CN111078997A - Information recommendation method and device - Google Patents

Information recommendation method and device Download PDF

Info

Publication number
CN111078997A
CN111078997A CN201911129660.9A CN201911129660A CN111078997A CN 111078997 A CN111078997 A CN 111078997A CN 201911129660 A CN201911129660 A CN 201911129660A CN 111078997 A CN111078997 A CN 111078997A
Authority
CN
China
Prior art keywords
information
user
target
users
group
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
Application number
CN201911129660.9A
Other languages
Chinese (zh)
Other versions
CN111078997B (en
Inventor
张静
狄潇然
张亚泽
郑小虎
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Bank of China Ltd
Original Assignee
Bank of China Ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Bank of China Ltd filed Critical Bank of China Ltd
Priority to CN201911129660.9A priority Critical patent/CN111078997B/en
Publication of CN111078997A publication Critical patent/CN111078997A/en
Application granted granted Critical
Publication of CN111078997B publication Critical patent/CN111078997B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

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
    • 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
    • G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
    • G06F17/10—Complex mathematical operations
    • G06F17/16—Matrix or vector computation, e.g. matrix-matrix or matrix-vector multiplication, matrix factorization

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Databases & Information Systems (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Mathematical Physics (AREA)
  • Data Mining & Analysis (AREA)
  • Pure & Applied Mathematics (AREA)
  • Mathematical Analysis (AREA)
  • Mathematical Optimization (AREA)
  • Computational Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • Computing Systems (AREA)
  • Algebra (AREA)
  • Software Systems (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)

Abstract

The application provides an information recommendation method and device, relates to the technical field of internet, and can improve accuracy and diversity of information recommendation results. The method comprises the following steps: acquiring a plurality of feature tags of a plurality of users in a user group, a plurality of attribute tags of a plurality of information in an information group, interactive information of the users in the user group and interactive information of the information in the information group; the interactive information of the information is used for representing the information browsed by the users; constructing a user-information preference model by using interactive information of a first target user in a user group and attribute tags of a plurality of information browsed by the first target user; the user-information preference model is used for recommending information for the first target user; constructing an information-user audience model by using the interactive information and the feature labels of the first target information in the information group; the first target information is recommended to users in the user group using an information-user audience model.

Description

Information recommendation method and device
Technical Field
The present application relates to the field of internet technologies, and in particular, to an information recommendation method and apparatus.
Background
At present, most mobile phone APPs have implemented a message push function, but when the mobile phone APP of the commercial bank provides a message push service to the customer, the same message is pushed to all customers, and the same message may not be the content of interest to all the customers. Although the mobile phone APP of the existing bank increases the user stickiness by providing the information search service, the financial information is various in types and contents, and the user is difficult to find the content in which the user is interested from the mass information. Moreover, the timeliness and accuracy problems of the user searching information are not questioned, so that troubles are brought to the user, the user experience is reduced, and the bank mobile phone APP is limited to be better developed.
For this reason, some of the mobile phones APP of the banks use a collaborative filtering algorithm to push information. I.e., users who have selected the same product are considered to have similar preferences, the target user is recommended the product selected by the neighbor user who has selected the same product as the target user. However, the method for recommending information by using the collaborative filtering algorithm is not suitable for new users without historical browsing records, and the amount of calculated data is large when the neighbor users are selected and screened, so that the method is complicated to implement. And the bank mobile phone APP adopts an algorithm based on information content recommendation to perform information recommendation. Namely, the user is considered to like the products similar to the products selected by the user in the history, namely, the products similar to the products selected by the target user in the history are recommended to the target user. However, the same algorithm for recommending based on information content cannot deal with the situation of new user usage, and the recommendation based on information content may result in a single recommended information content and cannot meet the customer requirements. Therefore, how to better recommend information by the bank mobile phone APP becomes a problem to be solved urgently.
Disclosure of Invention
The application provides an information recommendation method and device, which can be used for recommending information for a target user by combining user interest preference and information audience characteristics so as to improve the accuracy and diversity of information recommendation contents.
In order to achieve the purpose, the technical scheme is as follows:
in a first aspect, the present application provides an information recommendation method, which may include:
acquiring a plurality of feature tags of a plurality of users in a user group, a plurality of attribute tags of a plurality of information in an information group, interactive information of the users in the user group and interactive information of the information in the information group; the interactive information of the information is used for representing the information browsed by the users;
constructing a user-information preference model by using interactive information of a first target user in a user group and attribute tags of a plurality of information browsed by the first target user; the user-information preference model is used for recommending information for a first target user;
constructing an information-user audience model by using the interactive information and the feature labels of the first target information in the information group; the information-user audience model is used to recommend first target information to users in the user group.
In a second aspect, the present application provides an information recommendation apparatus, including: the device comprises an acquisition module, a construction module and a recommendation module. The acquisition module is used for acquiring a plurality of feature tags of a plurality of users in a user group, a plurality of attribute tags of a plurality of information in an information group, interactive information of the users in the user group and interactive information of the information in the information group; the interactive information of the information is used for representing a plurality of information browsed by the user, and the interactive information of the information is used for representing the information browsed by the plurality of users. The building module is used for building a user-information preference model by utilizing the interactive information of a first target user in the user group and the attribute tags of a plurality of information browsed by the first target user; and constructing an information-user audience model by using the interactive information and the feature tags of the first target information in the information group. And the recommending module is used for recommending information for the first target user by using the user-information preference model and recommending the first target information to users in the user group by using the information-user audience model.
In a third aspect, the present application provides an information recommendation apparatus, including: a processor and a memory. Wherein the memory is used to store one or more programs. The one or more programs include computer executable instructions that, when executed by the apparatus, cause the apparatus to perform the information recommendation method of any one of the first aspect and its various alternative implementations.
In a fourth aspect, the present application provides a computer-readable storage medium, in which instructions are stored, and when the instructions are executed by a computer, the computer executes the information recommendation method according to any one of the first aspect and various optional implementations thereof.
In a fifth aspect, the present application provides a computer program product comprising instructions, which when run on a computer, cause the computer to perform the information recommendation method of any one of the first aspect and its various alternative implementations.
In a sixth aspect, the present application provides a computer program product comprising instructions, which when run on a computer, cause the computer to perform the information recommendation method of any one of the first aspect and its various alternative implementations.
The information recommendation method and device provided by the application acquire a plurality of feature tags of a plurality of users in a user group, a plurality of attribute tags of a plurality of information in an information group, interactive information of the users in the user group, and interactive information of the information in the information group; the interactive information of the information is used for representing the information browsed by the users; constructing a user-information preference model by using interactive information of a first target user in a user group and attribute tags of a plurality of information browsed by the first target user; the user-information preference model is used for recommending information for a first target user; constructing an information-user audience model by using the interactive information and the feature labels of the first target information in the information group; the information-user audience model is used to recommend first target information to users in the user group. Compared with the prior art, the information recommendation is single and the data calculation amount is large. According to the information recommendation method, firstly, a plurality of information of the user and the information, such as the characteristic mark of the user, the interactive information of the user, the attribute mark of the information and the interactive information of the information, are used, and the information is generated in the process that the user actually browses the information, so that the information can better represent the preference of the user to the information or can better represent the user to which one piece of information is more suitable, and therefore, the accuracy of a recommendation result can be ensured. Secondly, information is directly recommended to the user by using the established user-information preference model and the established information-user audience model, and the browsing history of target neighbor users with similar characteristics to the user does not need to be searched each time for analysis and calculation, so that the information recommendation efficiency can be improved. Moreover, information recommendation is performed not only depending on the browsing history of the user, so that the diversity of recommendation results can be further ensured, and the problem of recommendation of new users and new information is solved.
Drawings
FIG. 1 is a first schematic diagram illustrating an information recommendation method according to an embodiment of the present application;
FIG. 2 is a diagram illustrating a second information recommendation method according to an embodiment of the present application;
FIG. 3 is a third schematic diagram of an information recommendation method according to an embodiment of the present application;
FIG. 4 is a first schematic structural diagram of an information recommendation device according to an embodiment of the present application;
FIG. 5 is a schematic structural diagram of an information recommendation device according to an embodiment of the present application;
fig. 6 is a third schematic structural diagram of an information recommendation device according to an embodiment of the present application.
Detailed Description
The information recommendation method and apparatus provided in the embodiments of the present application are described in detail below with reference to the accompanying drawings.
The terms "first" and "second" and the like in the description and drawings of the present application are used for distinguishing different objects or for distinguishing different processes for the same object, and are not used for describing a specific order of the objects.
Furthermore, the terms "including" and "having," and any variations thereof, as referred to in the description of the present application, are intended to cover non-exclusive inclusions. For example, a process, method, system, article, or apparatus that comprises a list of steps or elements is not limited to only those steps or elements but may alternatively include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.
It should be noted that in the embodiments of the present application, words such as "exemplary" or "for example" are used to indicate examples, illustrations or explanations. Any embodiment or design described herein as "exemplary" or "e.g.," is not necessarily to be construed as preferred or advantageous over other embodiments or designs. Rather, use of the word "exemplary" or "such as" is intended to present concepts related in a concrete fashion.
In the description of the present application, the meaning of "a plurality" means two or more unless otherwise specified.
An embodiment of the present application provides an information recommendation method, as shown in fig. 1, the method may include S101-S104:
s101, the platform acquires a plurality of feature tags of a plurality of users in a user group, a plurality of attribute tags of a plurality of information in an information group, interactive information of the users in the user group and interactive information of the information in the information group.
Specifically, the platform may be a big data processing platform or other platforms having functions of data acquisition, processing, pushing, and the like, which is not specifically limited in the embodiment of the present application.
The user group is composed of users using a certain APP, and the users include old users and new users, namely, part of the users in the user group are the old users who may have historical browsing records, and part of the users are the new users who may not have the historical browsing records. The feature tag of the user is a feature of each user in the user group, and may be multi-dimensional information of gender, age, academic history, occupation, asset condition, liability condition, commercial product purchase condition, and the like of the user. It is understood that each user has at least one feature tag, and the embodiment of the present application is not limited to the included scope of the feature tags.
The information set is composed of the information in the APP information database, and the information includes old information and new information, i.e. part of the information in the information set is the old information browsed by the user, and part of the information is the new information not browsed by the user. The attribute tag of the information is an attribute feature to which each piece of information in the information group belongs, and each piece of information may include one or more aspects, for example, one piece of information includes both political content and financial content. Exemplary, the attribute tags may be multi-dimensional information such as politics, finance, entertainment, internet, gardening, and the like. It should be understood that each piece of information has at least one attribute tag, and the embodiment of the present application is not limited to the range included in the attribute tag.
The feature tag and the information tag may be manually labeled or automatically extracted by using a natural language processing technology, which is not specifically limited in this embodiment of the present application.
The interactive information of the user is used for representing a plurality of information browsed by the user, namely the information browsing condition of each user. When a user browses a piece of information, a history record is left, which indicates that the user browses the piece of information, and meanwhile, a record is also left for a feature tag of the user and an attribute tag of the information. It will be appreciated that only older users who have browsed information will have the user's interactive information.
The interactive information of the information is used for representing the condition that the information is browsed by a plurality of users, namely, each piece of information is browsed. Similarly, it can be known from the history which users have browsed each piece of information, and the attribute tag of the piece of information and the feature tag of the user who has browsed the piece of information. It is understood that only the information browsed by the user has interactive information of the information.
S102, constructing a user-information preference model by using interactive information of a first target user in the user group and attribute tags of a plurality of information browsed by the first target user.
As a possible implementation manner, taking the first target user history browsing n pieces of information as an example, as shown in fig. 2, S102 may be implemented as the following steps S1021 to S1022:
s1021, constructing a user-information preference matrix by using the interactive information of the first target user in the user group and the attribute tags of the information browsed by the first target user.
The first target user is a user in the user group with the user interaction information. The first target user u browses n pieces of information historically, and can determine that the n pieces of information correspond to m attribute tags according to the attribute tags obtained in step S101, so that a user-information preference matrix X can be constructeduComprises the following steps:
Figure BDA0002277932680000051
wherein n represents that the first target user browses n pieces of information, m represents that the n pieces of information have m attribute labels in total, and xijIndicating the grade of the ith information of the first target user,
Figure BDA0002277932680000052
n、m、i、j、
Figure BDA0002277932680000053
are all positive integers.
Alternatively to this, the first and second parts may,
Figure BDA0002277932680000054
can indicate the number of times the first target user u browses the ith piece of information.
Illustratively, there are 4 attribute tags in total, economic, entertainment, military, political, for the 4 th attribute tag (e.g., political), if the 3 rd information has the 4 th attribute tag, e.g., the content of the 3 rd information includes "political" related keywords, then
Figure BDA0002277932680000061
The number of times that the user u browses the 3 rd information is represented, and the grade of the 3 rd information by the first target user is
Figure BDA0002277932680000062
If the 2 nd information does not have the 4 th attribute label (for example, the 2 nd information is entertainment, and the content does not include words related to politics), then
Figure BDA0002277932680000063
The value of (3) is 0, the first target user's score for the 3 rd information is 0. The scoring mechanism is used for counting the browsing times of a user for a certain information, that is, when the information has a certain label, the browsing times of the information can be regarded as the score of the information under the label. In addition to the scoring mechanism referred to above, it may be manually entered by the user. Specifically, the score of the first target user on the information may be generated through a certain scoring mechanism in the APP, so that after the user browses the information, the score of the information may be input on the interface according to the preference degree of the user on the information. The scoring may be performed in a tenth system or a percentile system, which is not specifically limited in the embodiment of the present application.
Of course, the above-mentioned various scoring mechanisms can be combined to score the information. For example, combining the two scoring mechanisms, for a certain piece of information belonging to a certain label, weights are set for two scores respectively generated by the two scoring mechanisms, and the two scores are weighted to obtain a final score of the information under the label.
Illustratively, a first target user u has historically browsed 4 pieces of information v1，v2，v3，v4The 4 pieces of information have 4 attribute labels s in total1，s2，s3，s4Wherein the information v1With attribute labels s1And s3V information v2With attribute labels s1And s2V information v3With attribute labels s3V information v4With attribute labels s2And s4. First target user u pairInformation v1Is scored as 8 points, the first target user u pairs the information v2The score of is 6, the first target user u pairs the information v3Is scored as 10 points, the first target user u pairs the information v4Score of (2). The score may be generated by any of the scoring mechanisms described above.
Thus, the user-information preference matrix for the first target user u is:
Figure BDA0002277932680000071
s1022, build the user-information preference model by using the user-information preference matrix.
Specifically, the process of constructing the user-information preference model by using the user-information preference matrix is a solution process of the multi-attribute decision problem. There are three subjects in the multi-attribute decision problem: alternative set (set of all information historically viewed by first target user u) V ═ V { (V)1,v2,…,vi,…,vnThe attribute set of the measurement scheme (the set of attribute labels of all the information) S ═ S1,s2,…,sj,…,smAn attribute weight vector representing the importance of an attribute (the preference of a first target user for an attribute tag)
Figure BDA0002277932680000072
0≤wj≤1，
Figure BDA0002277932680000073
Set of alternatives V ═ { V ═ V1,v2,…,vi…,vnEach of the schemes in (j) may contain m dimensions (corresponding to m attribute tags). All alternative vectors can be combined to form a multi-attribute decision matrix, namely a user-information preference matrix Xu. Wherein the attribute weight vector
Figure BDA0002277932680000074
In, wjM attribute labels representing informationThe weight of the jth attribute tag in (1). Illustratively, the information 1 includes attribute labels of politics, finance and the internet, and three attributes constitute all the attributes of the content of the information 1, wherein among the three attribute labels, politics accounts for 70% of the weight, finance accounts for 10% of the weight, and the internet accounts for 20% of the weight.
As a possible implementation manner, in the multi-attribute decision theory, the topsis (technique for order prediction by similarity to an ideal solution) method is a very effective multi-attribute decision method. The principle of the TOPSIS method is to rank according to the proximity of a limited number of alternatives to an idealized target, and is a ranking method approaching an ideal solution. Briefly, the basic principle of the TOPSIS method is to perform ranking by detecting the distance between a candidate solution and the optimal solution and the worst solution, and if the candidate solution is closest to the optimal solution and is further away from the worst solution, the optimal solution is obtained. And all the index values of the optimal solution reach the optimal values of all the evaluation indexes, and all the index values of the worst solution reach the worst values of all the evaluation indexes. Therefore, the TOPSIS method can be used to construct the user-information preference model
Figure BDA0002277932680000081
And utilizes the idea of TOPSIS approaching to ideal solution to find the optimal solution of user-information preference model
Figure BDA0002277932680000082
Wherein the optimal solution of the user-information preference model can be expressed as
Figure BDA0002277932680000083
Specifically, the TOPSIS method for constructing the user-information preference model can be realized by the following steps:
step one, normalizing the user-information preference matrix to obtain a matrix Eu＝[eij]n×m。
The normalization process has the advantages of improving the convergence rate of the model, improving the accuracy of the model and the like, so that the normalization process is firstly carried out on the user-information preference matrix so as to optimize the establishment of the subsequent model. The specific normalization processing method may be referred to in the prior art, and is not described in detail in the embodiments of the present application.
Step two, matrix E is pairedu＝[eij]n×mAnd carrying out weighting processing.
Since the importance of the attributes of different dimensions is different, the influence of the attribute weight needs to be considered when determining the ideal solution, so that the normalized matrix E needs to be consideredu＝[eij]n×mWeighting is carried out, namely:
Figure BDA0002277932680000084
for the above matrix Y in the TOPSIS methoduIs generally a solution based on a positive or negative ideal solution:
Figure BDA0002277932680000085
Figure BDA0002277932680000086
for the embodiment of the present application, since the attribute tag of the information is neither a benefit attribute nor a cost attribute, that is, for the information, the attribute is not an element constituting a certain optimal solution of the user-information preference model because the benefit generated by a certain attribute of the information is better, and the attribute is not an element constituting a certain optimal solution of the user-information preference model because the cost corresponding to a certain attribute is lower. The above solutions according to the positive and negative ideal solutions are not applicable to the embodiments of the present application.
In step S1021, the scores of the first target users for the information are obtained, and the scores of the users for the information with the same label are different from each other according to the matrix (1). Therefore, the user's score on the information can be used to obtain the preference degree of the user to the information attribute, that is, the user's score on the information can be used to obtain the average score corresponding to each attribute label, that is, the average scoreThe user's preference for having a certain label. For example, an ideal solution may be approximated
Figure BDA0002277932680000091
Expressed as:
Figure BDA0002277932680000092
wherein the content of the first and second substances,
Figure BDA0002277932680000093
represents the ideal solution of the desired approximation and,
Figure BDA0002277932680000094
represents an optimal solution of the user-information preference model,
Figure BDA0002277932680000095
attribute label s representing information of first target user u in normalized user-information preference matrixjAverage of multiple scores. For example, in the matrix (1),
Figure BDA0002277932680000096
i.e. for the current 4 pieces of information, the attribute tag s2The average score for user u is 2.
Step three, solving the optimal solution
Figure BDA0002277932680000097
The optimization model is built based on the toposis method, i.e. it is necessary to minimize the sum of the squares of the distances between the optimal solution and the ideal solution, then,
Figure BDA0002277932680000101
for convenience, let
Figure BDA0002277932680000102
And the Lagrange equation is constructed as follows:
Figure BDA0002277932680000103
are respectively paired
Figure BDA0002277932680000104
And λ and let equation be 0, i.e.:
Figure BDA0002277932680000105
solving equation (1) above yields the following solution:
Figure BDA0002277932680000106
for Lagrange's equation
Figure BDA0002277932680000107
In (1)
Figure BDA0002277932680000108
Solving second-order partial derivatives to construct a black plug Hesse matrix as follows:
Figure BDA0002277932680000111
due to the fact that
Figure BDA0002277932680000112
Therefore, it is not only easy to use
Figure BDA0002277932680000113
Is a positive definite matrix. Therefore, the temperature of the molten metal is controlled,
Figure BDA0002277932680000114
is the optimal solution for user-information preferences.
Step four, constructing a user-information preference model
Figure BDA0002277932680000115
According to the aboveSolving the optimal solution in the third step
Figure BDA0002277932680000116
The method of (1) obtaining the optimal solution corresponding to each attribute tag in the m attribute tags can be expressed as
Figure BDA0002277932680000117
Further build a user-information preference model
Figure BDA0002277932680000118
Each element in the user-information preference model can be represented by an optimal solution, i.e.
Figure BDA0002277932680000119
S103, recommending information for the first target user by using the user-information preference model.
As a possible implementation manner, information with the preference degree meeting the first condition is recommended for the first target user.
Specifically, a user-information preference model can be created for each user according to the above method, and the description can be solved according to the above ideal solution
Figure BDA00022779326800001110
The average value of the historical scores of the user for some information attribute labels and the user-information preference model of the user can be used to determine the score of the user for the information, wherein the score is the preference degree of the user for the information. In this case, the first condition may be information in which the user scores higher than a first predetermined threshold, and the information satisfying the first condition may be recommended to the user.
Illustratively, the user-information preference matrix for the first target user is:
Figure BDA0002277932680000121
the user-information preference model is:
Figure BDA0002277932680000122
at this time, the first information and the attribute tag included in the first information are obtained, and it is assumed that the first information includes the attribute tag: s1And s3. First, an average rating of a first target user for each attribute tag included in the first information is obtained. Then the attribute label s of the first target user pair can be obtained by referring to the user-information preference matrix of the first target user1Is (8+ 6)/2-7, the first target user is given the attribute label s3The average score of (8+ 10)/2-9. Then, a first score of the first target user for the first information is obtained according to the average score and the user-information preference model
Figure BDA0002277932680000123
The historical average score of the first target user for the attribute labels included in the first information is multiplied by the corresponding weight, and the sum of the historical average score and the corresponding weight is the first score. Assuming that the first predetermined threshold is 5 and the first score is 5.1 > 5, the first information is recommended to the first target user. The first preset threshold may be obtained according to an empirical value, which is not limited in this embodiment of the application. The recommended information is the information of the most preferred aspect of the user, so that the use interest of the user can be increased, and the client stickiness can be increased. Even a new piece of information can be recommended to the corresponding user according to the attribute label of the information so as to solve the problem of cold start of a new product.
Optionally, the first condition may also be a preset number of pieces of information whose preference degrees meet the preset condition, that is, the first scores higher than the first preset threshold may be sorted, and a certain number of pieces of information sorted in advance are recommended to the corresponding users, so that a situation that the satisfaction of the users is reduced by recommending an excessive number of pieces of information to the users is avoided. Correspondingly, when the number of the information with the preference degree meeting the preset condition (the first score is larger than the first preset threshold) is less than the preset number, recommending the information with the whole number to the corresponding user so as to meet the user requirement.
Optionally, because the information has timeliness, when the first scores of some information are the same or similar, a certain amount of information with the latest release time in the information can be recommended to the user, so that the information with higher quality can be recommended to the user, and the satisfaction degree of the user is improved.
S104, constructing an information-user audience model by using the interactive information and the feature labels of the first target information in the information group.
As a possible implementation manner, for example, the first target information history is browsed by d users, as shown in fig. 3, S104 may be implemented as the following steps S1041 to S1042:
s1041, constructing an information-user audience matrix by using the interactive information of the first target information in the information group and the feature tags of a plurality of users browsing the first target information.
Wherein, the first target information history is browsed by d users, and it can be determined that the d users correspond to c feature tags according to the feature tags obtained in step S101, so that an information-user audience matrix Z can be constructedvComprises the following steps:
Figure BDA0002277932680000131
wherein d indicates that d users in the information group browse the first target information, c indicates that the d users have c feature tags in total, and zabIndicating the grade of the first target information of the a-th user,
Figure BDA0002277932680000132
d、c、a、b、
Figure BDA0002277932680000133
are all positive integers;
alternatively to this, the first and second parts may,
Figure BDA0002277932680000134
may indicate the rating of the first target information by the user browsing the first target information. The score may be manually entered by the user, for example, the score may be entered through some scoring mechanism in the APPSo that the user can score the information according to the user's preference after browsing the information. The scoring may be performed in a tenth system or a percentile system, which is not specifically limited in the embodiment of the present application. Of course, other scoring mechanisms may be used to score the information. Or, combining multiple scoring mechanisms, and setting corresponding weight for each scoring mechanism, so as to obtain the final score of the first target information under each label.
Illustratively, there are 4 users u in the history1，u2，u3，u4Browsing the first target information, the 4 users have a total of 4 feature tags p1，p2，p3，p4Wherein user u1With a characteristic label p1And p3User u2With a characteristic label p1And p2User u3With a characteristic label p3User u4With a characteristic label p2And p4. User u1The first target information is scored as 8 points, and the user u2The first target information is scored as 6 points, user u3The first target information is scored as 10 points, user u4The score for the first target information is 2. The score may be generated by any of the scoring mechanisms described above.
Thus, the information-user audience matrix of the first target information v is:
Figure BDA0002277932680000141
s1042, constructing an information-user audience model by using the information-user audience matrix.
Specifically, the process of establishing the information-user audience matrix is a solution process of a multi-attribute decision problem. There are three subjects in the multi-attribute decision problem: alternative set (set of all users who have historically browsed first target information v) U ═ U { (U) }1,u2,…,ua,…,udThe attribute set of the measurement scheme (the characteristics of all the users)Set of token tags) P ═ { P ═ P1,p2,…,pb,…,pcAn attribute weight vector representing the importance of the attribute (the matching degree of the first target information and a feature label)
Figure BDA0002277932680000142
0≤wb≤1，
Figure BDA0002277932680000143
Alternative set U ═ U1,u2,…,ua,…,udEach of the schemes in (j) may contain c dimensions (corresponding to c feature labels). All the solution vectors are combined to form a multi-attribute decision matrix, i.e. an information-user audience matrix Zv. Wherein the attribute weight vector
Figure BDA0002277932680000144
In, wbThe weight of the b-th feature label in the c feature labels representing the information. Illustratively, user 1 contains feature tags of male, programmer and average monthly income 50K, and the three features constitute all the feature attributes of user 1, wherein of the three feature tags, male accounts for 30% of the weight, programmer accounts for 40% of the weight, and average monthly income 50K accounts for 30% of the weight.
As a possible implementation, the TOPSIS method can also be used to construct an information-user audience model
Figure BDA0002277932680000151
And utilizes the idea of TOPSIS method to approximate ideal solution to obtain the optimum solution of information-user audience model
Figure BDA0002277932680000152
Wherein, the optimal solution of the information-user audience model can be expressed as
Figure BDA0002277932680000153
Similar to the building steps of the user-information preference model, the building of the information-user audience model by using the TOPSIS method can be realized by the following steps:
step one, normalizing the information-user audience matrix to obtain a matrix Ev＝[eab]d×c。
Step two, matrix E is pairedv＝[eab]d×cAnd carrying out weighting processing.
Namely, it is
Figure BDA0002277932680000154
In step S1041, the scores of the users on the first target information are obtained, and the scores of the users with a same feature tag on the first target information are different from each other as shown in the matrix (2). Therefore, the matching degree between the first target information and the user attribute tag can be obtained by using the score of the user on the first target information, that is, the average score corresponding to each feature tag obtained by using the score of the user on the first target information is the matching degree between the feature tag and the first target information. For example, an ideal solution may be approximated
Figure BDA0002277932680000155
Expressed as:
Figure BDA0002277932680000156
wherein the content of the first and second substances,
Figure BDA0002277932680000157
represents the ideal solution of the desired approximation and,
Figure BDA0002277932680000158
represents an optimal solution of the above information-user audience model,
Figure BDA0002277932680000159
representing the corresponding characteristic label p calculated and obtained by the user after scoring the first target information v in the normalized information-user audience matrixbAverage of multiple scores. For example, in the matrix (2),
Figure BDA00022779326800001510
i.e. feature tag p for the current 4 users browsing the first target information v2Was scored as 2 on average.
Step three, solving the optimal solution
Figure BDA0002277932680000161
See step S1022 above for the optimal solution in the process of establishing the user-information preference model
Figure BDA0002277932680000162
Based on the same algorithm idea, the solving process of (1) obtains:
Figure BDA0002277932680000163
wherein the content of the first and second substances,
Figure BDA0002277932680000164
step four, constructing an information-user audience model
Figure BDA0002277932680000165
Solving the optimal solution according to the third step
Figure BDA0002277932680000166
The method of (1) obtaining the optimal solution corresponding to each feature tag in the c feature tags can be expressed as
Figure BDA0002277932680000167
Further build a user-information preference model
Figure BDA0002277932680000168
Each element in the user-information preference model can be represented by an optimal solution, i.e.
Figure BDA0002277932680000169
S105, recommending the first target information to users in the user group by using the information-user audience model.
As a possible implementation manner, the first target information is recommended to the users whose matching degree in the information-user audience model meets the second condition.
Specifically, an information-user audience model may be created for each piece of information according to the above method, and at this time, the description is solved according to the above ideal solution, so that the user can know the result
Figure BDA00022779326800001610
The information may be used to average historical scores for certain subscriber feature tags. That is, if the average rating of the information by the user with some feature tags in the history is 8 for the 40-year-old user 1 and 6 for the 40-year-old user 2, the average rating of the information for the feature tag of 40 years old is (8+ 6)/2-7. And an information-user audience model of the information, determining the scores of the information on the users, wherein the scores are the matching degree of the information and the users. In this case, the second condition may be set to the user whose score of the information to the user is higher than the second preset threshold, and the information may be recommended to the users who satisfy the second condition.
Illustratively, the information-user audience matrix of the first target information is:
Figure BDA0002277932680000171
the information-user audience model is:
Figure BDA0002277932680000172
at this time, the first user and the feature tag included in the first user are obtained, and it is assumed that the first user includes the feature tag: p is a radical of1And p3. First, an average score of first target information for each feature tag included in a first user is obtained. Then the first target information can be obtained by referring to the information-user audience matrix of the first target informationSign tag p1Is (8+ 6)/2-7, the first target information is corresponding to the feature label p3The average score of (8+ 10)/2-9. Then, according to the average score and the information-user audience model, the second score of the first target information to the first user is obtained as
Figure BDA0002277932680000173
The historical average score of the feature tag included in the first user is multiplied by the corresponding weight by using the first target information, and the sum of the results is the second score. If the second predetermined threshold is 5 and the second score is 5.1 > 5, the first target information is recommended to the first user. The second threshold may be obtained according to an empirical value, which is not limited in this embodiment of the application. At this time, the user recommended by the information is the user with higher matching degree with the information, so that the interest of the user can be increased, and the client stickiness can be increased. Even a new user can recommend corresponding information to the user according to the characteristic label of the user, and the recommended information can meet the requirements of the client so as to solve the problem of cold start of the new user.
Optionally, in order to avoid recommending too much information to the same user, the second scores higher than the second preset threshold corresponding to a certain user may be ranked, and a certain amount of information ranked first is recommended to the user, so as to avoid a situation that the satisfaction degree of the user is reduced by recommending too much information to the user. Correspondingly, when the number of the information with the matching degree meeting the preset condition (the second score is larger than the second preset threshold) is less than the preset number, recommending the information with the whole number to the corresponding user so as to meet the user requirement.
Optionally, because the information has timeliness, when the second scores of some information to the same user are the same or similar, a certain amount of information with the latest release time in the information can be recommended to the user, so that the information with higher quality can be recommended to the user, and the satisfaction degree of the user is improved.
Further, through the above steps, establishing the user-information preference model and the information-user audience model may already be implemented as user recommendation information, but in order to achieve a better recommendation effect, the information recommendation method of the embodiment of the present application may further include the following step S106:
s106, according to the recommendation results of the first preset condition fusion user-information preference model and the information-user audience model, the optimal E pieces of information are recommended to the first target user or the first target information is recommended to the optimal F users.
Optionally, a first preset condition may be set according to the experience value, for example, the weight of the user-information preference model in 60% of the recommendation result, and the weight of the information-user audience model in 40% of the recommendation result, to perform information fusion recommendation. For example, it is assumed that the first preset threshold and the second preset threshold are both 5; the first score of the user A to the information 1 is 8, and the second score of the information 1 to the user A is 6; a first rating of 6 for information 2 by UserA and a second rating of 8 for UserA by UserA is provided. Then, if no fusion recommendation is made, both the information 1 and the information 2 can be recommended to the user A. If the fusion recommendation is made, the fusion score of the information 1 is (8 × 60% +6 × 40%) -7.2; the fusion score for info 2 was (6 × 60% +8 × 40%) 6.8. If the preset threshold value of the fusion recommendation score is 7, the fusion score of the information 1 is larger than 7, the information 1 is recommended to the user A, and the fusion score of the information 2 is smaller than 7, the information 2 is not recommended to the user A. By fusing the recommendation results of the user-information preference model and the information-user audience model, the accuracy and diversity of the recommendation results are further ensured. The information recommendation result not only ensures that the recommended information meets the preference degree of the user, but also meets the matching degree of the recommended information and the user and meets the requirement.
Optionally, the number may be preset in the first preset condition, the scores greater than the preset threshold in the fusion scores of all the information of a certain user are sorted, and a certain number of pieces of information sorted in the first order are selected and recommended to the user, or the scores greater than the preset threshold in the fusion scores of all the users of a certain information are sorted, and a certain number of users sorted in the first order are selected and recommended to the users. That is, E pieces of information are preferentially selected from the recommendation results of the two models and recommended to the first target user, or F users are preferentially selected and the first target information is recommended to the F users. Where E and F may be positive integers set according to empirical values. This avoids situations where an excessive amount of information is recommended to the user, which reduces the user's satisfaction. Correspondingly, when the number of the information with the fusion score meeting the preset condition (larger than the preset threshold) is less than the preset number, the information with the total number is recommended to the corresponding user so as to meet the user requirement.
Optionally, because the information has timeliness, when the fusion scores of some information are the same or similar, a certain amount of information with the latest release time in the information can be recommended to the corresponding user, so that the information with higher quality can be recommended to the user, and the satisfaction degree of the user is improved.
The information recommendation method provided by the application obtains a plurality of feature tags of a plurality of users in a user group, a plurality of attribute tags of a plurality of information in the information group, interactive information of the users in the user group and interactive information of the information in the information group; the interactive information of the information is used for representing the information browsed by the users; constructing a user-information preference model by using interactive information of a first target user in a user group and attribute tags of a plurality of information browsed by the first target user; the user-information preference model is used for recommending information for a first target user; constructing an information-user audience model by using the interactive information and the feature labels of the first target information in the information group; the information-user audience model is used to recommend first target information to users in the user group. Compared with the prior art, the information recommendation is single and the data calculation amount is large. According to the information recommendation method, the constructed user-information preference model and the information-user audience model are used for recommending information for the user, so that the efficiency can be improved, the accuracy and diversity of recommendation results are ensured, and the recommendation problem of new users and new information is solved.
In the embodiment of the present application, the device may be divided into the functional modules or the functional units according to the method example, for example, each functional module or functional unit may be divided corresponding to each function, or two or more functions may be integrated into one processing module. The integrated module may be implemented in a form of hardware, or may be implemented in a form of a software functional module or a functional unit. The division of the modules or units in the embodiment of the present application is schematic, and is only a logic function division, and there may be another division manner in actual implementation.
Fig. 4 shows a schematic diagram of a possible structure of the device according to the above-described embodiment. The apparatus 400 includes an obtaining module 401, a constructing module 402, and a recommending module 403.
The acquiring module 401 is configured to acquire a plurality of feature tags of a plurality of users in a user group, a plurality of attribute tags of a plurality of information in an information group, interaction information of the users in the user group, and interaction information of the information in the information group; the interactive information of the information is used for representing a plurality of information browsed by the user, and the interactive information of the information is used for representing the information browsed by the plurality of users.
A building module 402, configured to build a user-information preference model by using the interaction information of a first target user in the user group and attribute tags of a plurality of information browsed by the first target user; the method is also used for constructing an information-user audience model by utilizing the interactive information and the feature labels of the first target information in the information group.
A recommending module 403 for recommending information for the first target user by using the user-information preference model and recommending the first target information to users in the user group by using the information-user audience model.
In a possible implementation manner, the building module 402 is configured to build a user-information preference model by using the interaction information of the first target user in the user group and the attribute tags of the pieces of information browsed by the first target user, and specifically may be:
constructing a user-information preference matrix by using interactive information of a first target user in a user group and attribute tags of a plurality of information browsed by the first target user; the user-information preference matrix XuComprises the following steps:
Figure BDA0002277932680000201
wherein n represents that the first target user browses n pieces of information, m represents that the n pieces of information have m attribute tags, and xijIndicating the grade of the ith information of the first target user,
Figure BDA0002277932680000202
n、m、i、j、
Figure BDA0002277932680000203
are all positive integers;
building user-information preference model using user-information preference matrix
Figure BDA0002277932680000204
0≤wj≤1，
Figure BDA0002277932680000205
Wherein, wjIndicating the preference degree of the j-th attribute tag by the first target user.
The building module 402 is configured to build an information-user audience model by using the interactive information and the feature tags of the first target information in the information group, and specifically may be:
constructing an information-user audience matrix by using interactive information of first target information in the information group and feature tags of a plurality of users browsing the first target information; the information-user audience matrix ZvComprises the following steps:
Figure BDA0002277932680000206
wherein d indicates that d users in the information group browse the first target information, c indicates that d users have c feature tags, zabIndicating the grade of the first target information of the a-th user,
Figure BDA0002277932680000211
Figure BDA0002277932680000212
d、c、a、b、
Figure BDA0002277932680000213
are all positive integers;
construction of information-user audience model using information-user audience matrix
Figure BDA0002277932680000214
0≤wb≤1，
Figure BDA0002277932680000215
Wherein the content of the first and second substances,
Figure BDA0002277932680000216
indicating the matching degree of the first target information and the b-th feature tag.
In one possible implementation, the recommending module 403 is configured to recommend information for the first target user by using the user-information preference model and recommend the first target information to the users in the user group by using the information-user audience model. The method specifically comprises the following steps:
recommending information with preference degree meeting a first condition for a first target user by using the user-information preference model;
recommending the first target information to the users in the user group by using the information-user audience model comprises: recommending the first target information to the user whose matching degree satisfies the second condition.
In a possible implementation manner, the recommending module 403, using the user-information preference model to recommend, for a first target user, information whose preference degree satisfies a first condition, includes: acquiring the first information and the attribute tag contained in the first information. And acquiring the average score of the first target user on the attribute tags included in each first information. And obtaining a first score of the first target user for the first information according to the average score and the user-information preference model. If the first score is higher than a first preset threshold value, the first information is information with the preference degree meeting a first condition, and the first information is recommended to the first target user. Recommending the first target information to the user whose matching degree meets the second condition by using the information-user audience model comprises the following steps: and acquiring the first user and the feature tag contained in the first user. And acquiring the average score of the first target information on the feature tags included by each first user. And obtaining a second score of the first target information to the first user according to the average score and the information-user audience model. If the second score is higher than a second preset threshold, the first user is a user with the matching degree meeting a second condition, and the first target information is recommended to the first user.
In a possible implementation manner, the recommending module 403 is further configured to combine the recommending results of the user-information preference model and the information-user audience model according to a first preset condition, and recommend the optimal E pieces of information to the first target user or recommend the first target information to the optimal F users; wherein E and F are positive integers.
The information recommendation device provided by the application acquires a plurality of feature tags of a plurality of users in a user group, a plurality of attribute tags of a plurality of information in the information group, interactive information of the users in the user group and interactive information of the information in the information group; the interactive information of the information is used for representing the information browsed by the users; constructing a user-information preference model by using interactive information of a first target user in a user group and attribute tags of a plurality of information browsed by the first target user; the user-information preference model is used for recommending information for a first target user; constructing an information-user audience model by using the interactive information and the feature labels of the first target information in the information group; the information-user audience model is used to recommend first target information to users in the user group. Compared with the prior art, the information recommendation is single and the data calculation amount is large. The information recommending device provided by the application utilizes the established user-information preference model and the established information-user audience model to recommend information for the user, can improve the efficiency, ensures the accuracy and diversity of recommendation results, and solves the recommendation problem of new users and new information.
FIG. 5 is a schematic diagram of a possible structure of the information recommendation apparatus. As shown in fig. 5, the apparatus 500 may include: a processor 501 and a communication interface 502. The processor 501 is configured to control and manage the actions of the apparatus 500, for example, to perform the steps performed by the obtaining module 401, the constructing module 402, and the recommending module 403, and/or to perform other processes of the techniques described herein. The communication interface 502 is used to support the communication of the apparatus with other network entities. The apparatus 500 may further comprise a memory 503 and a bus 504, the memory 503 being adapted to store program codes and data of the apparatus.
Memory 503 may include, among other things, volatile memory, such as random access memory; the memory may also include non-volatile memory, such as read-only memory, flash memory, a hard disk, or a solid state disk; the memory may also comprise a combination of memories of the kind described above.
The processor 501 may be any means that can implement or execute the various illustrative logical blocks, modules, and circuits described in connection with the disclosure herein. The processor may be a central processing unit, general purpose processor, digital signal processor, application specific integrated circuit, field programmable gate array or other programmable logic device, transistor logic device, hardware component, or any combination thereof. Which may implement or perform the various illustrative logical blocks, modules, and circuits described in connection with the disclosure. The processor may also be a combination of computing functions, e.g., comprising one or more microprocessors, DSPs, and microprocessors, among others.
The bus 504 may be an Extended Industry Standard Architecture (EISA) bus or the like. The bus 504 may be divided into an address bus, a data bus, a control bus, and the like. For ease of illustration, only one thick line is shown in FIG. 5, but this is not intended to represent only one bus or type of bus.
As shown in fig. 6, another possible structure diagram of an information recommendation device is provided in the present embodiment. The apparatus 600 comprises: a processing unit 601. The processing unit 601 is used for controlling and managing actions of the apparatus 400, for example, performing the steps performed by the obtaining module 401, the constructing module 402, the recommending module 403, and/or other processes for performing the techniques described herein. The apparatus may further comprise a storage unit 602 and a communication unit 603, the storage unit 602 being adapted to store program codes and data of the apparatus; the communication unit 603 is configured to support communication of the apparatus with other network entities.
With reference to fig. 5 and fig. 6, the processing unit 601 may be the processor 501 or the controller in the apparatus 500.
The storage unit 602 may be a memory or the like in the apparatus 500, which may include a volatile memory, such as a random access memory; the memory may also include non-volatile memory, such as read-only memory, flash memory, a hard disk, or a solid state disk; the memory may also comprise a combination of memories of the kind described above.
The communication unit 603 may be the communication interface 502 in the apparatus 500, or a transceiver, a transceiving circuit, etc.
Through the above description of the embodiments, it is clear to those skilled in the art that, for convenience and simplicity of description, the foregoing division of the functional modules is merely used as an example, and in practical applications, the above function distribution may be completed by different functional modules according to needs, that is, the internal structure of the device may be divided into different functional modules to complete all or part of the above described functions. For the specific working processes of the system, the apparatus and the unit described above, reference may be made to the corresponding processes in the foregoing method embodiments, and details are not described here again.
The embodiment of the present application further provides a computer-readable storage medium, where instructions are stored in the computer-readable storage medium, and when the information recommendation device executes the instructions, the device executes each step executed by the information recommendation device in the method flow shown in the foregoing method embodiment.
The computer readable storage medium may be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples (a non-exhaustive list) of the computer readable storage medium would include the following: an electrical connection having one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM), a register, a hard disk, an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing, or any other form of computer readable storage medium known in the art. An exemplary storage medium is coupled to the processor such the processor can read information from, and write information to, the storage medium. Of course, the storage medium may also be integral to the processor. The processor and the storage medium may reside in an Application Specific Integrated Circuit (ASIC). In embodiments of the present application, a computer readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device.
The above description is only an embodiment of the present application, but the scope of the present application is not limited thereto, and any changes or substitutions within the technical scope of the present disclosure should be covered by the scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.

Claims (11)

1. An information recommendation method, comprising:
acquiring a plurality of feature tags of a plurality of users in a user group, a plurality of attribute tags of a plurality of information in an information group, interactive information of the users in the user group and interactive information of the information in the information group; the interactive information of the user is used for representing a plurality of information browsed by the user, and the interactive information of the information is used for representing that the information is browsed by a plurality of users;
constructing a user-information preference model by using interactive information of a first target user in the user group and attribute tags of a plurality of information browsed by the first target user; the user-information preference model is used for recommending information for the first target user;
constructing an information-user audience model by using the interactive information and the feature labels of the first target information in the information group; and recommending the first target information to users in the user group by using the information-user audience model.
2. The information recommendation method of claim 1, wherein the constructing a user-information preference model using the interaction information of the first target user in the user group and the attribute tags of the plurality of information browsed by the first target user comprises:
constructing a user-information preference matrix by using interactive information of a first target user in the user group and attribute labels of a plurality of information browsed by the first target user; the user-information preference matrix XuComprises the following steps:
Figure FDA0002277932670000011
wherein n represents that the first target user browses n pieces of information, m represents that the n pieces of information have m attribute tags, and xijIndicating the grade of the ith piece of information of the first target user,
Figure FDA0002277932670000012
n、m、i、j、
Figure FDA0002277932670000013
are all positive integers;
constructing the user-information preference model using the user-information preference matrix
Figure FDA0002277932670000014
0≤wj≤1，
Figure FDA0002277932670000015
Wherein, wjIndicating the preference degree of the first target user to the jth attribute tag.
3. The information recommendation method according to claim 1, wherein the constructing an information-user audience model by using the interactive information and feature tags of the first target information in the information group comprises:
constructing an information-user audience matrix by using the interactive information of the first target information in the information group and the feature tags of a plurality of users who browse the first target information; the information-user audience matrix ZvComprises the following steps:
Figure FDA0002277932670000021
wherein d represents that d users in the information group browse the first target information, c represents that the d users have c feature tags, zabIndicating the grade of the first target information of the a-th user,
Figure FDA0002277932670000022
d、c、a、b、
Figure FDA0002277932670000023
are all positive integers;
constructing the information-user audience model using the information-user audience matrix
Figure FDA0002277932670000024
0≤wb≤1，
Figure FDA0002277932670000025
Wherein, wbIndicating the matching degree of the first target information and the b-th characteristic label.
4. The information recommendation method according to any one of claims 1-3,
the user-information preference model is specifically used for recommending information with preference degree meeting a first condition for the first target user;
the information-user audience model is specifically used for recommending the first target information to the users with the matching degree meeting a second condition.
5. The information recommendation method according to claim 4, wherein said recommending information for said first target user whose preference degree satisfies a first condition comprises:
acquiring first information and an attribute tag contained in the first information;
acquiring the average score of the first target user on the attribute tags contained in each first information;
obtaining a first score of the first target user on the first information according to the average score and the user-information preference model;
if the first score is higher than a first preset threshold value, the first information is information of which the preference degree meets a first condition, and the first information is recommended to the first target user.
6. The information recommendation method of claim 4, wherein the recommending the first target information to the user whose matching degree satisfies the second condition comprises:
acquiring a first user and a feature tag contained in the first user;
acquiring the average score of the first target information on the feature tags contained in each first user;
obtaining a second score of the first target information for the first user according to the average score and the information-user audience model;
and if the second score is higher than a second preset threshold value, the first user is a user with the matching degree meeting a second condition, and the first target information is recommended to the first user.
7. The information recommendation method according to claim 5 or 6, further comprising:
according to a first preset condition, the recommendation results of the user-information preference model and the information-user audience model are fused, and the optimal E pieces of information are recommended to the first target user or the first target information is recommended to the optimal F users; wherein E and F are positive integers.
8. An information recommendation apparatus for performing the information recommendation method of any one of claims 1 to 7.
9. An information recommendation apparatus, comprising: a processor and a memory, wherein the memory is used for storing one or more programs, the one or more programs comprise computer-executable instructions, and when the information recommendation device runs, the processor executes the computer-executable instructions stored in the memory to enable the information recommendation device to execute the information recommendation method according to any one of claims 1 to 7.
10. A computer-readable storage medium having instructions stored therein, wherein when the instructions are executed by a computer, the computer performs the information recommendation method according to any one of claims 1 to 7.
11. A computer program product comprising instructions for executing the method of information recommendation according to any one of claims 1 to 7 when said computer program product is run on a computer.
CN201911129660.9A 2019-11-18 2019-11-18 Information recommendation method and device Active CN111078997B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201911129660.9A CN111078997B (en) 2019-11-18 2019-11-18 Information recommendation method and device

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201911129660.9A CN111078997B (en) 2019-11-18 2019-11-18 Information recommendation method and device

Publications (2)

Publication Number Publication Date
CN111078997A true CN111078997A (en) 2020-04-28
CN111078997B CN111078997B (en) 2023-08-22

Family

ID=70311125

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201911129660.9A Active CN111078997B (en) 2019-11-18 2019-11-18 Information recommendation method and device

Country Status (1)

Country Link
CN (1) CN111078997B (en)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111538905A (en) * 2020-04-29 2020-08-14 中国银行股份有限公司 Object recommendation method and device
CN113592198A (en) * 2021-08-24 2021-11-02 腾讯科技（深圳）有限公司 Method for determining demand reference information, server and terminal
CN113836347A (en) * 2020-06-23 2021-12-24 北京达佳互联信息技术有限公司 Recommendation method and device and electronic equipment

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108711075A (en) * 2018-05-22 2018-10-26 阿里巴巴集团控股有限公司 A kind of Products Show method and apparatus
CN109543111A (en) * 2018-11-28 2019-03-29 广州虎牙信息科技有限公司 Recommendation information screening technique, device, storage medium and server
US20190114687A1 (en) * 2017-10-17 2019-04-18 Adobe Systems Incorporated Global Vector Recommendations Based on Implicit Interaction and Profile Data
CN110457581A (en) * 2019-08-02 2019-11-15 达而观信息科技（上海）有限公司 A kind of information recommended method, device, electronic equipment and storage medium

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20190114687A1 (en) * 2017-10-17 2019-04-18 Adobe Systems Incorporated Global Vector Recommendations Based on Implicit Interaction and Profile Data
CN108711075A (en) * 2018-05-22 2018-10-26 阿里巴巴集团控股有限公司 A kind of Products Show method and apparatus
CN109543111A (en) * 2018-11-28 2019-03-29 广州虎牙信息科技有限公司 Recommendation information screening technique, device, storage medium and server
CN110457581A (en) * 2019-08-02 2019-11-15 达而观信息科技（上海）有限公司 A kind of information recommended method, device, electronic equipment and storage medium

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111538905A (en) * 2020-04-29 2020-08-14 中国银行股份有限公司 Object recommendation method and device
CN111538905B (en) * 2020-04-29 2023-10-20 中国银行股份有限公司 Object recommendation method and device
CN113836347A (en) * 2020-06-23 2021-12-24 北京达佳互联信息技术有限公司 Recommendation method and device and electronic equipment
CN113592198A (en) * 2021-08-24 2021-11-02 腾讯科技（深圳）有限公司 Method for determining demand reference information, server and terminal
CN113592198B (en) * 2021-08-24 2023-09-26 腾讯科技（深圳）有限公司 Method, server and terminal for determining demand reference information

Also Published As

Publication number Publication date
CN111078997B (en) 2023-08-22

Similar Documents

Publication Publication Date Title
CN111061946B (en) Method, device, electronic equipment and storage medium for recommending scenerized content
CN105224699B (en) News recommendation method and device
US10963541B2 (en) Systems, methods, and apparatuses for implementing a related command with a predictive query interface
CN108829808B (en) Page personalized sorting method and device and electronic equipment
CN109033101B (en) Label recommendation method and device
CN108885624B (en) Information recommendation system and method
US8484099B1 (en) Method, medium, and system for behavior-based recommendations of product upgrades
EP4181026A1 (en) Recommendation model training method and apparatus, recommendation method and apparatus, and computer-readable medium
US20150339759A1 (en) Detecting product attributes associated with product upgrades based on behaviors of users
CN104199896A (en) Video similarity determining method and video recommendation method based on feature classification
CN105426550B (en) Collaborative filtering label recommendation method and system based on user quality model
CN111078997B (en) Information recommendation method and device
US20150154685A1 (en) Automated detection of new item features by analysis of item attribute data
Zuva et al. A survey of recommender systems techniques, challenges and evaluation metrics
CN113077317A (en) Item recommendation method, device and equipment based on user data and storage medium
WO2023000491A1 (en) Application recommendation method, apparatus and device, and computer-readable storage medium
US11176486B2 (en) Building and matching electronic standards profiles using machine learning
CN114329207A (en) Multi-service information sequencing system, method, storage medium and electronic equipment
CN117421355A (en) Search recall method, device and equipment
CN105512298A (en) Interested content prediction method based on machine learning
CN105550282A (en) User interest forecasting method by utilizing multidimensional data
Singh et al. Relative similarity based approach for improving aggregate recommendation diversity
CN111353793A (en) CRM (customer relationship management) service recommendation method and device
Xie et al. Interactive resource recommendation with optimization by tag association and significance analysis
CN111831892A (en) Information recommendation method, information recommendation device, server and storage medium

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