CN103324742A - Method and equipment for recommending keywords - Google Patents
Method and equipment for recommending keywords Download PDFInfo
- Publication number
- CN103324742A CN103324742A CN2013102696431A CN201310269643A CN103324742A CN 103324742 A CN103324742 A CN 103324742A CN 2013102696431 A CN2013102696431 A CN 2013102696431A CN 201310269643 A CN201310269643 A CN 201310269643A CN 103324742 A CN103324742 A CN 103324742A
- Authority
- CN
- China
- Prior art keywords
- keyword
- object information
- word
- prioritization
- feedback
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Landscapes
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention provides a method and equipment for recommending keywords. The method for recommending the keywords includes receiving seed data from a user; providing at least one keyword to the user on the basis of the seed data; receiving operation of the user on the first keywords; updating the first keywords according to result information of the operation.
Description
Technical field
The present invention relates to information search and data mining field.More particularly, the method and apparatus that relates to a kind of recommended keywords.
Background technology
Along with the continuous development of infotech, search engine is used more and more widely.By search engine, general search subscriber (that is general netizen) can search relevant information.In addition, the advertisement promotion user also utilizes search engine that the advertisement promotion service is provided.Providing usually of advertisement promotion service is associated with keyword, when search subscriber is imported certain keyword and searched for, can provide and this keyword associated advertisement like this.For this reason, the advertisement promotion user need know the keyword that is associated with oneself advertisement to be promoted.Like this, the technology of a kind of recommended keywords of needs is helped the advertisement promotion user and is obtained suitable keyword.
In existing keyword recommended technology, obtain relevant keyword by input with waiting the relevant seed information analysis of promoting advertisement usually.For this reason, need a large amount of historical behavior data of advertisement promotion user to carry out calculated off-line as seed information.Yet when different advertisement promotion users' business is more close, the recommended keywords that analysis obtains based on seed information can't embody different advertisement promotion user's oneself characteristics with advolution.
In addition, less or when not having, the keyword that is difficult to provide accurately suitable is to the advertisement promotion user when an advertisement promotion user's historical behavior data.That is, the cold start-up problem face recommended in keyword.
Correspondingly, owing to advertisement to be promoted can not be distributed to exactly in keyword, general search subscriber also can't be obtained promotion message accurately by keyword.
Therefore, the technology that needs a kind of recommended keywords that can above-mentioned at least one technical matters of technology.
Summary of the invention
The object of the present invention is to provide a kind of method and apparatus of recommended keywords to solve above-mentioned at least one technical matters.
To achieve these goals, an aspect of of the present present invention provides a kind of method of recommended keywords, comprising: receive seed data from the user; Provide at least one first keyword to the user based on seed data; Receive the user to the operation of first keyword; Object information according to described operation upgrades first keyword.
Alternatively, provide at least one first keyword to comprise to user's step based on seed data: to obtain second keyword based on seed data; Select second keyword of predetermined quantity to offer the user as first keyword.
Alternatively, the step of upgrading first keyword according to the object information of described operation comprises: adjust the prioritization of the 3rd keyword according to the object information of described operation, wherein, the 3rd keyword is to select remaining second keyword after first keyword; In response to another operation from the user, select the 3rd keyword of the forward predetermined quantity of prioritization to offer the user as first keyword.
Alternatively, the step of upgrading first keyword according to the object information of described operation comprises: the prioritization of adjusting first keyword according to the object information of described operation.
Alternatively, the user comprises positive feedback operation and negative feedback operation to the operation of described at least one first keyword.
Alternatively, positive feedback operation comprises: to the selection of predetermined first keyword and/or to the screening that comprises of first keyword.
Alternatively, negative feedback operation comprises: to not the selecting of predetermined first keyword, to the shielding of predetermined first keyword, in the exclude filter of first keyword at least one.
Alternatively, be described predetermined first keyword with the corresponding object information of selection to predetermined first keyword.
Alternatively, with first keyword comprised the corresponding object information of screening for being used for comprising the information of screening.
Alternatively, with do not select corresponding object information for there not being selecteed first keyword to predetermined first keyword.
Alternatively, be conductively-closed and can't selecteed first keyword with the corresponding object information of shielding to predetermined first keyword.
Alternatively, with to the corresponding object information of the exclude filter of first keyword for being used for carrying out the information of exclude filter.
Alternatively, come according to first keyword of the object information adjustment conduct of described operation keyword to be sorted or the prioritization of the 3rd keyword by following manner: obtain the relevant keyword of the object information with operation in the keyword to be sorted; The prioritization of the keyword that adjustment is obtained in keyword to be sorted.
Alternatively, the step of the prioritization of the keyword that adjustment is obtained in keyword to be sorted comprises: improve the prioritization of the keyword relevant with object information positive feedback operation in the keyword to be sorted, reduce the prioritization of the relevant keyword of the object information of operating with negative feedback in the keyword to be sorted.
Alternatively, obtaining the step with the relevant keyword of object information operation in the keyword to be sorted comprises: obtain the feedback word, organize at least one word structure, attribute characteristics, the label from the object information of operation; From keyword to be sorted, obtain and feed back at least one relevant keyword in word, group word structure, attribute characteristics, the label.
Alternatively, the step that obtains the feedback word from the object information of operation comprises: the object information of operation is cut word to obtain the feedback word.
Alternatively, comprise from the step of object information acquisition group word structure of operation: the group word structure to the object information of operation is analyzed, with the group word structure that obtains to exist.
Alternatively, the step that obtains label from the object information of operation comprises: the label to the object information of operation is analyzed, with the label that obtains to exist.
Alternatively, the step that obtains the attribute characteristics from the object information of operating comprises: the size of the attribute of the object information of analysis operation, and to determine the attribute characteristics.
Alternatively, the attribute characteristics comprise the high or low of volumes of searches.
Alternatively, in the step of the prioritization of keyword in keyword to be sorted that adjustment is obtained: the word frequency of feedback word is more high, and then the prioritization of the keyword relevant with this feedback word is changed more manyly by adjustment.
Alternatively, in the step of the prioritization of keyword in keyword to be sorted that adjustment is obtained: the quantity of the group word structure of a type is more many, and then the prioritization of the keyword relevant with the group word structure of the type is changed more manyly by adjustment.
Alternatively, in the step of the prioritization of keyword in keyword to be sorted that adjustment is obtained: the number of tags of a type is more high, and then the prioritization of the keyword relevant with the label of the type is changed more manyly by adjustment.
Alternatively, the step that obtains the feedback word from the object information of operation also comprises: the feedback word of selecting predetermined quantity according to word frequency order from high to low from the feedback word that obtains.
