device according to claim 19, it is characterised in that the feedback information include at least two search results and Each self-corresponding priority of at least two search results, the display module, including：Sequencing unit and display unit；

The sequencing unit, for the order according to the priority of at least two search result from high to low, to it is described extremely Few two search results carry out descending processing, obtain the search result of descending；

21. a kind of server, it is characterised in that the server includes processor and memory, is stored with the memory At least one instruction, at least one instruction are loaded by the processor and performed to realize as claim 1 to 13 is any The feedback method of described search result.

22. a kind of terminal, it is characterised in that the terminal includes processor and memory, is stored with least in the memory One instruction, at least one instruction are loaded as the processor and performed to realize searching as described in claims 14 or 15 The methods of exhibiting of hitch fruit.

23. a kind of computer-readable recording medium, it is characterised in that at least one instruction, institute are stored with the storage medium At least one instruction is stated to be loaded by the processor and performed to realize such as the search result as described in claim 1 to 13 is any Feedback method.

24. a kind of computer-readable recording medium, it is characterised in that at least one instruction, institute are stored with the storage medium At least one instruction is stated to be loaded as the processor and performed to realize the exhibition of the search result as described in claims 14 or 15 Show method.