CN108345419B - Information recommendation list generation method and device - Google Patents
Information recommendation list generation method and device Download PDFInfo
- Publication number
- CN108345419B CN108345419B CN201710056067.0A CN201710056067A CN108345419B CN 108345419 B CN108345419 B CN 108345419B CN 201710056067 A CN201710056067 A CN 201710056067A CN 108345419 B CN108345419 B CN 108345419B
- Authority
- CN
- China
- Prior art keywords
- recommended
- recommendation
- objects
- feature data
- preset
- 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.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9535—Search customisation based on user profiles and personalisation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
- G06F3/01—Input arrangements or combined input and output arrangements for interaction between user and computer
- G06F3/048—Interaction techniques based on graphical user interfaces [GUI]
- G06F3/0481—Interaction techniques based on graphical user interfaces [GUI] based on specific properties of the displayed interaction object or a metaphor-based environment, e.g. interaction with desktop elements like windows or icons, or assisted by a cursor's changing behaviour or appearance
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- General Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Human Computer Interaction (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The embodiment of the application discloses a method and a device for generating an information recommendation list, wherein the method comprises the following steps: the method comprises the steps that a terminal obtains characteristic data of each object to be recommended in an object set to be recommended, wherein the object set to be recommended comprises S1 objects to be recommended; the method comprises the steps that a terminal obtains a preset feature data set, the preset feature data set comprises feature data of S2 recommended objects, the feature data of S2 recommended objects comprise feature data of specified recommended objects, and S2 is smaller than or equal to S1; the terminal calculates the recommendation value of each object to be recommended according to the preset feature data set and feature data of S1 objects to be recommended; and the terminal selects a target recommendation object according to the recommendation value of each object to be recommended in the S1 objects to be recommended, and adds the target recommendation object to the designated display position in the recommendation list. By adopting the embodiment of the application, the method and the device have the advantages that the selection accuracy of the recommended objects can be improved, and the resource utilization rate of the recommendation list can be improved.
Description
Technical Field
The present application relates to the field of communications technologies, and in particular, to a method and an apparatus for generating an information recommendation list.
Background
Currently, with the increasing popularization of mobile phones, the mobile phones have increasingly close relationship with the daily life requirements of mobile phone users. The mobile phone user can contact with relatives and friends, read books, view information, plan out-trips, or play games, and the like through the mobile phone, and the mobile phone user can acquire corresponding information in a mode of downloading a corresponding application program (APP for short) through a mobile phone application market. The mobile phone recommendation system can recommend the APP to be downloaded to the mobile phone user on an information recommendation platform such as a mobile phone application market.
In the prior art, a mobile phone recommendation system can recommend more related information such as APP to a user according to the APP downloaded by the mobile phone user. In the prior art, by predicting the preference degree of a mobile phone user for certain articles, an APP with higher predicted preference degree is recommended to the mobile phone user. However, the recommendation method in the prior art only considers the information such as the APP which the user is interested in, does not consider the substitutability and other correlations between the existing information such as the APP and the recommended information such as the APP, and is poor in applicability. For example, if a mobile phone user downloads an APP of a travel category, the APP with higher user preference predicted by the prior art will include more APPs of travel categories, and thus there will be more alternative APPs of the same category in the recommendation list. When the tourism APP that the user needs only need one, the APP of other relevant tourism types will have taken more recommendation display position, have wasted recommendation list resource, and the suitability is poor.
Disclosure of Invention
The application provides a method and a device for generating an information recommendation list, which can improve the correlation between the recommendation value of an object to be recommended and the characteristic data of the object to be recommended, and improve the selection accuracy of the recommended object and the resource utilization rate of the recommendation list.
In a first aspect, the present application provides a method for generating an information recommendation list, which may include:
the method comprises the steps that a terminal obtains feature data of each object to be recommended in an object set to be recommended, wherein the object set to be recommended comprises S1 objects to be recommended, and S1 is an integer larger than 1;
the terminal acquires a preset feature data set, the preset feature data set comprises feature data of S2 recommended objects, the feature data of S2 recommended objects comprise feature data of specified recommended objects, and S2 is smaller than or equal to S1;
the terminal calculates the recommendation value of each object to be recommended according to the preset feature data set and the feature data of the S1 objects to be recommended;
and the terminal selects a target recommendation object according to the recommendation value of each object to be recommended in the S1 objects to be recommended, and adds the target recommendation object to a designated display position in a recommendation list.
In the implementation manner provided by the application, the recommendation value of each object to be recommended can be calculated according to the feature data of each object to be recommended and the preset feature data set. When the recommendation value of each object to be recommended is calculated, the feature data of the specified recommendation object can be added into the calculation of the recommendation value of the object to be recommended to be determined, the determination of the specified recommendation object can be determined according to the actual application scene, the operation is more flexible, the calculation accuracy of the recommendation value of the object to be recommended is higher, the algorithm complexity controllability for generating the recommendation list is improved, and the applicability is stronger. Furthermore, the implementation manner described in the application determines the objects to be recommended to be determined according to the feature data of the specified recommended objects, so that the correlation among the objects to be recommended can be controlled more flexibly, and the utilization rate of the display positions of the recommendation list is improved.
With reference to the first aspect, in a first possible implementation manner, the specified recommendation object includes a recommendation object that has been determined to be added to the recommendation list, and a feature similarity between the specified recommendation object and the target recommendation object is smaller than a preset similarity threshold.
According to the method and the device, the designated recommendation object can be flexibly selected, the feature data of the designated recommendation object is added into the calculation of the recommendation value of the object to be recommended to be determined, and the correlation among the objects to be recommended added to the recommendation list can be flexibly controlled. The specified recommendation objects can comprise the recommendation objects which are determined to be added to the recommendation list, the occurrence probability of replaceable objects to be recommended can be reduced, the effective utilization rate of the display positions of the recommendation list is further improved, and the resource utilization rate is improved.