Alternatively, the step of object information acquisition group word structure from operation also comprises: the group word structure of selecting the type of predetermined quantity according to the quantity order from high to low of the phrase structure of each type.
Alternatively, the step that obtains label from the object information of operation also comprises: the label of selecting the type of predetermined quantity according to the number of tags order from high to low of each type.
Alternatively, described method also comprises: receive the user to the operation again of first keyword; Upgrade first keyword according to the described object information of operation again.
To achieve these goals, another aspect of the present invention provides a kind of equipment of recommended keywords, comprising: receiving element receives seed data from the user; Recommendation unit provides at least one first keyword to the user based on seed data; Operating unit receives the user to the operation of first keyword; Updating block upgrades first keyword according to the object information of described operation.
Alternatively, recommendation unit comprises: first recommendation unit obtains second keyword based on seed data; Second recommendation unit selects second keyword of predetermined quantity to offer the user as first keyword.
Alternatively, updating block is adjusted the prioritization of the 3rd keyword according to the object information of described operation, wherein, the 3rd keyword is to select first keyword remaining second keyword afterwards, wherein, described equipment also comprises selected cell, in response to another operation from the user, selects the 3rd keyword of the forward predetermined quantity of prioritization to offer the user as first keyword.
Alternatively, updating block is adjusted the prioritization of first keyword according to the object information of described operation.
Alternatively, the user comprises positive feedback operation and negative feedback operation to the operation of described at least one first keyword.
Alternatively, positive feedback operation comprises: to the selection of predetermined first keyword and/or to the screening that comprises of first keyword.
Alternatively, negative feedback operation comprises: to not the selecting of predetermined first keyword, to the shielding of predetermined first keyword, in the exclude filter of first keyword at least one.
Alternatively, be described predetermined first keyword with the corresponding object information of selection to predetermined first keyword.
Alternatively, with first keyword comprised the corresponding object information of screening for being used for comprising the information of screening.
Alternatively, with do not select corresponding object information for there not being selecteed first keyword to predetermined first keyword.
Alternatively, be conductively-closed and can't selecteed first keyword with the corresponding object information of shielding to predetermined first keyword.
Alternatively, with to the corresponding object information of the exclude filter of first keyword for being used for carrying out the information of exclude filter.
Alternatively, updating block comprises: extraction unit, obtain the relevant keyword of the object information with operation in the keyword to be sorted; Adjustment unit is adjusted the prioritization of keyword in keyword to be sorted of obtaining.
Alternatively, adjustment unit improves the prioritization of the relevant keyword of the object information with the positive feedback operation in the keyword to be sorted, reduces the prioritization of the relevant keyword of the object information of operating with negative feedback in the keyword to be sorted.
Alternatively, extraction unit obtains feedback word, group word structure, attribute characteristics, the label at least one from the object information of operation, and obtains from keyword to be sorted and feed back at least one relevant keyword in word, group word structure, attribute characteristics, the label.
Alternatively, extraction unit is cut word to obtain the feedback word to the object information of operation.
Alternatively, extraction unit is analyzed the group word structure of the object information of operation, with the group word structure that obtains to exist.
Alternatively, extraction unit is analyzed the label of the object information of operation, with the label that obtains to exist.
Alternatively, the size of the attribute of the object information of extraction unit analysis operation is to determine the attribute characteristics.
Alternatively, the attribute characteristics comprise the high or low of volumes of searches.
Alternatively, the word frequency of feedback word is more high, and then the prioritization of the keyword relevant with this feedback word is changed more manyly by adjustment unit.
Alternatively, the quantity of the group word structure of a type is more high, and then the prioritization of the keyword relevant with the group word structure of the type is changed more manyly by adjustment unit.
Alternatively, the number of tags of a type is more many, and then the prioritization of the keyword relevant with the label of the type is changed more manyly by adjustment unit.
Alternatively, extraction unit is selected the feedback word of predetermined quantity from the feedback word that obtains according to word frequency order from high to low.
Alternatively, extraction unit is selected the group word structure of the type of predetermined quantity according to the quantity order from high to low of the phrase structure of each type.
Alternatively, extraction unit is selected the label of the type of predetermined quantity according to the number of tags order from high to low of each type.
Alternatively, operating unit receives the user to the operation again of first keyword, and updating block upgrades first keyword according to the described object information of operation again.
According to the method and apparatus of recommended keywords of the present invention, can upgrade the keyword of recommendation according to the user in real time to the operation that keyword is provided based on seed data.
In addition, according to the method and apparatus of recommended keywords of the present invention, when seed data more after a little while, accurate suitable keyword can be offered the user.
In addition, according to the method and apparatus of recommended keywords of the present invention, owing to advertisement to be promoted can be distributed to exactly in keyword, therefore general search subscriber can be obtained promotion message accurately by keyword.
Will be in ensuing description part set forth the present invention other aspect and/or advantage, some will be clearly by describing, and perhaps can learn through enforcement of the present invention.
Description of drawings
By the detailed description of carrying out below in conjunction with accompanying drawing, above and other objects of the present invention, characteristics and advantage will become apparent, wherein:
Fig. 1 illustrates the process flow diagram of the method for recommended keywords according to an embodiment of the invention;
Fig. 2 illustrates the process flow diagram of the method for recommended keywords according to another embodiment of the present invention;
Fig. 3 illustrates the block diagram of the equipment of recommended keywords according to an embodiment of the invention;
Fig. 4 illustrates the block diagram of the equipment of recommended keywords according to another embodiment of the present invention.
Embodiment
Below, describe different embodiments of the invention with reference to the accompanying drawings in detail.
Fig. 1 illustrates the process flow diagram of the method for recommended keywords according to an embodiment of the invention.
As shown in Figure 1, in step 101, receive seed data from the user.Seed data is the data relevant with the keyword of hope recommendation.For example, when hope obtained the keyword relevant with certain advertisement, seed data can be the industry of this advertisement, concrete product or the description of service etc.
In step 102, provide at least one first keyword to the user based on seed data.Can utilize existing keyword recommended technology to come to obtain keyword as first keyword based on seed data.For example, at least one first keyword with certain right of priority order can be offered the user.For example, can first keyword be shown to the user with the form of tabulating by display.Should be appreciated that, also can by other means first keyword be offered the user, thereby be known by the user.
In step 103, receive the user to the operation of first keyword.Can be for example to the selection of keyword, screening, shielding etc. to the operation of keyword.
In an example of the present invention, will be divided into positive feedback operation and negative feedback operation to the operation of keyword.Positive feedback operation can for example comprise: to the selection of keyword and/or to the screening that comprises of keyword.
Here, comprise the keyword that screening refers to filter out the screening conditions (key word that for example, is used for screening) that comprise input.
Negative feedback operation can for example comprise: to not the selecting of keyword, to the shielding of keyword, in the exclude filter of keyword at least one.
Here, can't be selected thereby the shielding of keyword is referred to temporarily will be scheduled to the keyword shielding, exclude filter refers to filter out the keyword of the screening conditions (key word that for example, is used for screening) that do not comprise input.
Should be appreciated that positive feedback operation of the present invention and negative feedback operation are not limited to top example.Can embody and to operate as positive feedback any operation of the positive attitude of predetermined keyword, can embody and to operate as negative feedback any operation of the passive attitude of predetermined keyword.
In step 104, upgrade first keyword according to the object information in the operation of step 103.
For example, adjust the prioritization of first keyword according to the object information of described operation, thereby upgrade the recommended priority of first keyword.Specifically, obtain as the first relevant keyword of the object information with operation in first keyword of keyword to be sorted, adjust the ordering of first keyword in first keyword to be sorted of obtaining.
Under the situation that will be divided into positive feedback operation and negative feedback operation to the operation of keyword, improve the ordering of first keyword relevant with object information positive feedback operation in first keyword to be sorted, reduce the ordering of the first relevant keyword of the object information of operating with negative feedback in first keyword to be sorted.
Can utilize existing data mining technology to obtain the keyword relevant with object information.
In addition, one embodiment of the present of invention provide a kind of technology of obtaining the keyword relevant with object information.Corresponding to above-mentioned positive feedback operation and negative feedback operation, and to the keyword of the corresponding object information of the selection of keyword for selecting; With keyword comprised the corresponding object information of screening for being used for comprising the information of screening; With do not select corresponding object information for there not being selecteed keyword to keyword; With to the corresponding object information of the shielding of keyword be conductively-closed and can't selecteed keyword; With to the corresponding object information of the exclude filter of keyword for being used for carrying out the information of exclude filter.
Can based on the relevant feedback word of object information of operation, group word structure, attribute characteristics, label at least one obtain first keyword of being correlated with object information operation as in first keyword of keyword to be sorted.
Specifically, can be at first obtain feedback word, group word structure, attribute characteristics, the label at least one from the object information of operation, from keyword to be sorted, obtain subsequently and feed back at least one first relevant keyword in word, group word structure, attribute characteristics, the label.
Here, the feedback word refers to the keyword that obtains of object information from operation.Can be by the object information of operation being cut word to obtain the feedback word.Alternatively, also can select the feedback word of predetermined quantity according to word frequency (that is, the number of times that in all feedback word candidates that obtain, occurs) order from high to low, rather than use the feedback word of all acquisitions.
Alternatively, the word frequency of feedback word is more high, and then the ordering of the keyword relevant with this feedback word is changed more manyly by adjustment in step 104.For example, for the feedback word that the object information from the positive feedback operation obtains, the word frequency of a feedback word is more high, and then the ordering of the keyword relevant with this feedback word is enhanced more manyly by adjustment; For the feedback word that the object information from the negative feedback operation obtains, the word frequency of a feedback word is more high, and then the ordering of the keyword relevant with this feedback word is lowered more manyly by adjustment.
Phrase structure refers to the language construct structure (for example, V-O construction, parallel construction, M-D (modifier-head) construction, subject-predicate phrase etc.) of the object information operated.Can be by the group word structure of object information of operation be analyzed, with the group word structure that obtains to exist.Alternatively, also can select the group word structure of the type of predetermined quantity according to quantity (that is, the quantity that the phrase structure of each type exists) order from high to low in all phrase structures that obtain, rather than use all types of groups of word structures that obtain.
Alternatively, the quantity of one type phrase structure is more many, and then the ordering of the keyword relevant with this kind phrase structure is changed more manyly by adjustment in step 104.For example, for the phrase structure that the object information from the positive feedback operation obtains, the quantity of one type phrase structure is more many, and then the ordering of the keyword relevant with the phrase structure of the type is enhanced more manyly by adjustment; For the phrase structure that the object information from the negative feedback operation obtains, the quantity of one type phrase structure is more many, and then the ordering of the keyword relevant with the phrase structure of the type is lowered more manyly by adjustment.
The attribute characteristics refer to the high or low characteristics of the attribute that can quantize (for example, volumes of searches, degree of contention) of the object information operated.For example, the attribute greater than predetermined threshold is considered to the attribute characteristics for high; It is low that the attribute that is not more than predetermined threshold is considered to the attribute characteristics.
Keyword relevant with the attribute characteristics in step 104 is adjusted.For example, for the attribute characteristics (for example, volumes of searches height or volumes of searches are low) that the object information from the positive feedback operation obtains, the ordering that improves the keyword relevant with these attribute characteristics; Attribute characteristics (for example, volumes of searches height or volumes of searches are low) for the object information from the negative feedback operation obtains reduce the ordering of the keyword relevant with these attribute characteristics.
Label refers to the various forms of labels (for example, tag along sort, geographical labels etc.) of the object information operated.Can be by the label of object information of operation be analyzed, with the label that obtains to exist.Alternatively, also can select the label of the type of predetermined quantity according to quantity (that is, the quantity that the label of each type exists) order from high to low in all labels that obtain, rather than use all types of labels that obtain.
Alternatively, one type number of tags is more many, and then the ordering of the keyword relevant with this kind label is changed more manyly by adjustment in step 104.For example, for the label that the object information from the positive feedback operation obtains, one type number of tags is more many, and then the ordering of the keyword relevant with the label of the type is enhanced more manyly by adjustment; For the label that the object information from the negative feedback operation obtains, one type number of tags is more many, and then the ordering of the keyword relevant with the label of the type is lowered more manyly by adjustment.
In addition, described method can be returned step 103 after step 104, continues to receive the user to the operation of first keyword, and the operation below carrying out subsequently.
Fig. 2 illustrates the process flow diagram of the method for recommended keywords according to another embodiment of the present invention.
As shown in Figure 2, in step 201, receive seed data.Seed data is the data relevant with the keyword of hope recommendation.For example, when hope obtained the keyword relevant with certain advertisement, seed data can be the industry of this advertisement, concrete product or the description of service etc.
In step 202, obtain at least one second keyword based on seed data.Can utilize existing keyword recommended technology to come to obtain keyword as second keyword based on seed data.For example, can obtain to have at least one second keyword of certain right of priority order.
In step 203, select second keyword of predetermined quantity to offer the user as first keyword.For example, can select second keyword of predetermined quantity as first keyword according to right of priority order from high to low.For example, can first keyword be shown to the user with the form of tabulating by display.Should be appreciated that, also can by other means first keyword be offered the user, thereby be known by the user.
In step 204, receive the user to the operation of first keyword.Can be for example to the selection of keyword, screening, shielding etc. to the operation of keyword.
In an example of the present invention, will be divided into positive feedback operation and negative feedback operation to the operation of keyword.Positive feedback operation can for example comprise: to the selection of keyword and/or to the screening that comprises of keyword.
Here, comprise the keyword that screening refers to filter out the screening conditions (key word that for example, is used for screening) that comprise input.
Negative feedback operation can for example comprise: to not the selecting of keyword, to the shielding of keyword, in the exclude filter of keyword at least one.
Here, can't be selected thereby the shielding of keyword is referred to temporarily will be scheduled to the keyword shielding, exclude filter refers to filter out the keyword of the screening conditions (key word that for example, is used for screening) that do not comprise input.
Should be appreciated that positive feedback operation of the present invention and negative feedback operation are not limited to top example.Can embody and to operate as positive feedback any operation of the positive attitude of predetermined keyword, can embody and to operate as negative feedback any operation of the passive attitude of predetermined keyword.
In step 205, upgrade the 3rd keyword according to the object information in the operation of step 204.Here, the 3rd keyword is remaining second keyword after selecting first keyword among second keyword.
For example, adjust the prioritization of the 3rd keyword according to the object information of described operation, thereby upgrade the recommended priority of the 3rd keyword.Specifically, obtain as the 3rd relevant keyword of the object information with operation in the 3rd keyword of keyword to be sorted, adjust the ordering of the 3rd keyword in the 3rd keyword to be sorted of obtaining.
Under the situation that will be divided into positive feedback operation and negative feedback operation to the operation of keyword, improve the ordering of three keyword relevant with object information positive feedback operation in the 3rd keyword to be sorted, reduce the ordering of the 3rd relevant keyword of the object information of operating with negative feedback in the 3rd keyword to be sorted.
Can utilize existing data mining technology to obtain the keyword relevant with object information.
In addition, one embodiment of the present of invention provide a kind of technology of obtaining the keyword relevant with object information.Corresponding to above-mentioned positive feedback operation and negative feedback operation, and to the keyword of the corresponding object information of the selection of keyword for selecting; With keyword comprised the corresponding object information of screening for being used for comprising the information of screening; With do not select corresponding object information for there not being selecteed keyword to keyword; With to the corresponding object information of the shielding of keyword be conductively-closed and can't selecteed keyword; With to the corresponding object information of the exclude filter of keyword for being used for carrying out the information of exclude filter.
Can based on the relevant feedback word of object information of operation, group word structure, attribute characteristics, label at least one obtain the 3rd keyword of being correlated with object information operation as in the 3rd keyword of keyword to be sorted.
Specifically, can be at first obtain feedback word, group word structure, attribute characteristics, the label at least one from the object information of operation, from keyword to be sorted, obtain subsequently and feed back at least one the 3rd relevant keyword in word, group word structure, attribute characteristics, the label.
Here, the feedback word refers to the keyword that obtains of object information from operation.Can be by the object information of operation being cut word to obtain the feedback word.Alternatively, also can select the feedback word of predetermined quantity according to word frequency (that is, the number of times that in all feedback word candidates that obtain, occurs) order from high to low, rather than use the feedback word of all acquisitions.
Alternatively, the word frequency of feedback word is more high, and then the ordering of the keyword relevant with this feedback word is changed more manyly by adjustment in step 205.For example, for the feedback word that the object information from the positive feedback operation obtains, the word frequency of a feedback word is more high, and then the ordering of the keyword relevant with this feedback word is enhanced more manyly by adjustment; For the feedback word that the object information from the negative feedback operation obtains, the word frequency of a feedback word is more high, and then the ordering of the keyword relevant with this feedback word is lowered more manyly by adjustment.
Phrase structure refers to the language construct structure (for example, V-O construction, parallel construction, M-D (modifier-head) construction, subject-predicate phrase etc.) of the object information operated.Can be by the group word structure of object information of operation be analyzed, with the group word structure that obtains to exist.Alternatively, also can select the group word structure of the type of predetermined quantity according to quantity (that is, the quantity that the phrase structure of each type exists) order from high to low in all phrase structures that obtain, rather than use all types of groups of word structures that obtain.
Alternatively, the quantity of one type phrase structure is more many, and then the ordering of the keyword relevant with this kind phrase structure is changed more manyly by adjustment in step 205.For example, for the phrase structure that the object information from the positive feedback operation obtains, the quantity of one type phrase structure is more many, and then the ordering of the keyword relevant with the phrase structure of the type is enhanced more manyly by adjustment; For the phrase structure that the object information from the negative feedback operation obtains, the quantity of one type phrase structure is more many, and then the ordering of the keyword relevant with the phrase structure of the type is lowered more manyly by adjustment.
The attribute characteristics refer to the high or low characteristics of the attribute that can quantize (for example, volumes of searches, degree of contention) of the object information operated.For example, it is high being considered to the attribute characteristics greater than the attribute of predetermined threshold; It is low that the attribute that is not more than predetermined threshold is considered to the attribute characteristics.
Keyword relevant with the attribute characteristics in step 205 is adjusted.For example, for the attribute characteristics (for example, volumes of searches height or volumes of searches are low) that the object information from the positive feedback operation obtains, the ordering that improves the keyword relevant with these attribute characteristics; Attribute characteristics (for example, volumes of searches height or volumes of searches are low) for the object information from the negative feedback operation obtains reduce the ordering of the keyword relevant with these attribute characteristics.
Label refers to the various forms of labels (for example, tag along sort, geographical labels etc.) of the object information operated.Can be by the label of object information of operation be analyzed, with the label that obtains to exist.Alternatively, also can select the label of the type of predetermined quantity according to quantity order from high to low, rather than use all types of labels that obtain.
Alternatively, one type number of tags is more many, and then the ordering of the keyword relevant with this kind label is changed more manyly by adjustment in step 205.For example, for the label that the object information from the positive feedback operation obtains, one type number of tags is more many, and then the ordering of the keyword relevant with the label of the type is enhanced more manyly by adjustment; For the label that the object information from the negative feedback operation obtains, one type number of tags is more many, and then the ordering of the keyword relevant with the label of the type is lowered more manyly by adjustment.
In step 206, in response to another operation from the user, among the 3rd keyword that has upgraded, select the 3rd keyword of the forward predetermined quantity of prioritization to offer the user as first keyword.For example, be presented at before or after the first previous keyword with the tabular form order.Should be appreciated that at this moment, described selecteed the 3rd keyword is not re-used as the 3rd keyword and exists.Here, another operation of user can be the various operations by user interface (for example, touch-screen, keyboard, mouse etc.) input.
In addition, described method can continue in response to another operation from the user after 206, execution in step 206, and the 3rd keyword of the predetermined quantity that other prioritization is forward offers the user as first keyword.
In addition, described method can be returned step 204 after step 205 or 206, continues to receive the user to the operation of first keyword, and carries out operation subsequently subsequently.In addition, also can after step 205 or 206, judge whether to receive the user to the operation of first keyword and/or judge whether to receive described another operation; When receiving the user to the operation of first keyword, carry out step 205; When not receiving the operation of user to first keyword, continue above-mentioned judgement or end operation; When receiving described another when operation, carry out step 206; When not receiving described another when operation, continue above-mentioned judgement or end operation.
Fig. 3 illustrates the block diagram of the equipment of recommended keywords according to an embodiment of the invention.
As shown in Figure 3, the equipment 300 of recommended keywords comprises according to an embodiment of the invention: receiving element 310, recommendation unit 320, operating unit 330, updating block 340.
Receiving element 310 receives the seed data of user's input.
Recommendation unit 320 provides at least one first keyword to the user based on seed data.Can utilize existing keyword recommended technology to come to obtain keyword as first keyword based on seed data.
In an example of the present invention, will be divided into positive feedback operation and negative feedback operation to the operation of keyword.Positive feedback operation can for example comprise: to the selection of keyword and/or to the screening that comprises of keyword.
Here, comprise the keyword that screening refers to filter out the screening conditions (key word that for example, is used for screening) that comprise input.
Negative feedback operation can for example comprise: to not the selecting of keyword, to the shielding of keyword, in the exclude filter of keyword at least one.
Here, can't be selected thereby the shielding of keyword is referred to temporarily will be scheduled to the keyword shielding, exclude filter refers to filter out the keyword of the screening conditions (key word that for example, is used for screening) that do not comprise input.
Should be appreciated that positive feedback operation of the present invention and negative feedback operation are not limited to top example.Can embody and to operate as positive feedback any operation of the positive attitude of predetermined keyword, can embody and to operate as negative feedback any operation of the passive attitude of predetermined keyword.
Updating block 340 upgrades first keyword according to the object information of operation.
For example, updating block 340 is adjusted the prioritization of first keyword according to the object information of described operation, thereby upgrades the recommended priority of first keyword.
Specifically, updating block 340 comprises extraction unit 341 and adjustment unit 342.
Extraction unit 341 obtains as the first relevant keyword of the object information with operation in first keyword of keyword to be sorted.Adjustment unit 342 is adjusted the ordering of first keyword in first keyword to be sorted of obtaining.
Under the situation that will be divided into positive feedback operation and negative feedback operation to the operation of keyword, adjustment unit 342 improves the ordering of the first relevant keyword of the object information with the positive feedback operation in first keyword to be sorted, reduces the ordering of the first relevant keyword of the object information of operating with negative feedback in first keyword to be sorted.
Can utilize existing data mining technology to obtain the keyword relevant with object information.
In addition, one embodiment of the present of invention provide a kind of technology of obtaining the keyword relevant with object information.Corresponding to above-mentioned positive feedback operation and negative feedback operation, and to the keyword of the corresponding object information of the selection of keyword for selecting; With keyword comprised the corresponding object information of screening for being used for comprising the information of screening; With do not select corresponding object information for there not being selecteed keyword to keyword; With to the corresponding object information of the shielding of keyword be conductively-closed and can't selecteed keyword; With to the corresponding object information of the exclude filter of keyword for being used for carrying out the information of exclude filter.
Extraction unit 341 can based on the relevant feedback word of object information of operation, group word structure, attribute characteristics, label at least one obtain first keyword of being correlated with object information operation as in first keyword of keyword to be sorted.
Specifically, extraction unit 341 can be at first obtains feedback word, group word structure, attribute characteristics, the label at least one from the object information of operation, obtains from keyword to be sorted subsequently and feeds back at least one first relevant keyword in word, group word structure, attribute characteristics, the label.
Here, the feedback word refers to the keyword that obtains of object information from operation.Extraction unit 341 can be by cutting word to obtain the feedback word to the object information of operation.Alternatively, extraction unit 341 also can be selected the feedback word of predetermined quantity according to word frequency (that is, the number of times that occurs) order from high to low in all feedback word candidates that obtain, rather than uses all types of feedback words that obtain.
Alternatively, the word frequency of feedback word is more high, and then the ordering of the keyword relevant with this feedback word is changed more manyly by adjustment unit 342.For example, for the feedback word that the object information from the positive feedback operation obtains, the word frequency of a feedback word is more high, and then the ordering of the keyword relevant with this feedback word is enhanced more manyly by adjustment unit 342; For the feedback word that the object information from the negative feedback operation obtains, the word frequency of a feedback word is more high, and then the ordering of the keyword relevant with this feedback word is lowered more manyly by adjustment unit 342.
Phrase structure refers to the language construct structure (for example, V-O construction, parallel construction, M-D (modifier-head) construction, subject-predicate phrase etc.) of the object information operated.Extraction unit 341 can be by analyzing the group word structure of object information of operation, with the group word structure that obtains to exist.Alternatively, extraction unit 341 also can be according to quantity (namely, the quantity that the phrase structure of each type exists in all phrase structures that obtain) order is from high to low selected the group word structure of the type of predetermined quantity, rather than uses all types of groups of word structures that obtain.
Alternatively, the quantity of one type phrase structure is more many, and then the ordering of the keyword relevant with this kind phrase structure is changed more manyly by adjustment unit 342.For example, for the phrase structure that the object information from the positive feedback operation obtains, the quantity of one type phrase structure is more many, and then the ordering of the keyword relevant with the phrase structure of the type is enhanced more manyly by adjustment unit 342; For the phrase structure that the object information from the negative feedback operation obtains, the quantity of one type phrase structure is more many, and then the ordering of the keyword relevant with the phrase structure of the type is lowered more manyly by adjustment unit 342.
The attribute characteristics refer to the high or low characteristics of the attribute that can quantize (for example, volumes of searches, degree of contention) of the object information operated.For example, it is high being considered to the attribute characteristics greater than the attribute of predetermined threshold; It is low that the attribute that is not more than predetermined threshold is considered to the attribute characteristics.
Adjustment unit 342 is adjusted the keyword relevant with the attribute characteristics.For example, for the attribute characteristics (for example, volumes of searches height or volumes of searches are low) that the object information from the positive feedback operation obtains, the ordering that adjustment unit 342 improves the keyword relevant with these attribute characteristics; For the attribute characteristics (for example, volumes of searches height or volumes of searches are low) that the object information from the negative feedback operation obtains, adjustment unit 342 reduces the ordering of the keyword relevant with these attribute characteristics.
Label refers to the various forms of labels (for example, tag along sort, geographical labels etc.) of the object information operated.Extraction unit 341 can be by analyzing the label of object information of operation, with the label that obtains to exist.Alternatively, extraction unit 341 also can be selected the label of the type of predetermined quantity according to quantity order from high to low, rather than uses all types of labels that obtain.
Alternatively, one type number of tags is more many, and then the ordering of the keyword relevant with this kind label is changed more manyly by adjustment unit 342.For example, for the label that the object information from the positive feedback operation obtains, one type number of tags is more many, and then the ordering of the keyword relevant with the label of the type is enhanced more manyly by adjustment unit 342; For the label that the object information from the negative feedback operation obtains, one type number of tags is more many, and then the ordering of the keyword relevant with the label of the type is lowered more manyly by adjustment unit 342.
In addition, operating unit 330 also can receive the user to the operation again of first keyword, and updating block 340 upgrades first keyword according to the described object information of operation again.
Fig. 4 illustrates the block diagram of the equipment of recommended keywords according to another embodiment of the present invention
As shown in Figure 4, the equipment 400 of recommended keywords according to another embodiment of the present invention comprises: receiving element 410, recommendation unit 420, operating unit 430, updating block 440, selected cell 450.
Receiving element 410 receives the seed data of user's input.
First recommendation unit 421 obtains second keyword based on seed data.Second recommendation unit 422 selects second keyword of predetermined quantity to offer the user as first keyword.
In an example of the present invention, will be divided into positive feedback operation and negative feedback operation to the operation of keyword.Positive feedback operation can for example comprise: to the selection of keyword and/or to the screening that comprises of keyword.
Here, comprise the keyword that screening refers to filter out the screening conditions (key word that for example, is used for screening) that comprise input.
Negative feedback operation can for example comprise: to not the selecting of keyword, to the shielding of keyword, in the exclude filter of keyword at least one.
Here, can't be selected thereby the shielding of keyword is referred to temporarily will be scheduled to the keyword shielding, exclude filter refers to filter out the keyword of the screening conditions (key word that for example, is used for screening) that do not comprise input.
Should be appreciated that positive feedback operation of the present invention and negative feedback operation are not limited to top example.Can embody and to operate as positive feedback any operation of the positive attitude of predetermined keyword, can embody and to operate as negative feedback any operation of the passive attitude of predetermined keyword.
Updating block 440 upgrades the 3rd keyword according to the object information of described operation, and wherein, the 3rd keyword is to select first keyword remaining second keyword afterwards.For example, updating block 440 is adjusted the prioritization of the 3rd keyword according to the object information of described operation.
It is identical with updating block shown in Figure 3 340 that updating block 440 can be implemented as.At this moment, the keyword to be sorted as the operand of updating block 340 becomes the 3rd keyword.
In addition, selected cell 450 can be repeatedly in response to another operation from the user, offer the user with the 3rd keyword of repeatedly that prioritization is forward predetermined quantity as first keyword.Should be appreciated that each selecteed the 3rd keyword will not be re-used as the 3rd keyword and exist.
In addition, operating unit 430 also can receive the user to the operation again of first keyword, and updating block 440 upgrades the 3rd keyword according to the described object information of operation again.
According to the method and apparatus of recommended keywords of the present invention, can come recommended keywords in real time to the operation that keyword is provided based on seed data according to the user.In addition, when seed data more after a little while, accurate suitable keyword can be offered the user.
In addition, should be appreciated that, can be implemented nextport hardware component NextPort according to each unit in the equipment of the recommended keywords of exemplary embodiment of the present invention.Those skilled in the art can for example use field programmable gate array (FPGA) or special IC (ASIC) to realize each unit according to the performed processing in each unit that limits.
In addition, said method according to the present invention may be implemented as the computer code in the computer readable recording medium storing program for performing.Those skilled in the art can realize described computer code according to the description to said method.When being performed, realizes described computer code said method of the present invention in computing machine.
Although specifically shown with reference to its exemplary embodiment and described the present invention, but it should be appreciated by those skilled in the art, under the situation that does not break away from the spirit and scope of the present invention that claim limits, can carry out various changes on form and the details to it.
Claims (42)
1. the method for a recommended keywords comprises:
Receive seed data from the user;
Provide at least one first keyword to the user based on seed data;
Receive the user to the operation of first keyword;
Object information according to described operation upgrades first keyword.
2. method according to claim 1 wherein, provides at least one first keyword to comprise to user's step based on seed data:
Obtain second keyword based on seed data;
Select second keyword of predetermined quantity to offer the user as first keyword.
3. method according to claim 2, wherein, the step of upgrading first keyword according to the object information of described operation comprises:
Adjust the prioritization of the 3rd keyword according to the object information of described operation, wherein, the 3rd keyword is to select remaining second keyword after first keyword;
In response to another operation from the user, select the 3rd keyword of the forward predetermined quantity of prioritization to offer the user as first keyword.
4. method according to claim 1, wherein, the step of upgrading first keyword according to the object information of described operation comprises:
Adjust the prioritization of first keyword according to the object information of described operation.
5. method according to claim 1, wherein, the user comprises positive feedback operation and negative feedback operation to the operation of described at least one first keyword.
6. method according to claim 5, wherein, the positive feedback operation comprises: to the selection of predetermined first keyword and/or to the screening that comprises of first keyword.
7. method according to claim 5, wherein, the negative feedback operation comprises: to not the selecting of predetermined first keyword, to the shielding of predetermined first keyword, in the exclude filter of first keyword at least one.
8. method according to claim 6, wherein, with the corresponding object information of selection to predetermined first keyword be described predetermined first keyword, and first keyword comprised the corresponding object information of screening for being used for comprising the information of screening.
9. method according to claim 7, wherein, with do not select corresponding object information for there not being selecteed first keyword to predetermined first keyword, with the corresponding object information of shielding to predetermined first keyword be conductively-closed and can't selecteed first keyword, with to the corresponding object information of the exclude filter of first keyword for being used for carrying out the information of exclude filter.
10. according to claim 3 or 4 described methods, wherein, come according to first keyword of the object information adjustment conduct of described operation keyword to be sorted or the prioritization of the 3rd keyword by following manner: obtain the relevant keyword of the object information with operation in the keyword to be sorted; The prioritization of the keyword that adjustment is obtained in keyword to be sorted.
11. method according to claim 10, wherein, the step of the prioritization of the keyword that adjustment is obtained in keyword to be sorted comprises: improve the prioritization of the keyword relevant with object information positive feedback operation in the keyword to be sorted, reduce the prioritization of the relevant keyword of the object information of operating with negative feedback in the keyword to be sorted.
12. method according to claim 10 wherein, is obtained the step with the relevant keyword of object information operation in the keyword to be sorted and is comprised:
Obtain the feedback word, organize at least one word structure, attribute characteristics, the label from the object information of operation;
From keyword to be sorted, obtain and feed back at least one relevant keyword in word, group word structure, attribute characteristics, the label.
13. method according to claim 12, wherein, the step of feeding back word from the object information acquisition of operation comprises: the object information of operation is cut word to obtain the feedback word;
Wherein, comprise from the step of object information acquisition group word structure of operation: the group word structure to the object information of operation is analyzed, with the group word structure that obtains to exist;
Wherein, the step that obtains label from the object information of operation comprises: the label to the object information of operation is analyzed, with the label that obtains to exist;
Wherein, the step that obtains the attribute characteristics from the object information of operating comprises: the size of the attribute of the object information of analysis operation, and to determine the attribute characteristics.
14. method according to claim 12, wherein, the attribute characteristics comprise the high or low of volumes of searches.
15. method according to claim 13, wherein, in the step of the prioritization of keyword in keyword to be sorted that adjustment is obtained: the word frequency of feedback word is more high, and then the prioritization of the keyword relevant with this feedback word is changed more manyly by adjustment.
16. method according to claim 13, wherein, in the step of the prioritization of keyword in keyword to be sorted that adjustment is obtained: the quantity of the group word structure of a type is more many, and then the prioritization of the keyword relevant with the group word structure of the type is changed more manyly by adjustment.
17. method according to claim 13, wherein, in the step of the prioritization of keyword in keyword to be sorted that adjustment is obtained: the number of tags of a type is more high, and then the prioritization of the keyword relevant with the label of the type is changed more manyly by adjustment.
18. method according to claim 13, wherein, the step of feeding back word from the object information acquisition of operation also comprises: the feedback word of selecting predetermined quantity according to word frequency order from high to low from the feedback word that obtains.
19. method according to claim 13 wherein, also comprises from the step of object information acquisition group word structure of operation: the group word structure of selecting the type of predetermined quantity according to the quantity order from high to low of the phrase structure of each type.
20. method according to claim 13, wherein, the step that obtains label from the object information of operating also comprises: the label of selecting the type of predetermined quantity according to the number of tags order from high to low of each type.
21. method according to claim 1 also comprises:
Receive the user to the operation again of first keyword;
Upgrade first keyword according to the described object information of operation again.
22. the equipment of a recommended keywords comprises:
Receiving element receives seed data from the user;
Recommendation unit provides at least one first keyword to the user based on seed data;
Operating unit receives the user to the operation of first keyword;
Updating block upgrades first keyword according to the object information of described operation.
23. equipment according to claim 22, wherein, recommendation unit comprises:
First recommendation unit obtains second keyword based on seed data;
Second recommendation unit selects second keyword of predetermined quantity to offer the user as first keyword.
24. equipment according to claim 23, wherein, updating block is adjusted the prioritization of the 3rd keyword according to the object information of described operation, and wherein, the 3rd keyword is to select remaining second keyword after first keyword,
Wherein, described equipment also comprises selected cell, in response to another operation from the user, selects the 3rd keyword of the forward predetermined quantity of prioritization to offer the user as first keyword.
25. equipment according to claim 22, wherein, updating block is adjusted the prioritization of first keyword according to the object information of described operation.
26. equipment according to claim 22, wherein, the user comprises positive feedback operation and negative feedback operation to the operation of described at least one first keyword.
27. equipment according to claim 26, wherein, positive feedback operation comprises: to the selection of predetermined first keyword and/or to the screening that comprises of first keyword.
28. equipment according to claim 26, wherein, negative feedback operation comprises: to not the selecting of predetermined first keyword, to the shielding of predetermined first keyword, in the exclude filter of first keyword at least one.
29. equipment according to claim 27, wherein, with the corresponding object information of selection to predetermined first keyword be described predetermined first keyword, and first keyword comprised the corresponding object information of screening for being used for comprising the information of screening.
30. equipment according to claim 28, wherein, with do not select corresponding object information for there not being selecteed first keyword to predetermined first keyword, with the corresponding object information of shielding to predetermined first keyword be conductively-closed and can't selecteed first keyword, with to the corresponding object information of the exclude filter of first keyword for being used for carrying out the information of exclude filter.
31. according to claim 24 or 25 described equipment, wherein, updating block comprises:
Extraction unit obtains the relevant keyword of the object information with operation in the keyword to be sorted;
Adjustment unit is adjusted the prioritization of keyword in keyword to be sorted of obtaining.
32. equipment according to claim 31, wherein, adjustment unit improves the prioritization of the relevant keyword of the object information with the positive feedback operation in the keyword to be sorted, reduces the prioritization of the relevant keyword of the object information of operating with negative feedback in the keyword to be sorted.
33. equipment according to claim 31, wherein, extraction unit obtains feedback word, group word structure, attribute characteristics, the label at least one from the object information of operation, and obtains from keyword to be sorted and feed back at least one relevant keyword in word, group word structure, attribute characteristics, the label.
34. equipment according to claim 33, wherein, extraction unit is cut word with acquisition feedback word to the object information of operation,
Wherein, extraction unit is analyzed the group word structure of the object information of operation, with the group word structure that obtains to exist,
Wherein, extraction unit is analyzed the label of the object information of operation, with the label that obtains to exist,
Wherein, the size of the attribute of the object information of extraction unit analysis operation is to determine the attribute characteristics.
35. equipment according to claim 33, wherein, the attribute characteristics comprise the high or low of volumes of searches.
36. equipment according to claim 34, wherein, the word frequency of feedback word is more high, and then the prioritization of the keyword relevant with this feedback word is changed more manyly by adjustment unit.
37. equipment according to claim 34, wherein, the quantity of the group word structure of a type is more high, and then the prioritization of the keyword relevant with the group word structure of the type is changed more manyly by adjustment unit.
38. equipment according to claim 34, wherein, the number of tags of a type is more many, and then the prioritization of the keyword relevant with the label of the type is changed more manyly by adjustment unit.
39. equipment according to claim 34, wherein, extraction unit is selected the feedback word of predetermined quantity from the feedback word that obtains according to word frequency order from high to low.
40. equipment according to claim 34, wherein, extraction unit is selected the group word structure of the type of predetermined quantity according to the quantity order from high to low of the phrase structure of each type.
41. equipment according to claim 34, wherein, extraction unit is selected the label of the type of predetermined quantity according to the number of tags order from high to low of each type.
42. equipment according to claim 22, wherein, operating unit receives the user to the operation again of first keyword, and updating block upgrades first keyword according to the described object information of operation again.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310269643.1A CN103324742B (en) | 2013-06-28 | 2013-06-28 | The method and apparatus of recommended keywords |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310269643.1A CN103324742B (en) | 2013-06-28 | 2013-06-28 | The method and apparatus of recommended keywords |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103324742A true CN103324742A (en) | 2013-09-25 |
| CN103324742B CN103324742B (en) | 2017-03-29 |
Family
ID=49193485
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201310269643.1A Active CN103324742B (en) | 2013-06-28 | 2013-06-28 | The method and apparatus of recommended keywords |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103324742B (en) |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103888466A (en) * | 2014-03-28 | 2014-06-25 | 北京搜狗科技发展有限公司 | User interest discovering method and device |
| CN104361063A (en) * | 2014-11-04 | 2015-02-18 | 北京字节跳动网络技术有限公司 | User interest discovering method and device |
| CN105468901A (en) * | 2014-11-20 | 2016-04-06 | 韩国韩医学研究院 | Apparatus and method for setting multi-hierarchy event alarm according user response |
| WO2016134580A1 (en) * | 2015-02-28 | 2016-09-01 | 华为技术有限公司 | Data query method and apparatus |
| CN108376339A (en) * | 2017-09-11 | 2018-08-07 | 加拿大辉莱广告公司 | Advertising specialty recommends method and device |
| CN110362737A (en) * | 2018-04-08 | 2019-10-22 | 优视科技有限公司 | Method for pushing, device and the server of recommendation |
| CN110598016A (en) * | 2019-09-11 | 2019-12-20 | 腾讯科技（深圳）有限公司 | Method, device, equipment and medium for recommending multimedia information |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101650731A (en) * | 2009-08-31 | 2010-02-17 | 浙江大学 | Method for generating suggested keywords of sponsored search advertisement based on user feedback |
| CN102387207A (en) * | 2011-10-21 | 2012-03-21 | 华为技术有限公司 | Push method and system based on user feedback information |
| CN102982042A (en) * | 2011-09-07 | 2013-03-20 | 中国移动通信集团公司 | Personalization content recommendation method and platform and system |
-
2013
- 2013-06-28 CN CN201310269643.1A patent/CN103324742B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101650731A (en) * | 2009-08-31 | 2010-02-17 | 浙江大学 | Method for generating suggested keywords of sponsored search advertisement based on user feedback |
| CN102982042A (en) * | 2011-09-07 | 2013-03-20 | 中国移动通信集团公司 | Personalization content recommendation method and platform and system |
| CN102387207A (en) * | 2011-10-21 | 2012-03-21 | 华为技术有限公司 | Push method and system based on user feedback information |
Non-Patent Citations (2)
| Title |
|---|
| YIFAN CHEN ET AL.: ""Advertising Keyword Suggestion Based on Concept Hierarchy"", 《WSDM’08 PROCEEDINGS OF THE INTERNATIONAL CONFERENCE ON WEB SEARCH AND DATA MINING》, 31 December 2008 (2008-12-31), pages 251 - 260 * |
| 张铭芮等: ""移动互联网广告推荐算法研究"", 《电信科学》 * |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103888466A (en) * | 2014-03-28 | 2014-06-25 | 北京搜狗科技发展有限公司 | User interest discovering method and device |
| CN103888466B (en) * | 2014-03-28 | 2018-01-02 | 北京搜狗科技发展有限公司 | user interest discovery method and device |
| CN104361063A (en) * | 2014-11-04 | 2015-02-18 | 北京字节跳动网络技术有限公司 | User interest discovering method and device |
| CN104361063B (en) * | 2014-11-04 | 2018-03-16 | 北京字节跳动网络技术有限公司 | user interest discovery method and device |
| CN105468901A (en) * | 2014-11-20 | 2016-04-06 | 韩国韩医学研究院 | Apparatus and method for setting multi-hierarchy event alarm according user response |
| WO2016134580A1 (en) * | 2015-02-28 | 2016-09-01 | 华为技术有限公司 | Data query method and apparatus |
| CN108376339A (en) * | 2017-09-11 | 2018-08-07 | 加拿大辉莱广告公司 | Advertising specialty recommends method and device |
| CN110362737A (en) * | 2018-04-08 | 2019-10-22 | 优视科技有限公司 | Method for pushing, device and the server of recommendation |
| CN110362737B (en) * | 2018-04-08 | 2022-02-01 | 阿里巴巴（中国）有限公司 | Recommended content pushing method and device and server |
| CN110598016A (en) * | 2019-09-11 | 2019-12-20 | 腾讯科技（深圳）有限公司 | Method, device, equipment and medium for recommending multimedia information |
Also Published As
| Publication number | Publication date |
|---|---|
| CN103324742B (en) | 2017-03-29 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11907244B2 (en) | Modifying field definitions to include post-processing instructions | |
| US10268758B2 (en) | Method and system of acquiring semantic information, keyword expansion and keyword search thereof | |
| CN103324742A (en) | Method and equipment for recommending keywords | |
| US20210318851A1 (en) | Systems and Methods for Dataset Merging using Flow Structures | |
| US20200057958A1 (en) | Identification and application of hyperparameters for machine learning | |
| US10942733B2 (en) | Open-source-license analyzing method and apparatus | |
| WO2021098648A1 (en) | Text recommendation method, apparatus and device, and medium | |
| US10713291B2 (en) | Electronic document generation using data from disparate sources | |
| CN104933100A (en) | Keyword recommendation method and device | |
| CN109871311B (en) | Method and device for recommending test cases | |
| CN102930054A (en) | Data search method and data search system | |
| CN112966081B (en) | Method, device, equipment and storage medium for processing question and answer information | |
| CN112651236B (en) | Method and device for extracting text information, computer equipment and storage medium | |
| CN107291745B (en) | Data index management method and device | |
| CN106503274A (en) | A kind of Data Integration and searching method and server | |
| CN103136213A (en) | Method and device for providing related words | |
| CN110941702A (en) | Retrieval method and device for laws and regulations and laws and readable storage medium | |
| CN103870553A (en) | Input resource pushing method and system | |
| CN114330329A (en) | Service content searching method and device, electronic equipment and storage medium | |
| EP3079083A1 (en) | Providing app store search results | |
| CN106919593B (en) | Searching method and device | |
| US20180322199A1 (en) | Optimizing the allocation of jobs on job portal | |
| CN110362694A (en) | Data in literature search method, equipment and readable storage medium storing program for executing based on artificial intelligence | |
| CN104156492A (en) | Method and device for prompting search content | |
| US20190163810A1 (en) | Search User Interface |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |