CN111199459B - Commodity recommendation method, commodity recommendation device, electronic equipment and storage medium - Google Patents

Commodity recommendation method, commodity recommendation device, electronic equipment and storage medium Download PDF

Info

Publication number
CN111199459B
CN111199459B CN201911394011.1A CN201911394011A CN111199459B CN 111199459 B CN111199459 B CN 111199459B CN 201911394011 A CN201911394011 A CN 201911394011A CN 111199459 B CN111199459 B CN 111199459B
Authority
CN
China
Prior art keywords
user
group
commodity
users
vector
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
Application number
CN201911394011.1A
Other languages
Chinese (zh)
Other versions
CN111199459A (en
Inventor
赵豫陕
张军涛
肖淋峰
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.)
Shenzhen Mengtian Technology Co ltd
Original Assignee
Shenzhen Mengtian Technology Co 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 Shenzhen Mengtian Technology Co ltd filed Critical Shenzhen Mengtian Technology Co ltd
Priority to CN201911394011.1A priority Critical patent/CN111199459B/en
Publication of CN111199459A publication Critical patent/CN111199459A/en
Application granted granted Critical
Publication of CN111199459B publication Critical patent/CN111199459B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q30/00—Commerce
    • G06Q30/06—Buying, selling or leasing transactions
    • G06Q30/0601—Electronic shopping [e-shopping]
    • G06Q30/0631—Item recommendations
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
    • G06F16/36—Creation of semantic tools, e.g. ontology or thesauri
    • G06F16/367—Ontology
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G06N3/045—Combinations of networks
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/08—Learning methods
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Business, Economics & Management (AREA)
  • General Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • Data Mining & Analysis (AREA)
  • Economics (AREA)
  • Computational Linguistics (AREA)
  • Strategic Management (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Health & Medical Sciences (AREA)
  • General Business, Economics & Management (AREA)
  • Human Resources & Organizations (AREA)
  • Software Systems (AREA)
  • Mathematical Physics (AREA)
  • Computing Systems (AREA)
  • Molecular Biology (AREA)
  • Biophysics (AREA)
  • General Health & Medical Sciences (AREA)
  • Marketing (AREA)
  • Development Economics (AREA)
  • Evolutionary Computation (AREA)
  • Finance (AREA)
  • Accounting & Taxation (AREA)
  • Artificial Intelligence (AREA)
  • Biomedical Technology (AREA)
  • Tourism & Hospitality (AREA)
  • Databases & Information Systems (AREA)
  • Animal Behavior & Ethology (AREA)
  • Game Theory and Decision Science (AREA)
  • Quality & Reliability (AREA)
  • Operations Research (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

The embodiment of the disclosure discloses a commodity recommendation method, a commodity recommendation device, electronic equipment and a storage medium, wherein the commodity recommendation method comprises the following steps: acquiring a knowledge graph according to the association relation between users and the interaction record of the users and commodities; grouping the plurality of users based on the knowledge graph to obtain a plurality of user groups; inputting the vector of any user group and the vector of any group promotion commodity into a pre-trained commodity purchase prediction model to obtain the probability of the user group purchasing the group promotion commodity; determining a strategy for recommending the group promotion commodities to the user groups according to the probability that the user groups purchase the group promotion commodities, and recommending the group promotion commodities to the user groups according to the strategy. According to the technical scheme, the proper piecing promotion commodity can be recommended to each user group, personalized recommendation can be achieved, and the piecing success rate of the piecing promotion commodity can be improved.

Description

Commodity recommendation method, commodity recommendation device, electronic equipment and storage medium
Technical Field
The embodiment of the disclosure relates to the technical field of computer application, in particular to a commodity recommendation method, a commodity recommendation device, electronic equipment and a storage medium.
Background
In order to reduce commodity inventory and enhance user viscosity, current e-commerce platforms generally adopt a group promotion mode to uniformly display to users, namely: after the commodity to be promoted by the group is extracted, the commodity to be promoted by the group is uniformly displayed to the user in the appointed module, and all the commodity promoted by the group seen by the user are identical.
The electronic commerce adopts the sales mode of group promotion, so that the commodity inventory is reduced, the activity of the user is improved, and the lasting development of the platform is promoted. However, the current group promotion mode of each platform has a large defect, specifically, when facing different users, the group promotion commodities displayed to all users are the same, and the individuation of the group promotion commodities is not achieved, so that the conversion rate of the group promotion commodities is not high enough.
Disclosure of Invention
In view of this, the embodiments of the present disclosure provide a method, an apparatus, an electronic device, and a storage medium for recommending goods, so as to implement personalized recommendation and improve the success rate of grouping for promoting goods by grouping.
Other features and advantages of embodiments of the present disclosure will be apparent from the following detailed description, or may be learned by practice of embodiments of the disclosure in part.
In a first aspect, an embodiment of the present disclosure provides a commodity recommendation method, including:
acquiring a knowledge graph according to the association relationship between users and the interaction record of the users and commodities, wherein the knowledge graph comprises a plurality of users and the association relationship between the users;
grouping the plurality of users based on the knowledge graph to obtain a plurality of user groups;
inputting the vector of any user group and the vector of any group promotion commodity into a pre-trained commodity purchase prediction model to obtain the probability of the user group purchasing the group promotion commodity;
determining a strategy for recommending the group promotion commodities to the user groups according to the probability that the user groups purchase the group promotion commodities, and recommending the group promotion commodities to the user groups according to the strategy.
In an embodiment, obtaining the knowledge graph according to the association relationship between the users and the interaction record of the users and the commodity includes:
generating a first knowledge graph according to the association relation between users and the interaction record of the users and the commodities, wherein the first knowledge graph comprises a plurality of commodities, a plurality of users and the association relation between the commodities and the users;
and generating the knowledge graph according to the relation among the users in the first knowledge graph.
In an embodiment, grouping the plurality of users based on the knowledge-graph to obtain a plurality of user groups includes:
randomly selecting a first number of users from the knowledge graph as a central user group;
for any central user in the central user group, taking the central user and the user associated with the central user as a user group;
if any user does not belong to any user group, the user is added into the central user group, if any user belongs to a plurality of user groups, the user is randomly added into any user group, and the user is removed from other user groups.
In one embodiment, determining a policy for recommending the plurality of group promotion items to the plurality of user groups according to a probability that each group of users purchased each group promotion item, and recommending the plurality of group promotion items to the plurality of user groups according to the policy comprises:
and selecting the group promotion commodity with highest probability from the group promotion commodities for any user group, and recommending the group promotion commodity to all users of the user group.
In one embodiment, the vector of any user group is an average vector of vectors of users included in the user group.
In one embodiment, the interaction records include one or more of purchase records, collection records, click records, evaluation records, and return records;
the association between the users includes one or more of an invitation relationship, and a geographic relationship, wherein the geographic relationship includes native and/or residential land.
In one embodiment, the commodity purchase prediction model is trained by the following steps:
acquiring a training sample set, wherein the training sample comprises a user group vector, a commodity vector and a label for representing the probability of purchasing the commodity by the user group;
determining an initialized merchandise purchase prediction model, wherein the initialized merchandise purchase prediction model includes a target layer for outputting a probability that a user group purchases a merchandise;
and using a machine learning method, taking the user group vector and the commodity vector in the training samples in the training sample set as the input of an initialized commodity purchase prediction model, taking the labels corresponding to the input user group vector and the commodity vector as the output of the initialized commodity purchase prediction model, and training to obtain the commodity purchase prediction model.
In one embodiment, the initialized merchandise purchase prediction model includes an input layer, a first hidden layer, a second hidden layer, a softmax layer, and an output layer, the first hidden layer and the second hidden layer employing a ReLU function as an activation function.
In a second aspect, an embodiment of the present disclosure further provides a commodity recommendation apparatus, including:
the system comprises a knowledge graph acquisition unit, a commodity analysis unit and a commodity analysis unit, wherein the knowledge graph acquisition unit is used for acquiring a knowledge graph according to the association relationship among users and the interaction record of the users and the commodity, and the knowledge graph comprises a plurality of users and the association relationship among the users;
the user grouping unit is used for grouping the plurality of users based on the knowledge graph to obtain a plurality of user groups;
the probability prediction unit is used for inputting the vector of any user group and the vector of any group promotion commodity into a pre-trained commodity purchase prediction model to obtain the probability of the user group buying the group promotion commodity;
the strategy determining and commodity recommending unit is used for determining a strategy for recommending the plurality of the group promotion commodities to the plurality of the user groups according to the probability that each group promotion commodity is purchased by each user group, and recommending the plurality of the group promotion commodities to the plurality of the user groups according to the strategy.
In an embodiment, the knowledge-graph obtaining unit is configured to:
generating a first knowledge graph according to the association relation between users and the interaction record of the users and the commodities, wherein the first knowledge graph comprises a plurality of commodities, a plurality of users and the association relation between the commodities and the users;
and generating the knowledge graph according to the relation among the users in the first knowledge graph.
In an embodiment, the user grouping unit is configured to:
randomly selecting a first number of users from the knowledge graph as a central user group;
for any central user in the central user group, taking the central user and the user associated with the central user as a user group;
if any user does not belong to any user group, the user is added into the central user group, if any user belongs to a plurality of user groups, the user is randomly added into any user group, and the user is removed from other user groups.
In an embodiment, the policy determination and merchandise recommendation unit is configured to:
and selecting the group promotion commodity with highest probability from the group promotion commodities for any user group, and recommending the group promotion commodity to all users of the user group.
In one embodiment, the vector of any user group is an average vector of vectors of users included in the user group.
In one embodiment, the interaction records include one or more of purchase records, collection records, click records, evaluation records, and return records;
the association between the users includes one or more of an invitation relationship, and a geographic relationship, wherein the geographic relationship includes native and/or residential land.
In one embodiment, the commodity purchase prediction model is trained by the following modules:
the system comprises a sample acquisition module, a sample analysis module and a sample analysis module, wherein the sample acquisition module is used for acquiring a training sample set, and the training sample comprises a user group vector, a commodity vector and a label used for representing the probability of purchasing the commodity by the user group;
a model determination module for determining an initialized merchandise purchase prediction model, wherein the initialized merchandise purchase prediction model includes a target layer for outputting a probability of a user group purchasing a merchandise;
and the model training module is used for utilizing a machine learning device, taking the user group vector and the commodity vector in the training sample set as the input of the initialized commodity purchase prediction model, taking the labels corresponding to the input user group vector and the commodity vector as the output of the initialized commodity purchase prediction model, and training to obtain the commodity purchase prediction model.
In one embodiment, the initialized merchandise purchase prediction model includes an input layer, a first hidden layer, a second hidden layer, a softmax layer, and an output layer, the first hidden layer and the second hidden layer employing a ReLU function as an activation function.
In a third aspect, an embodiment of the present disclosure further provides an electronic device, including:
one or more processors;
a memory for storing one or more programs;
the one or more programs, when executed by the one or more processors, cause the one or more processors to implement the instructions of the method of any of the first aspects.
In a fourth aspect, the presently disclosed embodiments also provide a computer readable storage medium having stored thereon a computer program which, when executed by a processor, implements the steps of the method according to any of the first aspects.
According to the embodiment of the disclosure, a knowledge graph is obtained according to the association relation among users and the interaction record of the users and the commodities, a plurality of user groups are obtained by grouping the users based on the knowledge graph, any user group vector and any group promotion commodity vector are input into a pre-trained commodity purchase prediction model to obtain the probability of purchasing the group promotion commodity, a strategy of recommending the group promotion commodity to the plurality of user groups is determined according to the probability of purchasing the group promotion commodity by each user group, and the group promotion commodity is recommended to the plurality of user groups according to the strategy, so that the personalized recommendation can be realized and the group promotion success rate of the group promotion commodity can be improved.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present disclosure, the following description will briefly explain the drawings required to be used in the description of the embodiments of the present disclosure, and it is apparent that the drawings in the following description are only some of the embodiments of the present disclosure, and other drawings may be obtained according to the contents of the embodiments of the present disclosure and these drawings without inventive effort for those skilled in the art.
Fig. 1 is a schematic flow chart of a commodity recommendation method according to an embodiment of the present disclosure;
FIG. 2 is a flow chart of a training method of a commodity purchase prediction model provided by an embodiment of the present disclosure;
FIG. 3 is a schematic flow chart of another method for recommending commodities according to an embodiment of the present disclosure;
FIG. 4 is a partial content of a first knowledge-graph example provided by an embodiment of the present disclosure;
FIG. 5 is an example of a knowledge-graph generated according to FIG. 4, provided by an embodiment of the present disclosure;
fig. 6 is a schematic structural diagram of a commodity recommendation device according to an embodiment of the present disclosure;
FIG. 7 is a schematic diagram of a training module of a merchandise purchase prediction model provided by an embodiment of the present disclosure;
fig. 8 shows a schematic structural diagram of an electronic device suitable for use in implementing embodiments of the present disclosure.
Detailed Description
In order to make the technical problems solved, the technical solutions adopted and the technical effects achieved by the embodiments of the present disclosure more clear, the technical solutions of the embodiments of the present disclosure will be described in further detail below with reference to the accompanying drawings, and it is apparent that the described embodiments are only some embodiments, but not all embodiments of the present disclosure. All other embodiments, which are derived by a person skilled in the art from the embodiments of the present disclosure without creative efforts, fall within the protection scope of the embodiments of the present disclosure.
It should be noted that the terms "system" and "network" in the embodiments of the present disclosure are often used interchangeably herein. References to "and/or" in the embodiments of the present disclosure are intended to encompass any and all combinations of one or more of the associated listed items. The terms first, second and the like in the description and in the claims and drawings are used for distinguishing between different objects and not for limiting a particular order.
It should be further noted that, in the embodiments of the present disclosure, the following embodiments may be implemented separately, or may be implemented in combination with each other, which is not specifically limited by the embodiments of the present disclosure.
The names of messages or information interacted between the various devices in the embodiments of the present disclosure are for illustrative purposes only and are not intended to limit the scope of such messages or information.
The technical solutions of the embodiments of the present disclosure are further described below with reference to the accompanying drawings and through specific implementations.
Fig. 1 is a schematic flow chart of a commodity recommendation method provided in an embodiment of the present disclosure, where the embodiment is applicable to a case of recommending a group of promoted commodities to a user group, and the method may be performed by a commodity recommendation device configured in an electronic device, as shown in fig. 1, where the commodity recommendation method includes:
in step S110, a knowledge graph is obtained according to the association relationship between users and the interaction record of the users and the commodity, wherein the knowledge graph includes a plurality of users and the association relationship between the plurality of users.
Wherein the interaction records include one or more of a plurality of purchase records, collection records, click records, evaluation records, and return records, for example.
Wherein the association relationship between the users includes one or more of various types, such as an invitation relationship, and a geographical (e.g., home, residence, etc.) relationship.
The step can be implemented by adopting various methods, for example, a first knowledge graph can be generated according to the association relationship between users and the interaction record of the users and the commodities, wherein the first knowledge graph comprises a plurality of commodities, a plurality of users and the association relationship between the commodities and the users. And generating the knowledge graph according to the relation between the users in the first knowledge graph.
In step S120, the plurality of users are grouped based on the knowledge graph to obtain a plurality of user groups.
For example, a first number of users may be randomly selected from the knowledge-graph as a central user group; and regarding any central user in the central user group, taking the central user and the user associated with the central user as a user group. For the grouping, if any user does not belong to any user group, the user can be added into the central user group, if any user belongs to a plurality of user groups, the user can be randomly added into any affiliated user group, and the user is removed from other affiliated user groups.
In step S130, the vector of any user group and the vector of any group-promoted commodity are input to a pre-trained commodity purchase prediction model, so as to obtain the probability of the user group purchasing the group-promoted commodity.
The commodity purchase prediction model can determine the probability value of the vector of the input group promotion commodity purchased by the user group corresponding to the input user/user group vector, and can be obtained through training in various modes, so long as the commodity purchase prediction model has the functions, and the specific training mode is not limited in this embodiment.
For example, fig. 2 is a flow chart of a training method of a commodity purchase prediction model according to an embodiment of the present disclosure, where the commodity purchase prediction model may be obtained by training through the steps shown in fig. 2:
in step S210, a training sample set is obtained, where the training sample includes a user group vector, a vector of merchandise, and a label for representing a probability that the user group purchases the merchandise.
The user group vector may be a vector corresponding to a user, or may be a vector corresponding to a user group.
For example, the group promotion commodity clicked by all users in each user group is defined as a positive sample of the user group, and the positive sample is marked as 1; and randomly selecting a part of the group-spliced promoted commodities which are not clicked by each user group as a negative sample of the user group, and marking the negative sample as 0.
In step S220, an initialized merchandise purchase prediction model is determined, wherein the initialized merchandise purchase prediction model includes a target tier for outputting a probability that a user group purchases a merchandise.
In step S230, the user group vector and the commodity vector in the training sample set are used as input of the initialized commodity purchase prediction model, and the label corresponding to the input user group vector and the commodity vector is used as output of the initialized commodity purchase prediction model, and the commodity purchase prediction model is obtained through training.
The initialized merchandise purchase prediction model may include, for example, an input layer, a first hidden layer, a second hidden layer, a softmax layer, and an output layer, where the first hidden layer and the second hidden layer use a ReLU function as an activation function.
In step S140, a policy for recommending the plurality of group promotion products to the plurality of user groups is determined according to the probability that each group promotion product is purchased by each user group, and the plurality of group promotion products are recommended to the plurality of user groups according to the policy.
For example, for any user group, the group promotion commodity with the highest probability can be selected from the group promotion commodities, and recommended to all users of the user group.
The vector of the user included in any user group can be averaged to be used as the user group vector.
Acquiring a knowledge graph according to the association relationship between users and the interaction record of the users and commodities, wherein the knowledge graph comprises a plurality of users and the association relationship between the users;
grouping the plurality of users based on the knowledge graph to obtain a plurality of user groups;
inputting the vector of any user group and the vector of any group promotion commodity into a pre-trained commodity purchase prediction model to obtain the probability of the user group purchasing the group promotion commodity;
determining a strategy for recommending the group promotion commodities to the user groups according to the probability that the user groups purchase the group promotion commodities, and recommending the group promotion commodities to the user groups according to the strategy.
According to the method, a knowledge graph is obtained according to the association relation among users and the interaction record of the users and commodities, a plurality of user groups are obtained by grouping the users based on the knowledge graph, any user group vector and any group promotion commodity vector are input into a pre-trained commodity purchase prediction model, the probability that the user group purchases the group promotion commodity is obtained, a strategy of recommending the plurality of group promotion commodities to the plurality of user groups is determined according to the probability that each user group purchases each group promotion commodity, and the plurality of group promotion commodities are recommended to the plurality of user groups according to the strategy, so that the suitable group promotion commodity can be recommended to each user group, personalized recommendation can be realized, and the group success rate of the group promotion commodity can be improved.
Fig. 3 is a schematic flow chart of another commodity recommendation method according to an embodiment of the present disclosure, where improvement optimization is performed based on the foregoing embodiment. As shown in fig. 3, the commodity recommendation method according to the present embodiment includes:
in step S310, a first knowledge graph is generated according to the association relationship between the users and the interaction record of the users and the commodities, wherein the first knowledge graph includes a plurality of commodities, a plurality of users, and the association relationship between the commodities and the users.
The first knowledge-graph may be represented by a single triplet (h, r, t), e.g. (Zhang three, age, 21), (Lifour, clicked, commodity 1). Some basic data are needed for constructing the first knowledge graph, basic characteristics of a user can be obtained by basic information of the user, interaction records of the user and commodities can be obtained by user logs, and basic characteristics of the commodities can be obtained by basic information of the commodities.
And constructing a first knowledge graph according to the triples, wherein a simplified part of the knowledge graph is shown in fig. 4. Fig. 4 may be represented by the following triplets: (user 1, native place, guangdong), (user 1, clicked, article 1), (user 1, purchased, article 2), (article 2, clicked, user 4), (Guangdong, are the usual places of the user, user 3), (user 3, invitation registration, user 2).
In step S320, the knowledge graph is generated according to the relationship between the users in the first knowledge graph.
After the first knowledge graph is constructed, the knowledge graph can be reconstructed, namely only the relationship between users is reserved. Taking fig. 4 as an example, the knowledge-graph obtained after the reconstruction according to the first knowledge-graph shown in fig. 4 is shown in fig. 5. Taking the user 1 as a starting point, the reconstruction method takes the user 1 as an example, the user 1 can walk to the user 3, the user 4 and the user 2 along the arrow, but the user 1 and the user 3 can be understood to be in a one-layer relationship without passing other users in the middle of the walk to the user 3 and the user 4, the user 2 is in a two-layer relationship with the user 1 when the user 2 needs to pass through the user 3, and the like, and the users with the one-layer relationship can be connected by the same method.
In step S330, the plurality of users are grouped based on the knowledge graph to obtain a plurality of user groups.
After the knowledge graph of the user dimension is constructed in step S320, each user may be numbered sequentially, for example, about 500 ten thousand users are numbered 1 to 500 ten thousand.
After numbering the users, 1 ten thousand random numbers are selected from 1 to 500 ten thousand, i.e., 1 ten thousand users are randomly selected. The 1-ten thousand users are taken as the center point, and the users which are in one-layer relation with the users are found to form 1-ten thousand user groups. Taking fig. 3 as an example, if the user 1 is selected as the center point, the user 1, the user 3, and the user 4 constitute one user group.
Because the center point is randomly selected, it may be that some users are not programmed to a user group or are programmed to multiple user groups. Aiming at the first situation, the user who is selected as the center point can find the group joining closest to the user, and if a plurality of closest groups exist, one group joining is selected randomly; for the second case, one group may be randomly reserved for the user, with the remaining groups removing the user.
Through the operation, the grouping of the users is completed, and meanwhile, each user is guaranteed to have only one corresponding group. A vector is randomly initialized for each group, and the corresponding group is represented by a vector.
In step S340, the vector of any user group and the vector of any group-promoted commodity are input to a pre-trained commodity purchase prediction model, so as to obtain the probability of the user group purchasing the group-promoted commodity.
The steps relate to the following:
(1) And (3) building a commodity purchase prediction model:
for example, a vector may be randomly initialized for each collage promotional item.
(2) Training positive and negative samples of commodity purchase prediction model and constructing:
for example, a positive sample for each user may be defined as: the group promotion commodity clicked by all users in each user group is defined as a positive sample of the user group, and the positive sample of the user group is the positive sample of each user in the group;
for each user's negative sample, a portion of the group promotion items not clicked by each user group may be randomly selected as the negative sample of the user group, for example, the number of negative samples may be controlled to have a ratio to the number of positive samples of 3: 1.
(3) And (3) building a commodity purchase prediction model:
for example, the initialized commodity purchase prediction model may be configured as a neural network model having a two-layer hidden layer neural network and one softmax layer. The activation function of each hidden neural network can be set as a ReLU function, the input of the model is the vector of the user (the vector of the user is the vector of the user group where the user is) and the vector of the collage promotion commodity, the output of the model is a value between 0 and 1, the probability that the user clicks the commodity can be understood, the larger the value is the more likely the user clicks the commodity, the threshold value is set to be 0.5, namely when the output value of the model is more than or equal to 0.5, the user can be considered to click the commodity, and when the output value of the model is less than or equal to 0.5, the commodity can be considered not to be clicked.
In step S350, for any user group, a group promotion product with the highest probability is selected from the group promotion products, and is recommended to all users of the user group.
After the model is built and trained, when the user is predicted, only the vector of the user and the vector of the group promotion commodity are input into the model, and the commodity with the highest score is selected to be recommended to the user. Because the vectors of the users within each user group are the same, the recommended items seen by the users within each user group are the same and are of most interest to them.
According to the embodiment, the knowledge graph is utilized to group the users, different group promotion commodities are respectively recommended to each user group, personalized recommendation can be achieved, and the group success rate of the group promotion commodities can be improved.
As an implementation of the method shown in the foregoing figures, an embodiment of a commodity recommendation apparatus is provided, and fig. 6 shows a schematic structural diagram of the commodity recommendation apparatus provided in this embodiment, where the embodiment of the apparatus corresponds to the embodiment of the method shown in fig. 1 to 5, and the apparatus may be specifically applied to various electronic devices. As shown in fig. 6, the commodity recommendation apparatus according to the present embodiment includes a knowledge-graph acquisition unit 610, a user grouping unit 620, a probability prediction unit 630, and a policy determination and commodity recommendation unit 640.
The knowledge-graph obtaining unit 610 is configured to obtain a knowledge graph according to an association relationship between users and an interaction record of the users with a commodity, where the knowledge graph includes a plurality of users and the association relationship between the plurality of users.
The user grouping unit 620 is configured to group the plurality of users based on the knowledge graph to obtain a plurality of user groups.
The probability prediction unit 630 is configured to input the vector of any user group and the vector of any group promotion commodity into a pre-trained commodity purchase prediction model to obtain the probability of the user group purchasing the group promotion commodity.
The policy determining and commodity recommending unit 640 is configured to determine a policy for recommending the plurality of group-promoted commodities to the plurality of user groups according to a probability that each group-promoted commodity is purchased by each user group, and recommend the plurality of group-promoted commodities to the plurality of user groups according to the policy.
In an embodiment, the knowledge-graph obtaining unit 610 is configured to further: generating a first knowledge graph according to the association relation between users and the interaction record of the users and the commodities, wherein the first knowledge graph comprises a plurality of commodities, a plurality of users and the association relation between the commodities and the users; and generating the knowledge graph according to the relation among the users in the first knowledge graph.
In an embodiment, the user grouping unit 620 is configured to further:
randomly selecting a first number of users from the knowledge graph as a central user group;
for any central user in the central user group, taking the central user and the user associated with the central user as a user group;
if any user does not belong to any user group, the user is added into the central user group, if any user belongs to a plurality of user groups, the user is randomly added into any user group, and the user is removed from other user groups.
In an embodiment, the policy determination and merchandise recommendation unit 640 is configured to further:
and selecting the group promotion commodity with highest probability from the group promotion commodities for any user group, and recommending the group promotion commodity to all users of the user group.
In one embodiment, the vector of any user group is an average vector of vectors of users included in the user group.
In one embodiment, the interaction records include one or more of purchase records, collection records, click records, evaluation records, and return records.
In one embodiment, the association between the users includes one or more of an invitation relationship and a geographical relationship, wherein the geographical relationship includes native and/or residential places.
Fig. 7 is a schematic structural diagram of a training module of a commodity purchase prediction model provided in this embodiment, and as shown in fig. 7, the commodity purchase prediction model is obtained by training by a sample obtaining module 710, a model determining module 720 and a model training module 730:
the sample acquisition module 710 is configured to acquire a set of training samples, wherein the training samples include a user group vector, a vector of merchandise, and a label representing a probability that the user group purchased the merchandise.
The model determination module 720 is configured to determine an initialized merchandise purchase prediction model, wherein the initialized merchandise purchase prediction model includes a target tier for outputting a probability that a user group purchases a merchandise.
The model training module 730 is configured to use machine learning means to use a user group vector and a commodity vector in a training sample in the training sample set as input of an initialized commodity purchase prediction model, and use labels corresponding to the input user group vector and commodity vector as output of the initialized commodity purchase prediction model, and train to obtain the commodity purchase prediction model.
Further, the initialized commodity purchase prediction model includes an input layer, a first hidden layer, a second hidden layer, a softmax layer, and an output layer, the first hidden layer and the second hidden layer employing a ReLU function as an activation function.
The commodity recommendation device provided by the embodiment can execute the commodity recommendation method provided by the embodiment of the method, and has the corresponding functional modules and beneficial effects of the execution method.
Referring now to fig. 8, a schematic diagram of an electronic device 800 suitable for use in implementing embodiments of the present disclosure is shown. The terminal devices in the embodiments of the present disclosure may include, but are not limited to, mobile terminals such as mobile phones, notebook computers, digital broadcast receivers, PDAs (personal digital assistants), PADs (tablet computers), PMPs (portable multimedia players), in-vehicle terminals (e.g., in-vehicle navigation terminals), and the like, and stationary terminals such as digital TVs, desktop computers, and the like. The electronic device shown in fig. 8 is merely an example and should not be construed to limit the functionality and scope of use of the disclosed embodiments.
As shown in fig. 8, the electronic device 800 may include a processing means (e.g., a central processor, a graphics processor, etc.) 801, which may perform various appropriate actions and processes according to a program stored in a Read Only Memory (ROM) 802 or a program loaded from a storage means 808 into a Random Access Memory (RAM) 803. In the RAM 803, various programs and data required for the operation of the electronic device 800 are also stored. The processing device 801, the ROM 802, and the RAM 803 are connected to each other by a bus 804. An input/output (I/O) interface 805 is also connected to the bus 804.
In general, the following devices may be connected to the I/O interface 805: input devices 806 including, for example, a touch screen, touchpad, keyboard, mouse, camera, microphone, accelerometer, gyroscope, and the like; an output device 807 including, for example, a Liquid Crystal Display (LCD), speakers, vibrators, etc.; storage 808 including, for example, magnetic tape, hard disk, etc.; communication means 809. The communication means 809 may allow the electronic device 800 to communicate wirelessly or by wire with other devices to exchange data. While fig. 8 shows an electronic device 800 having various means, it is to be understood that not all of the illustrated means are required to be implemented or provided. More or fewer devices may be implemented or provided instead.
In particular, according to embodiments of the present disclosure, the processes described above with reference to flowcharts may be implemented as computer software programs. For example, embodiments of the present disclosure include a computer program product comprising a computer program embodied on a computer readable medium, the computer program comprising program code for performing the method shown in the flow chart. In such an embodiment, the computer program may be downloaded and installed from a network via communication device 809, or installed from storage device 808, or installed from ROM 802. The above-described functions defined in the methods of the embodiments of the present disclosure are performed when the computer program is executed by the processing device 801.
It should be noted that, the computer readable medium described above in the embodiments of the present disclosure may be a computer readable signal medium or a computer readable storage medium, or any combination of the two. The computer readable storage medium can be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or a combination of any of the foregoing. More specific examples of the computer-readable storage medium may include, but are not limited to: 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 or flash memory), 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. In the disclosed embodiments, 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. In contrast, in the disclosed embodiments, the computer-readable signal medium may comprise a data signal propagated in baseband or as part of a carrier wave, with the computer-readable program code embodied therein. Such a propagated data signal may take any of a variety of forms, including, but not limited to, electro-magnetic, optical, or any suitable combination of the foregoing. A computer readable signal medium may also be any computer readable medium that is not a computer readable storage medium and that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device. Program code embodied on a computer readable medium may be transmitted using any appropriate medium, including but not limited to: electrical wires, fiber optic cables, RF (radio frequency), and the like, or any suitable combination of the foregoing.
The computer readable medium may be contained in the electronic device; or may exist alone without being incorporated into the electronic device.
The computer readable medium carries one or more programs which, when executed by the electronic device, cause the electronic device to:
acquiring a knowledge graph according to the association relationship between users and the interaction record of the users and commodities, wherein the knowledge graph comprises a plurality of users and the association relationship between the users;
grouping the plurality of users based on the knowledge graph to obtain a plurality of user groups;
inputting the vector of any user group and the vector of any group promotion commodity into a pre-trained commodity purchase prediction model to obtain the probability of the user group purchasing the group promotion commodity;
determining a strategy for recommending the group promotion commodities to the user groups according to the probability that the user groups purchase the group promotion commodities, and recommending the group promotion commodities to the user groups according to the strategy.
Computer program code for carrying out operations for embodiments of the present disclosure may be written in one or more programming languages, including an object oriented programming language such as Java, smalltalk, C ++ and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any kind of network, including a Local Area Network (LAN) or a Wide Area Network (WAN), or may be connected to an external computer (for example, through the Internet using an Internet service provider).
The flowcharts and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to various embodiments of the present disclosure. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that, in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.
The units involved in the embodiments of the present disclosure may be implemented by means of software, or may be implemented by means of hardware. The name of the unit does not in any way constitute a limitation of the unit itself, for example the first acquisition unit may also be described as "unit acquiring at least two internet protocol addresses".
The foregoing description is only of the preferred embodiments of the disclosed embodiments and is presented for purposes of illustration of the principles of the technology being utilized. It will be appreciated by those skilled in the art that the scope of the disclosure in the embodiments of the disclosure is not limited to the specific combination of the above technical features, but also encompasses other technical features formed by any combination of the above technical features or their equivalents without departing from the spirit of the disclosure. Such as the technical solution formed by mutually replacing the above-mentioned features and the technical features with similar functions (but not limited to) disclosed in the embodiments of the present disclosure.

Claims (10)

1. A commodity recommendation method, comprising:
acquiring a knowledge graph according to the association relationship between users and the interaction record of the users and commodities, wherein the knowledge graph comprises a plurality of users and the association relationship between the users;
randomly selecting a first number of users from the knowledge graph as a central user group;
for any central user in the central user group, taking the central user and the user associated with the central user as a user group; if any user does not belong to any user group, adding the user into the central user group, if any user belongs to a plurality of user groups, randomly adding the user into any user group, and removing the user from other user groups;
inputting the vector of any user group and the vector of any group promotion commodity into a pre-trained commodity purchase prediction model to obtain the probability of the user group purchasing the group promotion commodity; wherein, the vector of any user group and the vector of any group promotion commodity are obtained by random initialization;
determining a strategy for recommending the group promotion commodities to the user groups according to the probability that the user groups purchase the group promotion commodities, and recommending the group promotion commodities to the user groups according to the strategy: the group promotion merchandise recommended by the plurality of users of the same user group is the same.
2. The method of claim 1, wherein obtaining a knowledge-graph based on the association between users and the record of interactions between users and merchandise comprises:
generating a first knowledge graph according to the association relation between users and the interaction record of the users and the commodities, wherein the first knowledge graph comprises a plurality of commodities, a plurality of users and the association relation between the commodities and the users;
and generating the knowledge graph according to the relation among the users in the first knowledge graph.
3. The method of claim 1, wherein determining a policy for recommending the plurality of patch promoted items to the plurality of user groups based on a probability that each patch promoted item was purchased by each user group, and recommending the plurality of patch promoted items to the plurality of user groups based on the policy comprises:
and selecting the group promotion commodity with highest probability from the group promotion commodities for any user group, and recommending the group promotion commodity to all users of the user group.
4. The method of claim 1 wherein the any one of the user group vectors is an average vector of vectors of users included in the user group.
5. The method of claim 1, wherein the interaction records include one or more of purchase records, collection records, click records, evaluation records, and return records;
the association between the users includes one or more of an invitation relationship, and a geographic relationship, wherein the geographic relationship includes native and/or residential land.
6. The method according to any one of claims 1-5, wherein the commodity purchase prediction model is trained by:
acquiring a training sample set, wherein the training sample comprises a user group vector, a commodity vector and a label for representing the probability of purchasing the commodity by the user group;
determining an initialized merchandise purchase prediction model, wherein the initialized merchandise purchase prediction model includes a target layer for outputting a probability that a user group purchases a merchandise;
and using a machine learning method, taking the user group vector and the commodity vector in the training samples in the training sample set as the input of an initialized commodity purchase prediction model, taking the labels corresponding to the input user group vector and the commodity vector as the output of the initialized commodity purchase prediction model, and training to obtain the commodity purchase prediction model.
7. The method of claim 6, wherein the initialized commodity purchase prediction model comprises an input layer, a first hidden layer, a second hidden layer, a softmax layer, and an output layer, the first hidden layer and the second hidden layer employing a ReLU function as an activation function.
8. A commodity recommendation device, comprising:
the system comprises a knowledge graph acquisition unit, a commodity analysis unit and a commodity analysis unit, wherein the knowledge graph acquisition unit is used for acquiring a knowledge graph according to the association relationship among users and the interaction record of the users and the commodity, and the knowledge graph comprises a plurality of users and the association relationship among the users;
the user grouping unit is used for randomly selecting a first number of users from the knowledge graph as a central user group; for any central user in the central user group, taking the central user and the user associated with the central user as a user group; if any user does not belong to any user group, adding the user into the central user group, if any user belongs to a plurality of user groups, randomly adding the user into any user group, and removing the user from other user groups;
the probability prediction unit is used for inputting the vector of any user group and the vector of any group promotion commodity into a pre-trained commodity purchase prediction model to obtain the probability of the user group buying the group promotion commodity; wherein, the vector of any user group and the vector of any group promotion commodity are obtained by random initialization;
the strategy determining and commodity recommending unit is used for determining a strategy for recommending the plurality of group promotion commodities to the plurality of user groups according to the probability that each group promotion commodity is purchased by each user group, and recommending the plurality of group promotion commodities to the plurality of user groups according to the strategy: the group promotion merchandise recommended by the plurality of users of the same user group is the same.
9. An electronic device, comprising:
one or more processors;
a memory for storing one or more programs;
the instructions that when executed by the one or more processors cause the one or more processors to implement the method of any of claims 1-7.
10. A computer readable storage medium, on which a computer program is stored, characterized in that the computer program, when being executed by a processor, implements the steps of the method according to any of claims 1-7.
CN201911394011.1A 2019-12-30 2019-12-30 Commodity recommendation method, commodity recommendation device, electronic equipment and storage medium Active CN111199459B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201911394011.1A CN111199459B (en) 2019-12-30 2019-12-30 Commodity recommendation method, commodity recommendation device, electronic equipment and storage medium

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201911394011.1A CN111199459B (en) 2019-12-30 2019-12-30 Commodity recommendation method, commodity recommendation device, electronic equipment and storage medium

Publications (2)

Publication Number Publication Date
CN111199459A CN111199459A (en) 2020-05-26
CN111199459B true CN111199459B (en) 2024-02-09

Family

ID=70744478

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201911394011.1A Active CN111199459B (en) 2019-12-30 2019-12-30 Commodity recommendation method, commodity recommendation device, electronic equipment and storage medium

Country Status (1)

Country Link
CN (1) CN111199459B (en)

Families Citing this family (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111768231B (en) * 2020-06-24 2024-04-09 中国工商银行股份有限公司 Product information recommendation method and device
CN111768239A (en) * 2020-06-29 2020-10-13 腾讯科技（深圳）有限公司 Property recommendation method, device, system, server and storage medium
CN112261668B (en) * 2020-10-20 2022-07-19 北京邮电大学 Content caching method and device in mobile edge network and electronic equipment
CN112418452A (en) * 2020-11-19 2021-02-26 杭州闪援车管家汽车服务有限公司 Internet-based vehicle maintenance order combining method, server and system
CN112465594A (en) * 2020-11-27 2021-03-09 康键信息技术（深圳）有限公司 Live broadcast interaction method, device, equipment and storage medium
CN112330373A (en) * 2020-11-30 2021-02-05 中国银联股份有限公司 User behavior analysis method and device and computer readable storage medium
CN112395429A (en) * 2020-12-02 2021-02-23 上海三稻智能科技有限公司 Method, system and storage medium for determining, pushing and applying HS (high speed coding) codes based on graph neural network
CN112800207B (en) * 2021-01-13 2022-05-13 桂林电子科技大学 Commodity information recommendation method and device and storage medium
CN112800109A (en) * 2021-01-21 2021-05-14 蜜兔(杭州)网络科技有限公司 Information mining method and system
CN112419016A (en) * 2021-01-25 2021-02-26 南京石匠网络科技有限公司 Big data-based grouping method and system
CN113362131B (en) * 2021-06-02 2022-09-13 合肥工业大学 Intelligent commodity recommendation method based on map model and integrating knowledge map and user interaction

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105321089A (en) * 2014-07-16 2016-02-10 苏宁云商集团股份有限公司 Method and system for e-commerce recommendation based on multi-algorithm fusion
CN106067129A (en) * 2016-05-25 2016-11-02 北京小米移动软件有限公司 The method and apparatus of pushed information
CN106372961A (en) * 2016-08-23 2017-02-01 北京小米移动软件有限公司 Commodity recommendation method and device
CN106649774A (en) * 2016-12-27 2017-05-10 北京百度网讯科技有限公司 Artificial intelligence-based object pushing method and apparatus
CN109903117A (en) * 2019-01-04 2019-06-18 苏宁易购集团股份有限公司 A kind of knowledge mapping processing method and processing device for commercial product recommending
CN110322318A (en) * 2019-06-18 2019-10-11 中国平安财产保险股份有限公司 A kind of customer grouping method, apparatus and computer storage medium

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
TWI629655B (en) * 2015-02-16 2018-07-11 陳奕舟 A method for imrlementing a mobile group buying system

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105321089A (en) * 2014-07-16 2016-02-10 苏宁云商集团股份有限公司 Method and system for e-commerce recommendation based on multi-algorithm fusion
CN106067129A (en) * 2016-05-25 2016-11-02 北京小米移动软件有限公司 The method and apparatus of pushed information
CN106372961A (en) * 2016-08-23 2017-02-01 北京小米移动软件有限公司 Commodity recommendation method and device
CN106649774A (en) * 2016-12-27 2017-05-10 北京百度网讯科技有限公司 Artificial intelligence-based object pushing method and apparatus
CN109903117A (en) * 2019-01-04 2019-06-18 苏宁易购集团股份有限公司 A kind of knowledge mapping processing method and processing device for commercial product recommending
CN110322318A (en) * 2019-06-18 2019-10-11 中国平安财产保险股份有限公司 A kind of customer grouping method, apparatus and computer storage medium

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
知识图谱与数据应用——智能推荐;周晶 等;《电信科学》;全文 *

Also Published As

Publication number Publication date
CN111199459A (en) 2020-05-26

Similar Documents

Publication Publication Date Title
CN111199459B (en) Commodity recommendation method, commodity recommendation device, electronic equipment and storage medium
CN111242748B (en) Method, apparatus, and storage medium for recommending items to a user
US11586772B2 (en) Method and device for displaying information
CN113592535B (en) Advertisement recommendation method and device, electronic equipment and storage medium
CN112395022B (en) Information display method, information display device, electronic equipment and computer readable storage medium
CN113610582A (en) Advertisement recommendation method and device, storage medium and electronic equipment
CN112749323A (en) Method and device for constructing user portrait
CN111125502B (en) Method and device for generating information
CN111177582B (en) Method, device, electronic equipment and storage medium for determining friend user
CN116109374A (en) Resource bit display method, device, electronic equipment and computer readable medium
CN111026973A (en) Commodity interest degree prediction method and device and electronic equipment
US20240078585A1 (en) Method and apparatus for sharing information
CN112860999B (en) Information recommendation method, device, equipment and storage medium
CN109684015B (en) Interface data loading method and device, electronic equipment and storage medium
CN116433310A (en) Commodity recommendation method and device, electronic equipment and storage medium
CN113435960A (en) Virtual article display method and device, electronic equipment and computer readable medium
CN112506596A (en) Object display method and device, electronic equipment and computer readable storage medium
CN111563797A (en) House source information processing method and device, readable medium and electronic equipment
CN111738789A (en) Article information pushing method, device, equipment and storage medium
CN113177174B (en) Feature construction method, content display method and related device
CN117807428A (en) Object characteristic value recognition model training method, object recognition method and device
CN117807313A (en) Information recommendation method, device, electronic equipment and storage medium
CN112214306A (en) Content display weight value calculation method and device, electronic equipment and computer readable storage medium
CN114819747A (en) Message traffic quality evaluation method, electronic device, medium, and program product
CN117725219A (en) Enterprise entity relationship prediction method and device, storage medium and electronic 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