With reference to the first aspect or the first possible implementation manner of the first aspect, in a second possible implementation manner, the calculating, by the terminal, the recommendation value of the ith object to be recommended according to the preset feature data set and the feature data of the S1 objects to be recommended includes:
and the terminal takes the preset characteristic data set and the characteristic data of the ith object to be recommended as input values of a preset recommended value calculation model, and calculates the recommended value of the ith object to be recommended through the recommended value calculation model.
According to the method and the device, the recommendation value of each object to be recommended can be calculated through the recommendation value calculation model, calculation accuracy and calculation efficiency of the recommendation value of the object to be recommended can be improved, and the applicability is higher.
With reference to any one of the first aspect to the second possible implementation manner of the first aspect, in a third possible implementation manner, the selecting, by the terminal, a target recommendation object according to the recommendation value of each object to be recommended in the S1 objects to be recommended includes:
and the terminal selects the objects to be recommended which accord with the predefined selection rule from the S1 objects to be recommended, and selects the object to be recommended with the maximum recommendation value from the objects to be recommended which accord with the predefined selection rule as the target recommendation object.
According to the method and the device, the target recommendation object can be selected according to the predefined selection rule and the recommendation value of each object to be recommended, so that the selection controllability of each recommendation object in the recommendation list is further improved, and the resource utilization rate of the recommendation list is improved.
With reference to any one of the first aspect to the third possible implementation manner of the first aspect, in a fourth possible implementation manner, after the target recommendation object is added to a specified display position in a recommendation list, the method further includes:
and removing the characteristic data of the target recommendation object from the set of objects to be recommended, and adding the characteristic data of the target recommendation object to the preset characteristic data set.
According to the method and the device, the feature data in the preset feature data set can be updated according to the selected target recommendation object, the selection controllability of each recommendation object in the recommendation list is improved, the occurrence probability of the replaceable object to be recommended in the recommendation list is reduced, and the resource utilization rate of the recommendation list is improved.
With reference to any one of the first aspect to the fourth possible implementation manner of the first aspect, in a fifth possible implementation manner, the object to be recommended includes: at least one of an application program APP, audio/video data, a webpage and news information.
The implementation mode provided by the application can be suitable for selecting the objects to be recommended in more expression forms, the diversity of the information recommendation list is improved, and the applicability of the generation method of the information recommendation list is enhanced.
With reference to any one of the first aspect to the fifth possible implementation manner of the first aspect, in a sixth possible implementation manner, the feature data includes: at least one of an identification ID, a category attribute, an applicable platform, a number of times of reference, a click through rate, a number of downloads, and a data size.
In the implementation manner provided by the application, the feature data of the object to be recommended and the feature data in the preset feature data set can include multiple types of data, so that the selection accuracy of the object to be recommended can be improved, and the effectiveness of the recommendation list can be enhanced.
With reference to any one of the first aspect to the third possible implementation manner of the first aspect, in a seventh possible implementation manner, the predefined rule includes: the number of the objects to be recommended with the same or similar characteristics is not more than M1, or the version update date of the objects to be recommended is not later than the predefined date.
According to the method and the device, the target recommendation object can be selected according to the predefined selection rule and the recommendation value of each object to be recommended, the selection effectiveness of each recommendation object in the recommendation list is further improved, and the resource utilization rate of the recommendation list is improved.
In a second aspect, the present application provides an apparatus for generating an information recommendation list, which may include:
the recommendation system comprises an acquisition module, a recommendation module and a recommendation module, wherein the acquisition module is used for acquiring feature data of each object to be recommended in an object set to be recommended, the object set to be recommended comprises S1 objects to be recommended, and S1 is an integer greater than 1;
the obtaining module is further configured to obtain a preset feature data set, where the preset feature data set includes feature data of S2 recommended objects, the feature data of S2 recommended objects includes feature data of a specified recommended object, and S2 is less than or equal to S1;
the calculation module is used for calculating recommendation values of the objects to be recommended according to the preset feature data set acquired by the acquisition module and the feature data of the S1 objects to be recommended;
and the selecting module is used for selecting a target recommendation object according to the recommendation value of each object to be recommended in the S1 objects to be recommended calculated by the calculating module, and adding the target recommendation object to a specified display position in a recommendation list.
With reference to the second aspect, in a first possible implementation manner, the specified recommended object obtained by the obtaining module includes a recommended object that has already been determined to be added to the recommendation list, and a feature similarity between the specified recommended object and the target recommended object is smaller than a preset similarity threshold.
With reference to the second aspect or the first possible implementation manner of the second aspect, in a second possible implementation manner, the calculation module is configured to:
and taking the preset feature data set and the feature data of the ith object to be recommended as input values of a preset recommendation value calculation model, and calculating the recommendation value of the ith object to be recommended through the recommendation value calculation model.
With reference to any one of the second aspect to the second possible implementation manner of the second aspect, in a third possible implementation manner, the selecting module is configured to:
and selecting the objects to be recommended which accord with a predefined selection rule from the S1 objects to be recommended which are obtained by the obtaining module, and selecting the object to be recommended with the maximum recommendation value calculated by the calculating module as a target recommendation object from the objects to be recommended which accord with the predefined selection rule.
With reference to any one of the second aspect to any one of the third possible implementation manners of the second aspect, in a fourth possible implementation manner, the obtaining module is further configured to:
and removing the characteristic data of the target recommendation object from the set of objects to be recommended, and adding the characteristic data of the target recommendation object to the preset characteristic data set.
With reference to any one of the second aspect to the fourth possible implementation manner of the second aspect, in a fifth possible implementation manner, the object to be recommended includes: at least one of an application program APP, audio/video data, a webpage and news information.
With reference to any one of the second aspect to the fifth possible implementation manner of the second aspect, in a sixth possible implementation manner, the feature data includes: at least one of an identification ID, a category attribute, an applicable platform, a number of times of reference, a click through rate, a number of downloads, and a data size.
In a third aspect, the present application provides a terminal device, which may include: a memory and a processor;
the memory is used for storing a group of program codes;
the processor is configured to call the program code stored in the memory to execute the method provided by the first aspect.
In the implementation manner provided by the application, the recommendation value of each object to be recommended can be calculated according to the feature data of each object to be recommended and the preset feature data set, wherein the feature data of the object to be recommended which is determined to be recommended is also added to the calculation of the recommendation value of the next object to be recommended, the calculation accuracy of the recommendation value is higher, the algorithm complexity controllability for generating the recommendation list is improved, and the applicability is stronger. Furthermore, the implementation manner described in the application determines the object to be recommended to be determined according to the determined feature data of the object to be recommended, so that the correlation among the objects to be recommended can be more flexibly controlled, the occurrence probability of the replaceable object to be recommended is reduced, and the utilization rate of the display position of the recommendation list is improved.
Drawings
In order to more clearly describe the technical solutions in the embodiments of the present application, the drawings required to be used in the embodiments of the present application will be described below.
Fig. 1 is a schematic flow chart of a method for generating an information recommendation list according to an embodiment of the present application;
fig. 2 is another schematic flow chart of a method for generating an information recommendation list according to an embodiment of the present application;
fig. 3 is a schematic structural diagram of an information recommendation list generation apparatus provided by an embodiment of the present application;
fig. 4 is another schematic structural diagram of an information recommendation list generation apparatus provided in an embodiment of the present application.
Detailed Description
Embodiments of the present application are described below with reference to the drawings in the embodiments of the present application.
In a specific implementation, a terminal described in an embodiment of the present application includes: the mobile phone (mobile phone), the tablet computer (Pad), the wearable device, the personal computer assistant, and the like may be determined according to an actual application scenario, and are not limited herein. The following description will be given taking a mobile phone as an example.
The application platforms to which the method and the device for generating the information recommendation list are applicable include, but are not limited to, a mobile phone application market, mobile phone music, mobile phone video, mobile phone reading, mobile phone news information, mobile phone web pages and the like. For example, taking an application market as an example, the method and the apparatus for generating an information recommendation list provided in the embodiments of the present application may be applied to the huaji application market, the Baidu application market, the 360 application market, the millet application market, and the like. The information recommendation list described in the embodiments of the present application may include, but is not limited to: the system comprises an APP recommendation list, an audio and video data recommendation list, a webpage recommendation list, a news information recommendation list and the like. That is, the objects to be recommended described in the present application may include, but are not limited to: APP, audio/video data, web pages, news information, etc., which will be described below by taking APP as an example.
In the mobile phone recommendation system, the generation of the information recommendation list comprises the processes of prediction, recommendation and the like. What needs to be solved by the prediction is to predict the preference degree of the mobile phone user for each recommended object. The recommendation is to sort the recommendation objects according to the prediction result, for example, according to the predicted preference degree, sorting according to the order of preference degrees from high to low. The ranking strategies proposed in the field of ranking to rank (learning to rank) include a point-wise (single point optimization) strategy, a list-wise (list optimization) strategy and the like. The point-wise policy is to sort the objects to be recommended in descending order according to their predicted recommendation values, such as click-through-rate (CTR). In the point-wise strategy, the ranking of the objects to be recommended is performed from high preference degree to low preference degree according to the preference degree of the user of the objects to be recommended, the relevance such as replaceability among the objects to be recommended is not considered, and the applicability is poor.
The list-wise sorting strategy is to directly sort all the recommended objects to obtain an overall sequence, further directly optimize the overall sequence to obtain an optimized sequence by taking the overall sequence as a sample, and sort all the recommended objects according to the optimized sequence to obtain a recommendation list. The difficulty of the ranking strategy is how to label the recommendation list, and the probability of all ranking combinations needs to be calculated, which is difficult to implement. If there are n recommendation objects, the time complexity required by the sort strategy will be up to O (n × n |), where O () is an expression of time complexity. The larger n is, the higher the time complexity required by the sequencing sequence is, which cannot be directly applied to solving the problems in the industry, and the applicability is poor.
The embodiment of the application provides a method and a device for generating an information recommendation list, which can sequentially select a target recommendation object to be added to a designated display position in the recommendation list according to the characteristics of each object to be recommended and the characteristics of the sequenced recommendation objects. The implementation method provided by the embodiment of the application can consider the relevance such as replaceability among the recommendation objects, avoid the display positions of the recommendation list occupied by too many recommendation objects with higher replaceability, improve the resource utilization rate of the recommendation list, and improve the generation applicability of the information recommendation list.
Fig. 1 is a schematic flow chart of a method for generating an information recommendation list according to an embodiment of the present application. The method provided by the embodiment of the application comprises the following steps:
s101, the mobile phone obtains characteristic data of each object to be recommended in the object set to be recommended.
In some possible embodiments, the mobile phone may first obtain feature data of an object to be recommended from a platform such as an application market, and obtain an object set F1 to be recommended according to the obtained feature data of all objects to be recommended. The number of objects to be recommended included in F1 may be set to S1, and S1 is an integer greater than 1. When the mobile phone generates an APP recommendation list (hereinafter referred to as a recommendation list), the recommendation value of each object to be recommended can be determined according to the characteristic data of each object to be recommended, and then a target recommendation object can be selected according to the recommendation value of each object to be recommended and output to the designated display position of the recommendation list.
In some possible embodiments, before the mobile phone generates an APP recommendation list (hereinafter referred to as a recommendation list), the recommendation list may be initialized, and the recommendation list (list) may be initialized to be empty. Further, when the mobile phone performs an initialization operation of the recommendation list, an initial feature set F2 required for calculating recommendation values of the objects to be recommended may be set, and an ordered object set F3 may also be set. Wherein, the F2 includes initial feature data used for calculating recommendation values of the objects to be recommended, and the initial feature data includes: the feature data and the user feature data of each object to be recommended in the S1 objects to be recommended, and the combined features of the feature data and the user feature data of each object to be recommended may be determined specifically according to an actual application scenario or an application platform, and are not limited herein. The feature data of each object to be recommended may include, but is not limited to: identity (ID), product attribute, category attribute, applicable platform, number of times of reference, click through rate, number of downloads, and data size, etc., without limitation to this application. The user characteristic data may include, but is not limited to: user ID, APP selected by the user, historical download data or historical browsing data, and the like, which is not limited in this application.
Further, when the initialization is completed, if F3 is empty, after the mobile phone processes to obtain a recommendation object and sorts the recommendation object to the designated display position of the recommendation list, the sorted recommendation object may be added to F3, and further, whether a new target recommendation object needs to be selected may be determined according to the number of objects to be recommended (i.e., sorted objects) included in F3. If the number of the sorted objects included in F3 is greater than or equal to the predefined number, no new target recommended object is selected any more, and a recommendation list is generated from the sorted objects.
S102, the mobile phone obtains a preset feature data set.
In some possible embodiments, when the mobile phone determines the first target recommendation object in the recommendation list, the obtained preset feature data set may be the initial feature set F2 obtained by the above initialization. That is, in this application scenario, the preset feature data set includes the same feature data as the feature data included in the initial feature set F2 described above.
Further, in some possible embodiments, when the mobile phone determines other target recommended objects in the recommendation list, the feature data of the acquired preset feature data set may include the feature data included in F2 described above, and the feature data of one or more (S2) recommended objects that have already been determined. Wherein S2 is greater than 1 and less than or equal to S1.
And S103, calculating the recommendation value of each object to be recommended by the mobile phone according to the preset feature data set and the feature data of the S1 objects to be recommended.
In some possible embodiments, when the mobile phone calculates the recommendation value of any one of the objects to be recommended (set as the object 1 to be recommended) corresponding to F1, the recommendation value of the object 1 to be recommended may be calculated by using the feature data included in the acquired preset feature data set and the feature data of the object 1 to be recommended. In specific implementation, the mobile phone may use the feature data of the object 1 to be recommended and the feature data included in the preset feature data set as input values of a preset recommendation value calculation model, and calculate the recommendation value of the object 1 to be recommended through the recommendation value calculation model. Similarly, the mobile phone may calculate the recommendation value of the other object to be recommended included in F1.
It should be noted that the recommendation calculation model described in the present application may include, but is not limited to: logistic regression (logistic regression) models, decision tree models, deep learning models, factorization models, domain-aware factorization models, and combination models of any of the above. The method can be determined according to the requirements of the actual application scene, and is not limited herein. The recommendation value calculation model is used for determining a recommendation score (i.e. recommendation value) of an object to be recommended. The recommendation value calculation model can be obtained by training the generation data of the historical recommendation list through a machine learning algorithm. The generated data of the history recommendation list may include, but is not limited to, a label (label), the history recommendation list, and feature data of a recommendation object included in the history recommendation list. The label is a user operation behavior of the object to be recommended, such as purchasing or not purchasing, being clicked or not being clicked, and the like.
In specific implementation, the mobile phone may combine the feature data of the object 1 to be recommended with the feature data in the preset feature data set to obtain a combined feature, and then may calculate the recommendation value of the object 1 to be recommended according to the combined feature. The combination of the feature data may include a cartesian product feature combination. It should be noted that the preset feature data set may include feature data of the object to be recommended (i.e., the specified recommended object, which is set as the object to be recommended 2) whose recommendation has been determined. When the mobile phone calculates the recommendation value of the object to be recommended 1, the feature data of the object to be recommended 2 and the feature data of the object to be recommended 1 and F1 may be combined, and the recommendation value of the object to be recommended 1 may be calculated according to the combined feature data.
In some possible embodiments, after the mobile phone calculates the feature data of each object to be recommended through the recommendation value calculation model, a recommendation value adjustment rule of the object to be recommended may be predefined. The recommended value adjustment rule may be: if the feature similarity between the object 1 to be recommended and the determined recommended objects (namely, the sorted recommended objects) is greater than or equal to a preset similarity threshold, adjusting the recommendation value of the object 1 to be recommended downward; and if the feature similarity of the object 1 to be recommended and the sorted recommended objects is smaller than a preset similarity threshold, the recommended value of the object 1 to be recommended is adjusted upwards. For example, if the feature similarity between the object to be recommended 1 and the object to be recommended 2 is greater than or equal to the preset similarity threshold, it may be determined that the object to be recommended 1 and the object to be recommended 2 are similar objects, and the recommendation value of the object to be recommended 1 calculated at this time may be adjusted to a smaller value. If the feature similarity of the object to be recommended 1 and the object to be recommended 2 is smaller than the preset similarity threshold, it can be determined that the object to be recommended 1 and the object to be recommended 2 are not similar to each other, and the recommendation value of the object to be recommended 1 calculated at this time can be adjusted to be a large value. Wherein, the feature similarity may include: category similarity, information type similarity, or applicable platform similarity, etc.
And S104, selecting a target recommendation object by the mobile phone according to the recommendation value of each object to be recommended in the S1 objects to be recommended, and adding the target recommendation object to a designated display position in a recommendation list.
In some possible embodiments, after the mobile phone calculates the recommendation values of the objects to be recommended in F1, the target recommendation object with the largest recommendation value may be selected from the recommendation values of the objects to be recommended, and the target recommendation object is added to the designated display position in the recommendation list. The designated display position may be the last position in the recommendation list. That is, the target recommendation objects selected by the mobile phone each time are all placed at the last position of the recommendation list, and if the number of the selected target recommendation objects is equal to the predefined recommendation number, the recommendation list output to the user operation interface can be generated.
It should be noted that, in the implementation manner described in the present application, the recommendation value calculation of the object to be recommended that is determined later refers to the feature data of the previously determined sorted recommendation objects, so that the feature association between the objects to be recommended can be enhanced, the probability that a replaceable recommendation object appears in the recommendation list can be reduced, and the resource utilization rate of the recommendation list can be improved.
Further, in some possible embodiments, when the mobile phone selects a target recommendation object from each object to be recommended, the recommendation objects output to the recommendation list may be further filtered according to a predefined selection rule. Wherein, the predefined selection rule may include:
rule 1: the number of the objects to be recommended with the same or similar characteristics is not more than M1.
Wherein, M1 can be defined according to the practical application scenario, and is not limited herein. For example, if M1 is 2 and the predefined selection rule is that the number of objects to be recommended with the same or similar category in the recommendation list is not greater than 2, if there is already 2 recommended objects of category a in the sorted recommended objects, the target recommended object is not selected as the target recommended object even if the recommended value of the object to be recommended with category a is the largest when the target recommended object is selected according to the recommended value. At this time, the mobile phone may determine, as the target recommendation object, an object to be recommended of which the category is B and the recommendation value is only smaller than the maximum recommendation value. Therefore, the display position resources of the recommendation list can be prevented from being occupied by too many identical or similar objects to be recommended, and the resource utilization rate of the recommendation list is improved.
Rule 2: the online date of the object to be recommended is not later than the predefined date.
For example, if the predefined selection rule specifies that the recommendation objects displayed by the third display position of the recommendation list can only be online objects in the last week, the mobile phone determines the third target recommendation object, selects the online objects to be recommended in the last week from the objects to be recommended which are not sorted yet, and then determines the target recommendation object with the largest recommendation value from the selected objects to be recommended as the third target recommendation object. Therefore, the situation that too much display bit resources of the recommendation list are occupied by too low versions of the objects to be recommended can be avoided, the information effectiveness of the recommendation list is improved, and the applicability of the recommendation list is enhanced.
Further, in some possible embodiments, after the mobile phone determines the target recommendation object, the target recommendation object may be removed from the above F1 and added to F3, and then the feature data of the target recommendation object is added to F2 to serve as the reference feature data for determining the next target recommendation object. If the number of the sorted recommendation objects is greater than or equal to the predefined recommendation number or if F1 is empty, a recommendation list may be generated according to the sorted recommendation objects and output to the user operation interface.
Fig. 2 is another flowchart of a method for generating an information recommendation list according to an embodiment of the present application. The cyclic process of the method for generating the information recommendation list provided by the embodiment of the application comprises the following steps:
s201, data initialization.
In a specific implementation, the data initialization process includes initialization of data such as the recommendation value adjustment rule of the preset feature data set, the recommendation list, and the predefined selection rule of the target recommendation object, among the F1, F2, F3.
S202, calculating the recommendation value of each object to be recommended through a recommendation value calculation model.
In a specific implementation, the calculation manner of the recommendation value of each object to be recommended may refer to the implementation manner described in each step in the foregoing embodiment, which is not described herein again.
S203, determining the target recommendation object according to the recommendation value adjustment rule or the predefined selection rule of the target recommendation object.
In a specific implementation, the determination method of the target recommendation object may refer to the implementation methods described in the steps in the embodiments, and details are not described herein.
S204, updating F1, F2, F3 and the preset feature data set.
In a specific implementation, the updating manners of the F1, F2, F3 and the preset feature data set may join the implementation manners described in the related steps in the foregoing embodiments, and are not described herein again.
S205, determining whether the F1 is empty or whether the number of the objects to be recommended in the F3 meets the requirement.
If F1 is empty or the number of sorted recommended objects in F3 meets the requirement, go to step S206, otherwise go to step S202.
And S206, generating a recommendation list and outputting the recommendation list to a user operation interface.
In the implementation manner described in the application, each time the mobile phone determines a target recommendation object, the corresponding data set may be updated until all the objects to be recommended are sorted, or the number of sorted recommendation objects meets the requirement of a predefined number, and a recommendation list may be finally generated and output to the user operation interface to be displayed to the user.
It should be noted that, if the information recommendation list is a web page recommendation list, the feature data of the object to be recommended described in the present application may further include but is not limited to: the web page title, the main domain name of the web page, the word weight in the web page, the category of the web page, and the like can be determined according to the actual application scenario, and are not limited herein.
In the implementation manner described in the application, the determined feature data of the object to be recommended is added to the calculation of the recommendation value of the next object to be recommended, the calculation accuracy of the recommendation value is higher, the algorithm complexity controllability for generating the recommendation list is improved, and the applicability is stronger. Furthermore, the implementation manner described in the application determines the object to be recommended to be determined according to the determined feature data of the object to be recommended, so that the correlation among the objects to be recommended can be more flexibly controlled, the occurrence probability of the replaceable object to be recommended is reduced, and the utilization rate of the display position of the recommendation list is improved.
Fig. 3 is a schematic structural diagram of an apparatus for generating an information recommendation list according to an embodiment of the present application. The apparatus for generating an information recommendation list provided in the embodiment of the present application may specifically be a terminal device in the above embodiment, for example, a mobile phone, and is not limited herein. The apparatus for generating the information recommendation list (hereinafter referred to as a terminal) may include:
the obtaining module 31 is configured to obtain feature data of each object to be recommended in an object set to be recommended, where the object set to be recommended includes S1 objects to be recommended, and S1 is an integer greater than 1.
The obtaining module 31 is further configured to obtain a preset feature data set, where the preset feature data set includes feature data of S2 recommended objects, the feature data of S2 recommended objects includes feature data of specified recommended objects, and S2 is less than or equal to S1.
The calculating module 32 is configured to calculate a recommendation value of each object to be recommended according to the preset feature data set and the feature data of the S1 objects to be recommended, which are acquired by the acquiring module.
The selecting module 33 is configured to select a target recommendation object according to the recommendation value of each object to be recommended in the S1 objects to be recommended calculated by the calculating module, and add the target recommendation object to a designated display position in a recommendation list.
In some possible embodiments, the specified recommendation objects obtained by the obtaining module 31 include recommendation objects already determined to be added to the recommendation list, and the feature similarity between the specified recommendation object and the target recommendation object is smaller than a preset similarity threshold.
In some possible embodiments, the calculating module 32 is configured to:
and taking the preset feature data set and the feature data of the ith object to be recommended as input values of a preset recommendation value calculation model, and calculating the recommendation value of the ith object to be recommended through the recommendation value calculation model.
In some possible embodiments, the selecting module 33 is configured to:
selecting objects to be recommended which meet a predefined selection rule from the S1 objects to be recommended which are obtained by the obtaining module 31, and selecting the object to be recommended which has the largest recommendation value calculated by the calculating module 32 from the objects to be recommended which meet the predefined selection rule as a target recommendation object.
In some possible embodiments, the obtaining module 31 is further configured to:
and removing the characteristic data of the target recommendation object from the set of objects to be recommended, and adding the characteristic data of the target recommendation object to the preset characteristic data set.
In some possible embodiments, the object to be recommended includes: at least one of an application program APP, audio/video data, a webpage and news information.
In some possible embodiments, the characteristic data includes: at least one of an identification ID, a category attribute, an applicable platform, a number of times of reference, a click through rate, a number of downloads, and a data size.
In a specific implementation, the terminal provided in the present application may execute, through each module included in the terminal, an implementation manner described in each step in the foregoing embodiment, which is not described herein again.
Fig. 4 is another schematic structural diagram of an apparatus for generating an information recommendation list according to an embodiment of the present application. The generating device of the information recommendation list provided by the embodiment of the present application may be the terminal in the above embodiment, and may include a memory 41 and a processor 42, where the memory 41 and the processor 42 may be connected by a bus.
The memory 41 includes, but is not limited to, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM), or a portable read-only memory (CD-ROM), and the memory 41 is used for related instructions and data.
The processor 42 may be one or more Central Processing Units (CPUs), and in the case that the processor 42 is one CPU, the CPU may be a single-core CPU or a multi-core CPU.
The processor 42 is configured to read the program code stored in the memory 41 and perform the following operations:
acquiring characteristic data of each object to be recommended in an object set to be recommended, wherein the object set to be recommended comprises S1 objects to be recommended, and S1 is an integer greater than 1;
acquiring a preset feature data set, wherein the preset feature data set comprises feature data of S2 recommended objects, the feature data of S2 recommended objects comprise feature data of specified recommended objects, and S2 is less than or equal to S1;
calculating a recommendation value of each object to be recommended according to the preset feature data set and the feature data of the S1 objects to be recommended;
and selecting a target recommendation object according to the recommendation value of each object to be recommended in the S1 objects to be recommended, and adding the target recommendation object to a specified display position in a recommendation list.
In some possible embodiments, the specified recommendation object includes a recommendation object already determined to be added to the recommendation list, and the feature similarity between the specified recommendation object and the target recommendation object is smaller than a preset similarity threshold.
In some possible embodiments, the processor 42 is configured to:
and taking the preset feature data set and the feature data of the ith object to be recommended as input values of a preset recommendation value calculation model, and calculating the recommendation value of the ith object to be recommended through the recommendation value calculation model.
In some possible embodiments, the processor 42 is configured to:
and selecting the objects to be recommended which accord with a predefined selection rule from the S1 objects to be recommended, and selecting the object to be recommended with the maximum recommendation value from the objects to be recommended which accord with the predefined selection rule as a target recommendation object.
In some possible embodiments, the processor 42 is further configured to:
and removing the characteristic data of the target recommendation object from the set of objects to be recommended, and adding the characteristic data of the target recommendation object to the preset characteristic data set.
In some possible embodiments, the object to be recommended includes: at least one of an application program APP, audio/video data, a webpage and news information.
In some possible embodiments, the characteristic data includes: at least one of an identification ID, a category attribute, an applicable platform, a number of times of reference, a click through rate, a number of downloads, and a data size.
In the implementation manner provided by the application, the recommendation value of each object to be recommended can be calculated according to the feature data of each object to be recommended and the preset feature data set, wherein the feature data of the object to be recommended which is determined to be recommended is also added to the calculation of the recommendation value of the next object to be recommended, the calculation accuracy of the recommendation value is higher, the algorithm complexity controllability for generating the recommendation list is improved, and the applicability is stronger. Furthermore, the implementation manner described in the application determines the object to be recommended to be determined according to the determined feature data of the object to be recommended, so that the correlation among the objects to be recommended can be more flexibly controlled, the occurrence probability of the replaceable object to be recommended is reduced, and the utilization rate of the display position of the recommendation list is improved.
One of ordinary skill in the art will appreciate that all or part of the processes in the methods of the above embodiments may be implemented by hardware related to instructions of a computer program, which may be stored in a computer-readable storage medium, and when executed, may include the processes of the above method embodiments. And the aforementioned storage medium includes: various media capable of storing program codes, such as ROM or RAM, magnetic or optical disks, etc.
Claims (8)
1. A method for generating an information recommendation list is characterized by comprising the following steps:
the method comprises the steps that a terminal obtains feature data of each object to be recommended in an object set to be recommended, wherein the object set to be recommended comprises S1 objects to be recommended, and S1 is an integer larger than 1;
the terminal acquires a preset feature data set, wherein the preset feature data set comprises feature data of each object to be recommended in the S1 objects to be recommended, user feature data, combined features of the feature data of each object to be recommended and the user feature data, and feature data of S2 recommended objects, the feature data of the S2 recommended objects comprises feature data of a specified recommended object, and S2 is less than or equal to S1;
the terminal calculates the recommendation value of each object to be recommended according to the preset feature data set and the feature data of the S1 objects to be recommended;
the terminal selects a target recommendation object according to the recommendation value of each object to be recommended in the S1 objects to be recommended, and adds the target recommendation object to a designated display position in a recommendation list;
the terminal calculates the recommendation value of the ith object to be recommended according to the preset feature data set and the feature data of the S1 objects to be recommended, and includes:
and the terminal takes the preset characteristic data set and the characteristic data of the ith object to be recommended as input values of a preset recommended value calculation model, and calculates the recommended value of the ith object to be recommended through the recommended value calculation model.
2. The method of claim 1, wherein the specified recommended object comprises a recommended object that has been determined to be added to the recommendation list, and the characteristic similarity of the specified recommended object and the target recommended object is less than a preset similarity threshold.
3. The method according to any one of claims 1-2, wherein the terminal selecting the target recommendation object according to the recommendation value of each object to be recommended in the S1 objects to be recommended comprises:
and the terminal selects the objects to be recommended which accord with the predefined selection rule from the S1 objects to be recommended, and selects the object to be recommended with the maximum recommendation value from the objects to be recommended which accord with the predefined selection rule as the target recommendation object.
4. The method of any of claims 1-2, wherein after the target recommendation object is added to a specified presentation position in a recommendation list, the method further comprises:
and removing the characteristic data of the target recommendation object from the set of objects to be recommended, and adding the characteristic data of the target recommendation object to the preset characteristic data set.
5. An apparatus for generating an information recommendation list, comprising:
the recommendation system comprises an acquisition module, a recommendation module and a recommendation module, wherein the acquisition module is used for acquiring feature data of each object to be recommended in an object set to be recommended, the object set to be recommended comprises S1 objects to be recommended, and S1 is an integer greater than 1;
the obtaining module is further configured to obtain a preset feature data set, where the preset feature data set includes feature data of each object to be recommended in the S1 objects to be recommended, user feature data, a combined feature of the feature data of each object to be recommended and the user feature data, and feature data of S2 recommended objects, the feature data of S2 recommended objects includes feature data of a specified recommended object, and S2 is less than or equal to S1;
the calculation module is used for calculating recommendation values of the objects to be recommended according to the preset feature data set acquired by the acquisition module and the feature data of the S1 objects to be recommended;
the selecting module is used for selecting a target recommending object according to the recommendation value of each object to be recommended in the S1 objects to be recommended calculated by the calculating module, and adding the target recommending object to a specified display position in a recommending list;
wherein the computing module is to:
and taking the preset feature data set and the feature data of the ith object to be recommended as input values of a preset recommendation value calculation model, and calculating the recommendation value of the ith object to be recommended through the recommendation value calculation model.
6. The generation apparatus of claim 5, wherein the specified recommended object obtained by the obtaining module includes a recommended object that has been determined to be added to the recommendation list, and a feature similarity of the specified recommended object and the target recommended object is smaller than a preset similarity threshold.
7. The generation apparatus of any one of claims 5-6, wherein the selection module is to:
and selecting the objects to be recommended which accord with a predefined selection rule from the S1 objects to be recommended which are obtained by the obtaining module, and selecting the object to be recommended with the maximum recommendation value calculated by the calculating module as a target recommendation object from the objects to be recommended which accord with the predefined selection rule.
8. The generation apparatus of any of claims 5-6, wherein the acquisition module is further to:
and removing the characteristic data of the target recommendation object from the set of objects to be recommended, and adding the characteristic data of the target recommendation object to the preset characteristic data set.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710056067.0A CN108345419B (en) | 2017-01-25 | 2017-01-25 | Information recommendation list generation method and device |
| PCT/CN2017/115551 WO2018137420A1 (en) | 2017-01-25 | 2017-12-12 | Generating method and device for information recommendation list |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710056067.0A CN108345419B (en) | 2017-01-25 | 2017-01-25 | Information recommendation list generation method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108345419A CN108345419A (en) | 2018-07-31 |
| CN108345419B true CN108345419B (en) | 2021-06-08 |
Family
ID=62962843
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710056067.0A Active CN108345419B (en) | 2017-01-25 | 2017-01-25 | Information recommendation list generation method and device |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN108345419B (en) |
| WO (1) | WO2018137420A1 (en) |
Families Citing this family (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110827109B (en) * | 2018-08-14 | 2022-07-29 | 中国移动通信集团浙江有限公司 | Short message marketing method and device |
| CN110152298B (en) * | 2018-11-22 | 2022-03-15 | 腾讯科技（深圳）有限公司 | Equipment recommendation method and device for realizing virtual role and computer equipment |
| CN109408729B (en) * | 2018-12-05 | 2022-02-08 | 广州市百果园信息技术有限公司 | Recommended material determination method and device, storage medium and computer equipment |
| CN111401594A (en) * | 2018-12-29 | 2020-07-10 | 北京字节跳动网络技术有限公司 | Display information combination determination method and device, storage medium and electronic equipment |
| CN111026934A (en) * | 2019-12-20 | 2020-04-17 | 中科寒武纪科技股份有限公司 | Intelligent recommendation method and related equipment |
| CN115329208A (en) * | 2022-10-18 | 2022-11-11 | 平安银行股份有限公司 | Interest fee exemption scheme recommendation method and device, electronic equipment and medium |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103473244A (en) * | 2012-06-08 | 2013-12-25 | 富士通株式会社 | Device and method for recommending applications used in application group |
| CN104636419A (en) * | 2013-11-11 | 2015-05-20 | 国际商业机器公司 | Information processing device, information processing method |
| CN105843953A (en) * | 2016-04-12 | 2016-08-10 | 乐视控股（北京）有限公司 | Multimedia recommendation method and device |
| CN105989122A (en) * | 2015-02-12 | 2016-10-05 | 广东欧珀移动通信有限公司 | Application recommendation method and device |
| CN106227815A (en) * | 2015-07-22 | 2016-12-14 | Tcl集团股份有限公司 | The personalized application program function of a kind of multi-modal clue recommends method and system thereof |
| CN106326431A (en) * | 2016-08-25 | 2017-01-11 | 乐视控股（北京）有限公司 | Information recommendation method and device |
Family Cites Families (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP4247280B2 (en) * | 2007-02-14 | 2009-04-02 | 株式会社東芝 | RECOMMENDED PROGRAM INFORMATION PROVIDING DEVICE, RECOMMENDED PROGRAM INFORMATION PROVIDING METHOD, AND PROGRAM |
| JP2010231771A (en) * | 2009-03-05 | 2010-10-14 | Fujifilm Corp | Data reproduction device, content delivery system, and content delivery method |
| CN102831234B (en) * | 2012-08-31 | 2015-04-22 | 北京邮电大学 | Personalized news recommendation device and method based on news content and theme feature |
| US20140310304A1 (en) * | 2013-04-12 | 2014-10-16 | Ebay Inc. | System and method for providing fashion recommendations |
| CN103744966B (en) * | 2014-01-07 | 2018-06-22 | Tcl集团股份有限公司 | A kind of item recommendation method, device |
| CN105469263A (en) * | 2014-09-24 | 2016-04-06 | 阿里巴巴集团控股有限公司 | Commodity recommendation method and device |
| US10944707B2 (en) * | 2014-09-26 | 2021-03-09 | Line Corporation | Method, system and recording medium for providing video contents in social platform and file distribution system |
| CN105812937B (en) * | 2014-12-30 | 2019-05-24 | Tcl集团股份有限公司 | A kind of TV programme suggesting method and television program recommending device |
| CN105608117B (en) * | 2015-12-14 | 2019-12-10 | 微梦创科网络科技（中国）有限公司 | Information recommendation method and device |
| CN105930429A (en) * | 2016-04-19 | 2016-09-07 | 乐视控股（北京）有限公司 | Music recommendation method and apparatus |
| CN106021485B (en) * | 2016-05-19 | 2019-05-14 | 中国传媒大学 | A kind of polynary attribute cinematic data visualization system |
-
2017
- 2017-01-25 CN CN201710056067.0A patent/CN108345419B/en active Active
- 2017-12-12 WO PCT/CN2017/115551 patent/WO2018137420A1/en active Application Filing
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103473244A (en) * | 2012-06-08 | 2013-12-25 | 富士通株式会社 | Device and method for recommending applications used in application group |
| CN104636419A (en) * | 2013-11-11 | 2015-05-20 | 国际商业机器公司 | Information processing device, information processing method |
| CN105989122A (en) * | 2015-02-12 | 2016-10-05 | 广东欧珀移动通信有限公司 | Application recommendation method and device |
| CN106227815A (en) * | 2015-07-22 | 2016-12-14 | Tcl集团股份有限公司 | The personalized application program function of a kind of multi-modal clue recommends method and system thereof |
| CN105843953A (en) * | 2016-04-12 | 2016-08-10 | 乐视控股（北京）有限公司 | Multimedia recommendation method and device |
| CN106326431A (en) * | 2016-08-25 | 2017-01-11 | 乐视控股（北京）有限公司 | Information recommendation method and device |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108345419A (en) | 2018-07-31 |
| WO2018137420A1 (en) | 2018-08-02 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108345419B (en) | Information recommendation list generation method and device | |
| CN106651542B (en) | Article recommendation method and device | |
| CN109062919B (en) | Content recommendation method and device based on deep reinforcement learning | |
| CN110008973B (en) | Model training method, method and device for determining target user based on model | |
| CN110008397B (en) | Recommendation model training method and device | |
| CN103502899A (en) | Dynamic predictive modeling platform | |
| CN112307344B (en) | Object recommendation model, object recommendation method and device and electronic equipment | |
| CN108121814B (en) | Search result ranking model generation method and device | |
| CN111459993B (en) | Configuration updating method, device, equipment and storage medium based on behavior analysis | |
| US11061948B2 (en) | Method and system for next word prediction | |
| CN110659416B (en) | Recommendation method and recommendation device for browsing resources and readable storage medium | |
| CN110363580B (en) | Information recommendation method and device, computer equipment and storage medium | |
| WO2018161710A1 (en) | Method, device, and storage medium for recommending search term on basis of input search term | |
| CN112104505B (en) | Application recommendation method, device, server and computer readable storage medium | |
| CN111275205A (en) | Virtual sample generation method, terminal device and storage medium | |
| CN110598120A (en) | Behavior data based financing recommendation method, device and equipment | |
| CN110516164B (en) | Information recommendation method, device, equipment and storage medium | |
| CN110766513A (en) | Information sorting method and device, electronic equipment and readable storage medium | |
| CN111738774A (en) | Method, apparatus, computer device and storage medium for identifying potential target users | |
| CN115619448A (en) | User loss prediction method and device, computer equipment and storage medium | |
| CN108563648B (en) | Data display method and device, storage medium and electronic device | |
| US10579752B2 (en) | Generating a model based on input | |
| CN112765450A (en) | Recommended content determining method, recommended content determining device and storage medium | |
| CN113836388A (en) | Information recommendation method and device, server and storage medium | |
| CN108595395B (en) | Nickname generation method, device and equipment |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |