CN117808564B

CN117808564B - User data recommendation method and device based on artificial intelligence

Info

Publication number: CN117808564B
Application number: CN202410231813.5A
Authority: CN
Inventors: 孙勇; 胡泽平; 万青
Original assignee: Shenzhen Tobo Software Co ltd
Current assignee: Shenzhen Tobo Software Co ltd
Priority date: 2024-03-01
Filing date: 2024-03-01
Publication date: 2024-05-28
Anticipated expiration: 2044-03-01
Also published as: CN117808564A

Abstract

The application provides a user data recommendation method and device based on artificial intelligence, wherein the method comprises the following steps: establishing commodity description information and corresponding relation between a commodity introduction chart and a commodity recommendation identifier and a commodity recommendation list of the commodity through an artificial intelligent model; determining a target recommendation identifier and a target commodity recommendation list corresponding to the target input commodity according to the target commodity input request through the corresponding relation, and associating the target recommendation identifier, the target commodity recommendation list and the target commodity ID; determining a corresponding first recommendation identification set and a first commodity recommendation list set according to the current commodity ID set; and determining the association condition among all current transaction commodities in the current transaction request according to the first recommendation identification set and the first commodity recommendation list set, and generating commodity recommendation data according to the association condition, the first recommendation identification set and the first commodity recommendation list set. And the accuracy of the generated commodity recommendation data is improved.

Description

User data recommendation method and device based on artificial intelligence

Technical Field

The application relates to the field of data processing, in particular to a user data recommendation method and device based on artificial intelligence.

Background

In the current network environment, in order to improve the viscosity and retention rate of users, data recommendation is the weight of each platform, while data recommendation on an e-commerce platform is often embodied in the aspect of commodity recommendation, in order to enable users to quickly find commodities of the heart instrument or improve the consumption probability of the users on the platform, the existing commodity recommendation generally generates related recommended commodities according to the preference, commodity browsing records, commodity purchasing records and the like preset by the users as reference data related to commodity recommendation of the users.

In the process of generating data of recommended goods, various recommendation algorithms are involved, and the types of the existing recommendation algorithms are generally classified into: the four algorithms based on the artificial intelligence model, collaborative filtering, content and association rules have respective advantages and disadvantages.

In addition to the influence of the algorithm, the input parameters are also an important condition affecting the accuracy of the data when the data of the recommended commodity is generated, most of the current recommendation algorithms adopt the key words set by the grabbing merchant or text information as the input parameters, and the same commodity is often caused due to the regional culture difference of each merchant and the diversity of commodity names, the commodity names in different merchants are different, and the description of the commodity is also different, so that error data is easily generated when commodity recommendation data is generated, the commodity data recommended to a user is inaccurate, and the user experience is affected.

Disclosure of Invention

In view of the foregoing, the present application has been developed to provide an artificial intelligence based user data recommendation method and apparatus that overcomes or at least partially solves the foregoing, including:

The method relates to a user terminal, a merchant terminal, a main server and a recommendation server; the merchant terminal is used for sending a commodity input request to the main server; the commodity input request comprises commodity description information corresponding to input commodities and a commodity introduction chart; the main server is used for generating commodity IDs corresponding to the input commodities one by one and sending the commodity IDs and the commodity input requests to the recommendation server; the user is used for sending a transaction request to the main server; the transaction request comprises transaction information of at least two different transaction commodities, wherein the transaction commodities are recorded and bound with commodity IDs; the main server is used for determining a commodity ID set according to the transaction request and sending the commodity ID set to the recommendation server; wherein the commodity ID set comprises commodity IDs of all trade commodities in the trade request;

The method comprises the following steps:

The recommendation server establishes commodity description information and corresponding relation between a commodity introduction chart, a commodity recommendation identifier and a commodity recommendation list of the commodity through an artificial intelligent model; the commodity recommendation list comprises five different recommendation identifiers which have purchase relevance with commodities;

When a target commodity input request and a target commodity ID, which are sent by a main server and correspond to target input commodities, are received, determining a target recommendation identifier and a target commodity recommendation list, which correspond to the target input commodities, according to the target commodity input request through the corresponding relation by the recommendation server, and associating the target recommendation identifier, the target commodity recommendation list and the target commodity ID;

when a current commodity ID set corresponding to a current transaction request sent by a main server is received, determining a corresponding first recommendation identification set and a first commodity recommendation list set by the recommendation server according to the current commodity ID set;

The recommendation server determines the association condition among all current transaction commodities in the current transaction request according to the first recommendation identification set and the first commodity recommendation list set, and generates commodity recommendation data according to the association condition, the first recommendation identification set and the first commodity recommendation list set;

And the recommendation server pushes the commodity recommendation data to the user side.

Further, the artificial intelligence model comprises a first artificial intelligence sub-model and a second artificial intelligence sub-model; the corresponding relation comprises a first corresponding sub-relation and a second corresponding sub-relation; the step of establishing the corresponding relation between the commodity description information and the commodity introduction chart of the commodity, the commodity recommendation mark and the commodity recommendation list through the artificial intelligent model comprises the following steps:

The recommendation server establishes commodity description information of the commodity and a first corresponding sub-relationship between a commodity introduction chart and a recommendation identifier of the commodity through a first artificial intelligent sub-model;

And the recommendation server establishes a second corresponding sub-relationship between the recommendation identification and the commodity recommendation list through a second artificial intelligence sub-model.

Further, the step of determining the corresponding first recommendation identification set and the first commodity recommendation list set according to the current commodity ID set includes:

the recommendation server generates a first recommendation identification set by using recommendation identifications associated with all commodity IDs in the current commodity ID set;

And the recommendation server generates a commodity recommendation list corresponding to each recommendation identifier in the first recommendation identifier set into the first commodity recommendation list set.

Further, the step of determining the association condition between the current transaction commodities in the current transaction request according to the first recommendation identification set and the first commodity recommendation list set, and generating commodity recommendation data according to the association condition, the first recommendation identification set and the first commodity recommendation list set includes:

The recommendation server determines whether the first commodity recommendation list set contains recommendation identifiers in the first recommendation identifier set;

When the first commodity recommendation list set comprises recommendation identifiers in the first recommendation identifier set, the recommendation server deletes recommendation identifiers which are the same as the first commodity recommendation identifier set in the first commodity recommendation list set to obtain a second commodity recommendation list set;

and the recommendation server generates the commodity recommendation data according to the second commodity recommendation list set.

Further, the step of generating the commodity recommendation data by the recommendation server according to the second commodity recommendation list set includes:

the recommendation server sorts the recommendation identifications in the second commodity recommendation list set according to a preset weight rule, and screens out the recommendation identifications of the fifth highest rank to obtain a second recommendation identification set;

The recommendation server matches commodity IDs with association relation with the second recommendation identification set to obtain a first recommendation commodity ID set; the number of the commodity IDs matched with each recommendation identifier in the second recommendation identifier set is 2-5;

and the recommendation server generates the commodity recommendation data according to the first recommended commodity ID set.

Further, the method further comprises the steps of:

When the first commodity recommendation list set does not contain recommendation identifiers in the first recommendation identifier set, the recommendation server generates commodity recommendation data according to the first commodity recommendation list set.

Further, the step of generating the commodity recommendation data according to the first commodity recommendation list set includes:

The recommendation server sorts recommendation identifiers in the first commodity recommendation list set according to a preset weight rule, and screens out the recommendation identifiers with the top five ranks to obtain a third recommendation identifier set;

The recommendation server matches commodity IDs with association relation with the third recommendation identification set to obtain a second recommendation commodity ID set; the number of the commodity IDs matched with each recommendation identifier in the third recommendation identifier set is 2-5;

And the recommendation server generates the commodity recommendation data according to the second recommended commodity ID set.

The device relates to a user side, a merchant side, a main server and a recommendation server; the merchant terminal is used for sending a commodity input request to the main server; the commodity input request comprises commodity description information corresponding to input commodities and a commodity introduction chart; the main server is used for generating commodity IDs corresponding to the input commodities one by one and sending the commodity IDs and the commodity input requests to the recommendation server; the user is used for sending a transaction request to the main server; the transaction request comprises transaction information of at least two different transaction commodities, wherein the transaction commodities are recorded and bound with commodity IDs; the main server is used for determining a commodity ID set according to the transaction request and sending the commodity ID set to the recommendation server; wherein the commodity ID set comprises commodity IDs of all trade commodities in the trade request;

The device comprises:

The relation establishing module is used for establishing the corresponding relation between the commodity description information and the commodity introduction chart of the commodity, the recommendation identification of the commodity and the commodity recommendation list through the artificial intelligent model; the commodity recommendation list comprises five different recommendation identifiers which have purchase relevance with commodities;

The commodity input module is used for determining a target recommendation identifier and a target commodity recommendation list corresponding to the target input commodity according to the target commodity input request through the corresponding relation when receiving the target commodity input request and the target commodity ID, which are sent by the main server and correspond to the target input commodity, and associating the target recommendation identifier, the target commodity recommendation list and the target commodity ID;

the transaction request receiving module is used for determining a corresponding first recommendation identification set and a first commodity recommendation list set according to a current commodity ID set when receiving the current commodity ID set which is sent by the main server and corresponds to a current transaction request;

The commodity recommendation data generation module is used for determining the association condition among all current transaction commodities in the current transaction request according to the first recommendation identification set and the first commodity recommendation list set, and generating commodity recommendation data according to the association condition, the first recommendation identification set and the first commodity recommendation list set;

And the commodity recommendation data pushing module is used for pushing the commodity recommendation data to the user side.

A computer device comprising a processor, a memory and a computer program stored on the memory and capable of running on the processor, which when executed by the processor implements the steps of the artificial intelligence based user data recommendation method as described above.

A computer readable storage medium having stored thereon a computer program which when executed by a processor implements the steps of an artificial intelligence based user data recommendation method as described above.

The application has the following advantages:

In the embodiment of the application, aiming at the problems of resource waste caused by repeated or wrong recommendation data browsed by a user due to inaccurate recommendation commodities screened in a traditional recommendation data mode because of disordered commodity keywords and uneven correlation set by merchants in the prior art, the application provides a solution for determining commodity recommendation identifications of commodity description information and commodity introduction diagrams of commodities and recommending commodities according to the recommendation identifications by an artificial intelligent model, which comprises the following specific steps: establishing commodity description information and corresponding relation between a commodity introduction chart and a commodity recommendation identifier and a commodity recommendation list of the commodity through an artificial intelligent model; the commodity recommendation list comprises five different recommendation identifiers which have purchase relevance with commodities; when a target commodity input request and a target commodity ID, which are sent by a main server and correspond to target input commodities, are received, determining a target recommendation identifier and a target commodity recommendation list, which correspond to the target input commodities, according to the target commodity input request through the corresponding relation, and associating the target recommendation identifier, the target commodity recommendation list and the target commodity ID; when a current commodity ID set corresponding to a current transaction request sent by a main server is received, determining a corresponding first recommendation identification set and a first commodity recommendation list set according to the current commodity ID set; determining the association condition among all current transaction commodities in the current transaction request according to the first recommendation identification set and the first commodity recommendation list set, and generating commodity recommendation data according to the association condition, the first recommendation identification set and the first commodity recommendation list set; pushing the commodity recommendation data to the user side; the accuracy of the generated commodity recommendation data is improved by establishing the commodity description information and the corresponding relation between the commodity introduction chart, the commodity recommendation identification and the commodity recommendation list; the recommendation identification, the commodity recommendation list and the commodity ID are pre-associated, so that the efficiency of the generated commodity recommendation data is improved; and generating commodity recommendation data according to the association condition, the first recommendation identification set and the first commodity recommendation list set, so that the condition that commodity recommendation data repeated with purchased commodities are generated when the commodity recommendation data are generated is reduced.

Drawings

In order to more clearly illustrate the technical solutions of the present application, the following brief description will be given of the drawings required for the description of the present application, it being obvious that the drawings in the following description are only some embodiments of the present application, and that other drawings may be obtained from these drawings without inventive faculty for a person skilled in the art.

FIG. 1 is a flow chart of a user data recommendation method based on artificial intelligence according to an embodiment of the present application;

FIG. 2 is a block diagram of an artificial intelligence based user data recommendation device according to an embodiment of the present application;

fig. 3 is a schematic structural diagram of a computer device according to an embodiment of the present application.

Detailed Description

In order that the manner in which the above recited objects, features and advantages of the present application are obtained will become more readily apparent, a more particular description of the application briefly described above will be rendered by reference to the appended drawings. It will be apparent that the described embodiments are some, but not all, embodiments of the application. All other embodiments, which can be made by those skilled in the art based on the embodiments of the application without making any inventive effort, are intended to be within the scope of the application.

The inventors found by analyzing the prior art that: because of the regional culture difference and diversity of commodity names of each merchant, the same commodity setting keywords are different, or different commodity setting keywords are overlapped, so that the generated recommended commodity data can have wrong commodity data which is not related to the purchased commodity or duplicate data which is identical to the purchased commodity in a related commodity pushing scene after the purchased commodity is purchased. The problem of data accuracy is solved, the processing threshold of a merchant is greatly increased by restricting commodity introduction information and keywords provided by the merchant, and the user experience of a merchant user is reduced; meanwhile, related information of excessively restricting the commodity can also prevent the commodity propaganda of merchants; the commodity introduction chart is necessarily accompanied with related patterns of commodities, and although the description information of the commodities may deviate, the related patterns of the commodities have tendency, so that the accuracy of commodity recommendation data can be effectively improved by carrying out image recognition on the commodity introduction chart and combining the recognition information with text information corresponding to the commodities.

Referring to fig. 1, an artificial intelligence based user data recommendation method provided by an embodiment of the present application is shown, where the method relates to a user side, a merchant side, a main server, and a recommendation server; the merchant terminal is used for sending a commodity input request to the main server; the commodity input request comprises commodity description information corresponding to input commodities and a commodity introduction chart; the main server is used for generating commodity IDs corresponding to the input commodities one by one and sending the commodity IDs and the commodity input requests to the recommendation server; the user is used for sending a transaction request to the main server; the transaction request comprises transaction information of at least two different transaction commodities, wherein the transaction commodities are recorded and bound with commodity IDs; the main server is used for determining a commodity ID set according to the transaction request and sending the commodity ID set to the recommendation server; wherein the commodity ID set comprises commodity IDs of all trade commodities in the trade request;

The method comprises the following steps:

S110, the recommendation server establishes commodity description information of commodities and corresponding relations between commodity introduction diagrams, recommendation identifications of the commodities and commodity recommendation lists through an artificial intelligent model; the commodity recommendation list comprises five different recommendation identifiers which have purchase relevance with commodities;

S120, when a target commodity input request and a target commodity ID, which are sent by a main server and correspond to a target input commodity, are received, determining a target recommendation identifier and a target commodity recommendation list, which correspond to the target input commodity, according to the target commodity input request by the recommendation server through the corresponding relation, and associating the target recommendation identifier, the target commodity recommendation list and the target commodity ID;

S130, when a current commodity ID set corresponding to a current transaction request sent by a main server is received, determining a corresponding first recommendation identification set and a first commodity recommendation list set by the recommendation server according to the current commodity ID set;

S140, the recommendation server determines the association condition among all current transaction commodities in the current transaction request according to the first recommendation identification set and the first commodity recommendation list set, and generates commodity recommendation data according to the association condition, the first recommendation identification set and the first commodity recommendation list set;

And S150, pushing the commodity recommendation data to the user side by the recommendation server.

Next, a user data recommendation method based on artificial intelligence in the present exemplary embodiment will be further described.

As described in step S110, the recommendation server establishes the corresponding relationship between the commodity description information and the commodity introduction chart of the commodity, the recommendation identifier of the commodity and the commodity recommendation list through the artificial intelligent model; the commodity recommendation list comprises five different recommendation identifiers which have purchase relevance with commodities;

It should be noted that the recommendation identifiers included in the commodity recommendation list are marked with corresponding association levels, where the association levels include LV1, LV2, LV3. The recommendation identification types included in the commodity recommendation list are LV1 not less than two, LV2 not more than two and LV3 not more than one.

Note that, the recommended identifier of the commodity may be a character string composed of characters (the characters include any two or more of numerals, letters, and symbols), and information such as the type, classification, use, and the like of the commodity may be expressed by using different characters. The recommendation identification corresponding to each commodity is unique, and the same recommendation identification corresponds to a plurality of commodities. For example, the desk, electronic contest table, and computer table may all correspond to the same recommendation identifier "XX-XX-XX-XX".

The artificial intelligence model comprises a first artificial intelligence sub-model and a second artificial intelligence sub-model; the corresponding relation comprises a first corresponding sub-relation and a second corresponding sub-relation;

In an embodiment, the specific process of "building the corresponding relationship between the commodity description information and the commodity introduction chart of the commodity and the recommendation identifier and the commodity recommendation list of the commodity" in step S110 may be further described in conjunction with the following description.

It should be noted that, the first artificial intelligence sub-model preferably includes a first artificial neural network module for image recognition and a second artificial neural network module for establishing a first corresponding sub-relationship, and the first artificial neural network module is used for performing image recognition on the commodity introduction graph to determine a commodity name set of the commodity contained in the image, where the commodity introduction graph generally includes an article for assisting in commodity display, for example: the commodity introduction graph of the refrigerator often has objects such as vegetables, fruits and fish for assisting introduction, so that the commodity name set output by the first artificial neural network module needs to be screened and denoised according to commodity description information to confirm the commodity name corresponding to the commodity; specifically, text matching is performed on the commodity name set and commodity description information, the same or similar commodity names existing in the two texts are screened out, and the screened commodity names output by the first artificial neural network module are used as input of the second artificial neural network module. And finally, establishing a first corresponding sub-relationship between the commodity name and the recommended identification through a second artificial neural network module.

It should be noted that the first artificial neural network module may be a neural network whose structure includes 6 convolutional layers, 6 pooling layers, 6 batch normalization layers, 1 flat layer, 2 full connection layers, 2 Dropout layers, 1 Reshape layers, and 1 Softmax regression layer.

And establishing a second corresponding sub-relationship between the recommendation identification and the commodity recommendation list by the recommendation server through a second artificial intelligence sub-model as described in the following steps.

It should be noted that the second correspondence may be created based on purchasing habits and commodity association scores of a large number of different buyer users (including, but not limited to, one or more of age, gender, occupation, etc.), where the purchasing habits and commodity association scores are used to determine a relevance score between different recommendation identifications, and the recommendation identifications of the first five relevance scores are generated as a commodity recommendation list corresponding to the target recommendation identifications.

It should be noted that, since the commodity corresponding to the recommendation identifier is not unique, for example: iron door, timber, mixed material door, the door correspondence of different materials has different relevance scores with same commodity.

Therefore, the method can analyze the association rules of the association scores between the buying habits and the commodity association evaluations and the different recommendation identifications by utilizing an artificial intelligent model algorithm, and find the mapping rules between the association scores between the buying habits and the commodity association evaluations and the different recommendation identifications by self-learning and self-adapting characteristics of the artificial intelligent model so as to obtain the association score corresponding relation between the buying habits and the recommendation identifications between the commodity association evaluations.

For example: the method can utilize an artificial intelligent model algorithm to learn and train the artificial intelligent model by collecting the buying habits, commodity association evaluations and association scores among recommendation identifications of a large number of different volunteers (including but not limited to one or more of age, sex, occupation and the like), selecting the buying habits, commodity association evaluations and association scores among recommendation identifications of a plurality of buyer users as sample data, and enabling the artificial intelligent model to fit the corresponding relation among the buying habits, commodity association evaluations and the association scores among recommendation identifications by adjusting the weight among model structures and model nodes, so that the artificial intelligent model can accurately fit the association scores among different recommendation identifications.

The product association evaluation may be a survey evaluation of whether the volunteer collected by the e-commerce platform has association with the purchased product, for example, whether the product a and the product B are associated products (yes/no); the buying habit may be at least two of the goods contained in the order, the goods in the shopping cart, and the goods order of the last 5 times in the period each time the volunteer purchases the goods; the relevance score can be a score value of 100 points, the relevance grade can be 90-100 points to LV1, 75-89 points to LV2, 65-74 points to LV3, and less than 65 points to no grade.

As described in step S120, when receiving a target commodity input request and a target commodity ID, which are sent by a main server and correspond to a target input commodity, the recommendation server determines a target recommendation identifier and a target commodity recommendation list, which correspond to the target input commodity, according to the target commodity input request through the correspondence, and associates the target recommendation identifier, the target commodity recommendation list and the target commodity ID;

In an embodiment, the specific process of "when the target commodity input request and the target commodity ID corresponding to the target input commodity sent by the main server are received, the recommendation server determines the target recommendation identifier and the target commodity recommendation list corresponding to the target input commodity according to the target commodity input request through the correspondence, and associates the target recommendation identifier, the target commodity recommendation list and the target commodity ID" in step S120 may be further described in connection with the following description.

When the merchant user performs the process of loading the commodity, after the commodity is assigned with the commodity ID, the main server automatically sends information corresponding to the commodity to the recommendation server for commodity entry.

When an input request is received, a recommendation identifier corresponding to the commodity and a corresponding commodity recommendation list can be matched through a pre-established corresponding relation, wherein the commodity ID and the associated recommendation identifier are separately and independently processed in different servers, so that the associated recommendation identifier can be independently processed outside the commodity loading process, and the commodity loading efficiency of a merchant user is not negatively influenced.

As described in the above step S130, when a current commodity ID set corresponding to a current transaction request sent by a main server is received, the recommendation server determines a corresponding first recommendation identifier set and a first commodity recommendation list set according to the current commodity ID set;

in an embodiment, the specific process of step S130 "when the current product ID set corresponding to the current transaction request sent by the main server is received, the recommendation server determines the corresponding first recommendation identifier set and the first product recommendation list set according to the current product ID set" may be further described in conjunction with the following description.

The recommendation server generates the first recommendation identification set from recommendation identifications associated with each of the product IDs in the current product ID set;

It should be noted that, after receiving the commodity ID set in the transaction order, the recommendation server respectively matches recommendation identifiers corresponding to the respective commodity IDs in the commodity ID set, where the number of matched recommendation identifiers may include the same recommendation identifier, and generates the first recommendation identifier set from all the matched recommendation identifiers.

The recommendation server generates the first commodity recommendation list set from commodity recommendation lists corresponding to respective recommendation identifications in the first recommendation identification set, as described in the following steps.

It should be noted that, because the first recommendation identifier set may have the same recommendation identifier, when the first recommendation identifier set matches the commodity recommendation list, the same recommendation identifier is only matched for the first time, and repeated recommendation identifier matching is invalid, so as to reduce unnecessary operation times in subsequent processing steps.

As described in step S140, the recommendation server determines a correlation condition between each current transaction commodity in the current transaction request according to the first recommendation identification set and the first commodity recommendation list set, and generates commodity recommendation data according to the correlation condition, the first recommendation identification set and the first commodity recommendation list set;

In an embodiment, the specific process of "the recommendation server determines the association condition between each current transaction commodity in the current transaction request according to the first recommendation identification set and the first commodity recommendation list set, and generates commodity recommendation data according to the association condition, the first recommendation identification set and the first commodity recommendation list set" in step S140 may be further described in conjunction with the following description.

The recommendation server determines whether the first commodity recommendation list set contains recommendation identifiers in the first recommendation identifier set or not;

Since the commodity recommendation list itself is composed of recommendation identifiers, there is a case where the recommendation identifiers corresponding to the other commodities in the order are included in the commodity recommendation list corresponding to the recommendation identifier corresponding to the certain commodity in the same order, and in this case, there is a possibility that the recommended commodity is repeatedly generated, and therefore, in order to avoid the above-described case, the above-described determination is required.

When the first commodity recommendation list set includes recommendation identifiers in the first recommendation identifier set, the recommendation server deletes recommendation identifiers in the first commodity recommendation list set, which are the same as the first recommendation identifier set, to obtain a second commodity recommendation list set;

In one embodiment, the specific process of generating the commodity recommendation data by the recommendation server according to the second commodity recommendation list set in the above step may be further described in conjunction with the following description.

The recommendation server sorts the recommendation identifications in the second commodity recommendation list set according to a preset weight rule, and screens out the recommendation identifications of the fifth top rank to obtain a second recommendation identification set;

It should be noted that the preset weight rule includes a number of repeated occurrences a of the recommendation identifier in the first recommendation identifier set, a number of repeated occurrences B of the recommendation identifier in the second commodity recommendation list set, and an association level C of each recommendation identifier in each commodity recommendation list. Wherein, the weight ratio may be a: b: c=48%: 36%: and 16, calculating scores of all recommendation identifications in the second commodity recommendation list set according to the weight rule, sorting according to the scores, and screening out the recommendation identifications with the top five ranks.

The score calculation formula is

Score = N _A×48%+N_B×36%+N_C x 16%

Wherein N _A is the repeated occurrence number of the recommended marks in the first recommended mark set; n _B is the repeated occurrence number of the recommendation marks in the second commodity recommendation list set; n _C is the score value (LV 1 is 2, LV2 is 1, LV3 is 0.5) corresponding to the association level of the recommendation identifier.

It should be noted that, the weight value is a non-fixed value, and may be adaptively adjusted according to feedback conditions of the buyer user.

When parallel scores appear, the ranking is performed according to the weight condition of the weight rule, for example, when the scores are the same, the ranking is preferentially performed by the numerical value of A, the ranking is performed next by the numerical value of B, and finally the ranking is performed by the numerical value of C.

The recommendation server matches the commodity IDs with the association relation with the second recommendation identification set to obtain a first recommendation commodity ID set; the number of the commodity IDs matched with each recommendation identifier in the second recommendation identifier set is 2-5;

It should be noted that, the number of commodity IDs corresponding to the recommendation identifier is generally greater than 5, so that the recommended commodities need to be screened, and the screening rule may be in a manner of according to commodity sales, good evaluation rate, poor evaluation rate and the like, or may be in a manner of according to a business rule set up by the e-commerce platform.

The recommendation server generates the commodity recommendation data according to the first recommended commodity ID set as described in the following steps.

In another embodiment, the specific process of "the recommendation server determines the association condition between each current transaction commodity in the current transaction request according to the first recommendation identification set and the first commodity recommendation list set, and generates commodity recommendation data according to the association condition, the first recommendation identification set and the first commodity recommendation list set" in step S140 may be further described in conjunction with the following description.

When the first commodity recommendation list set does not include the recommendation identifier in the first recommendation identifier set, the recommendation server generates the commodity recommendation data according to the first commodity recommendation list set.

In one embodiment, the specific process of generating the commodity recommendation data by the recommendation server according to the first commodity recommendation list set in the above step may be further described in conjunction with the following description.

The score calculation formula is

Score = N _A×48%+N_B×36%+N_C x 16%

As described in step S150, the recommendation server pushes the commodity recommendation data to the user terminal.

For the device embodiments, since they are substantially similar to the method embodiments, the description is relatively simple, and reference is made to the description of the method embodiments for relevant points.

Referring to fig. 2, there is shown an artificial intelligence based user data recommending apparatus provided in an embodiment of the present application,

The method specifically comprises the following steps:

The relationship establishing module 210 is configured to establish, through the artificial intelligence model, a correspondence between the commodity description information and the commodity introduction chart of the commodity, the recommendation identifier of the commodity, and the commodity recommendation list; the commodity recommendation list comprises five different recommendation identifiers which have purchase relevance with commodities;

The commodity input module 220 is configured to determine, when a target commodity input request and a target commodity ID corresponding to a target input commodity sent by a main server are received, a target recommendation identifier and a target commodity recommendation list corresponding to the target input commodity according to the target commodity input request through the correspondence, and associate the target recommendation identifier, the target commodity recommendation list and the target commodity ID;

The transaction request receiving module 230 is configured to determine, when a current commodity ID set corresponding to a current transaction request sent by a main server is received, a corresponding first recommendation identifier set and a first commodity recommendation list set according to the current commodity ID set;

The commodity recommendation data generation module 240 is configured to determine a correlation condition between each current transaction commodity in the current transaction request according to the first recommendation identification set and the first commodity recommendation list set, and generate commodity recommendation data according to the correlation condition, the first recommendation identification set and the first commodity recommendation list set;

and the commodity recommendation data pushing module 250 is configured to push the commodity recommendation data to the user side.

In one embodiment of the invention, the artificial intelligence model includes a first artificial intelligence sub-model and a second artificial intelligence sub-model; the corresponding relation comprises a first corresponding sub-relation and a second corresponding sub-relation; the relationship establishment module 210 includes:

The first sub-key establishment sub-module is used for establishing commodity description information of the commodity and a first corresponding sub-relationship between a commodity introduction chart and a recommendation identifier of the commodity through the first artificial intelligent sub-model by the recommendation server;

and the second sub-key establishment sub-module is used for establishing a second corresponding sub-relationship between the recommendation identification and the commodity recommendation list when the recommendation server establishes the recommendation identification and the commodity recommendation list through a second artificial intelligence sub-model.

In one embodiment of the present invention, the transaction request receiving module 230 includes:

A first recommendation identification set generating sub-module, configured to generate a recommendation identification set associated with each product ID in the current product ID set;

and the first commodity recommendation list set generation sub-module is used for generating the commodity recommendation list corresponding to each recommendation identifier in the first recommendation identifier set into the first commodity recommendation list set.

In an embodiment of the present invention, the commodity recommendation data generation module 240 includes:

a recommendation identification determining sub-module, configured to determine whether the first commodity recommendation list set includes recommendation identifications in the first recommendation identification set;

The second commodity recommendation list set determining submodule is used for deleting the recommendation identifiers which are the same as the first recommendation identifier set in the first commodity recommendation list set when the first commodity recommendation list set contains the recommendation identifiers in the first recommendation identifier set, so as to obtain a second commodity recommendation list set;

And the first commodity recommendation data generation sub-module is used for generating the commodity recommendation data according to the second commodity recommendation list set.

In an embodiment of the present invention, the first merchandise recommendation data generation sub-module includes:

The second recommendation identification set determining unit is used for sequencing recommendation identifications in the second commodity recommendation list set according to a preset weight rule, screening out the recommendation identifications of the fifth top ranking, and obtaining a second recommendation identification set;

The first recommended commodity ID set determining unit is used for matching commodity IDs with association relation with the second recommended identification set to obtain a first recommended commodity ID set; the number of the commodity IDs matched with each recommendation identifier in the second recommendation identifier set is 2-5;

And the first commodity recommendation data generation unit is used for generating the commodity recommendation data according to the first recommended commodity ID set.

In an embodiment of the present invention, further includes:

And the second commodity recommendation data generation sub-module is used for generating the commodity recommendation data according to the first commodity recommendation list set when the first commodity recommendation list set does not contain the recommendation identifiers in the first recommendation identifier set.

In an embodiment of the present invention, the generating the second merchandise recommendation data includes:

The third recommendation identification set determining unit is used for sequencing recommendation identifications in the first commodity recommendation list set according to a preset weight rule, screening out the fifth recommendation identifications in the ranking, and obtaining a third recommendation identification set;

The second recommended commodity ID set unit is used for matching commodity IDs with association relation with the third recommended identification set to obtain a second recommended commodity ID set; the number of the commodity IDs matched with each recommendation identifier in the third recommendation identifier set is 2-5;

And the second commodity recommendation data unit is used for generating commodity recommendation data according to the second recommended commodity ID set.

Referring to fig. 3, a computer device of the present invention for an artificial intelligence based user data recommendation method is shown, which may specifically include the following:

The computer device 12 described above is embodied in the form of a general purpose computing device, and the components of the computer device 12 may include, but are not limited to: one or more processors or processing units 16, a system memory 28, a bus 18 that connects the various system components, including the system memory 28 and the processing units 16.

Bus 18 represents one or more of several types of bus structures, including a memory bus or memory controller, a peripheral bus, an accelerated graphics port, a processor, and a local bus using any of a variety of bus architectures. By way of example, and not limitation, such architectures include Industry Standard Architecture (ISA) bus, micro channel architecture (MAC) bus, enhanced ISA bus, video Electronics Standards Association (VESA) local bus, and Peripheral Component Interconnect (PCI) bus.

Computer device 12 typically includes a variety of computer system readable media. Such media can be any available media that is accessible by computer device 12 and includes both volatile and nonvolatile media, removable and non-removable media.

The system memory 28 may include computer system readable media in the form of volatile memory, such as Random Access Memory (RAM) 30 and/or cache memory 32. The computer device 12 may further include other removable/non-removable, volatile/nonvolatile computer system storage media. By way of example only, storage system 34 may be used to read from or write to non-removable, nonvolatile magnetic media (commonly referred to as a "hard disk drive"). Although not shown in fig. 3, a magnetic disk drive for reading from and writing to a removable non-volatile magnetic disk (e.g., a "floppy disk"), and an optical disk drive for reading from or writing to a removable non-volatile optical disk such as a CD-ROM, DVD-ROM, or other optical media may be provided. In such cases, each drive may be coupled to bus 18 through one or more data medium interfaces. The memory may include at least one program product having a set (e.g., at least one) of program modules 42, the program modules 42 being configured to carry out the functions of embodiments of the invention.

Program/utility 40 having a set (at least one) of program modules 42 may be stored in, for example, a memory, such program modules 42 including, but not limited to, an operating system, one or more application programs, other program modules, and program data, each or some combination of which may include an implementation of a network environment. Program modules 42 generally perform the functions and/or methods of the embodiments described herein.

The computer device 12 may also communicate with one or more external devices 14 (e.g., keyboard, pointing device, display 24, camera, etc.), one or more devices that enable a user to interact with the computer device 12, and/or any devices (e.g., network card, modem, etc.) that enable the computer device 12 to communicate with one or more other computing devices. Such communication may occur through the I/O interface 22. Moreover, the computer device 12 may also communicate with one or more networks such as a Local Area Network (LAN), a Wide Area Network (WAN) and/or a public network (e.g., the Internet) via a network adapter 20. As shown in fig. 3, the network adapter 20 communicates with other modules of the computer device 12 via the bus 18. It should be appreciated that although not shown in fig. 3, other hardware and/or software modules may be used in connection with computer device 12, including, but not limited to: microcode, device drivers, processing units 16, external disk drive arrays, RAID systems, tape drives, storage systems 34, and the like.

The processing unit 16 executes various functional applications and data processing by running programs stored in the system memory 28, for example, to implement an artificial intelligence-based user data recommendation method provided by an embodiment of the present invention.

That is, the processing unit 16 realizes when executing the program: establishing commodity description information and corresponding relation between a commodity introduction chart and a commodity recommendation identifier and a commodity recommendation list of the commodity through an artificial intelligent model; the commodity recommendation list comprises five different recommendation identifiers which have purchase relevance with commodities; when a target commodity input request and a target commodity ID, which are sent by a main server and correspond to target input commodities, are received, determining a target recommendation identifier and a target commodity recommendation list, which correspond to the target input commodities, according to the target commodity input request through the corresponding relation, and associating the target recommendation identifier, the target commodity recommendation list and the target commodity ID; when a current commodity ID set corresponding to a current transaction request sent by a main server is received, determining a corresponding first recommendation identification set and a first commodity recommendation list set according to the current commodity ID set; the recommendation server determines the association condition among all current transaction commodities in the current transaction request according to the first recommendation identification set and the first commodity recommendation list set, and generates commodity recommendation data according to the association condition, the first recommendation identification set and the first commodity recommendation list set; pushing the commodity recommendation data to the user side.

In an embodiment of the present application, the present application further provides a computer readable storage medium having stored thereon a computer program which, when executed by a processor, implements an artificial intelligence based user data recommendation method as provided in all embodiments of the present application:

That is, the program is implemented when executed by a processor: establishing commodity description information and corresponding relation between a commodity introduction chart and a commodity recommendation identifier and a commodity recommendation list of the commodity through an artificial intelligent model; the commodity recommendation list comprises five different recommendation identifiers which have purchase relevance with commodities; when a target commodity input request and a target commodity ID, which are sent by a main server and correspond to target input commodities, are received, determining a target recommendation identifier and a target commodity recommendation list, which correspond to the target input commodities, according to the target commodity input request through the corresponding relation, and associating the target recommendation identifier, the target commodity recommendation list and the target commodity ID; when a current commodity ID set corresponding to a current transaction request sent by a main server is received, determining a corresponding first recommendation identification set and a first commodity recommendation list set according to the current commodity ID set; the recommendation server determines the association condition among all current transaction commodities in the current transaction request according to the first recommendation identification set and the first commodity recommendation list set, and generates commodity recommendation data according to the association condition, the first recommendation identification set and the first commodity recommendation list set; pushing the commodity recommendation data to the user side.

Any combination of one or more computer readable media may be employed. The computer readable medium may be a computer-readable signal medium or a computer-readable storage medium. 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 (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 (EPOM 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 this document, 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 computer readable signal medium may include a propagated data signal with computer readable program code embodied therein, either in baseband or as part of a carrier wave. 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.

Computer program code for carrying out operations of the present invention may be written in any combination of 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 remote computers, 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). In this specification, each embodiment is described in a progressive manner, and each embodiment is mainly described by differences from other embodiments, and identical and similar parts between the embodiments are all enough to be referred to each other.

While preferred embodiments of the present application have been described, additional variations and modifications in those embodiments may occur to those skilled in the art once they learn of the basic inventive concepts. It is therefore intended that the following claims be interpreted as including the preferred embodiment and all such alterations and modifications as fall within the scope of the embodiments of the application.

Finally, it is further noted that relational terms such as first and second, and the like are used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Moreover, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or terminal that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or terminal. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article, or terminal device that comprises the element.

The above description of the user data recommendation method and device based on artificial intelligence provided by the application applies specific examples to illustrate the principle and implementation of the application, and the above description of the examples is only used for helping to understand the method and core idea of the application; meanwhile, as those skilled in the art will have variations in the specific embodiments and application scope in accordance with the ideas of the present application, the present description should not be construed as limiting the present application in view of the above.

Claims

1. The method is used for a scene of associated commodity pushing after purchasing commodities and is characterized by comprising a user side, a merchant side, a main server and a recommendation server; the merchant terminal is used for sending a commodity input request to the main server; the commodity input request comprises commodity description information corresponding to input commodities and a commodity introduction chart; the main server is used for generating commodity IDs corresponding to the input commodities one by one and sending the commodity IDs and the commodity input requests to the recommendation server; the user is used for sending a transaction request to the main server; the transaction request comprises transaction information of at least two different transaction commodities, wherein the transaction commodities are recorded and bound with commodity IDs; the main server is used for determining a commodity ID set according to the transaction request and sending the commodity ID set to the recommendation server; wherein the commodity ID set comprises commodity IDs of all trade commodities in the trade request; the recommendation identification corresponding to each commodity is unique, and the same recommendation identification corresponds to a plurality of similar commodities;

The method comprises the following steps:

the recommendation server determines the association condition among all current transaction commodities in the current transaction request according to the first recommendation identification set and the first commodity recommendation list set, and generates commodity recommendation data according to the association condition, the first recommendation identification set and the first commodity recommendation list set; specifically, the recommendation server determines whether the first commodity recommendation list set includes recommendation identifiers in the first recommendation identifier set; when the first commodity recommendation list set comprises recommendation identifiers in the first recommendation identifier set, the recommendation server deletes recommendation identifiers which are the same as the first commodity recommendation identifier set in the first commodity recommendation list set to obtain a second commodity recommendation list set; the recommendation server generates the commodity recommendation data according to the second commodity recommendation list set;

2. The method of claim 1, wherein the artificial intelligence model comprises a first artificial intelligence sub-model and a second artificial intelligence sub-model; the corresponding relation comprises a first corresponding sub-relation and a second corresponding sub-relation; the step of establishing the corresponding relation between the commodity description information and the commodity introduction chart of the commodity, the commodity recommendation mark and the commodity recommendation list through the artificial intelligent model comprises the following steps:

3. The method of claim 1, wherein the step of determining the corresponding first set of recommended identifications and first set of recommended lists of items from the current set of item IDs comprises:

4. The method of claim 3, wherein the step of the recommendation server generating the merchandise recommendation data from the second set of merchandise recommendation lists comprises:

5. A method according to claim 3, further comprising the step of:

6. The method of claim 5, wherein the step of generating the merchandise recommendation data from the first set of merchandise recommendation lists comprises:

7. The device is used for a scene of associated commodity pushing after purchasing commodities and is characterized by comprising a user side, a merchant side, a main server and a recommendation server; the merchant terminal is used for sending a commodity input request to the main server; the commodity input request comprises commodity description information corresponding to input commodities and a commodity introduction chart; the main server is used for generating commodity IDs corresponding to the input commodities one by one and sending the commodity IDs and the commodity input requests to the recommendation server; the user is used for sending a transaction request to the main server; the transaction request comprises transaction information of at least two different transaction commodities, wherein the transaction commodities are recorded and bound with commodity IDs; the main server is used for determining a commodity ID set according to the transaction request and sending the commodity ID set to the recommendation server; wherein the commodity ID set comprises commodity IDs of all trade commodities in the trade request; the recommendation identification corresponding to each commodity is unique, and the same recommendation identification corresponds to a plurality of similar commodities;

The device comprises:

The commodity recommendation data generation module is used for determining the association condition among all current transaction commodities in the current transaction request according to the first recommendation identification set and the first commodity recommendation list set, and generating commodity recommendation data according to the association condition, the first recommendation identification set and the first commodity recommendation list set; specifically, the recommendation server determines whether the first commodity recommendation list set includes recommendation identifiers in the first recommendation identifier set; when the first commodity recommendation list set comprises recommendation identifiers in the first recommendation identifier set, the recommendation server deletes recommendation identifiers which are the same as the first commodity recommendation identifier set in the first commodity recommendation list set to obtain a second commodity recommendation list set; the recommendation server generates the commodity recommendation data according to the second commodity recommendation list set;

8. A computer device comprising a processor, a memory and a computer program stored on the memory and capable of running on the processor, which when executed by the processor implements the artificial intelligence based user data recommendation method according to any one of claims 1 to 6.

9. A computer readable storage medium, wherein a computer program is stored on the computer readable storage medium, which computer program, when executed by a processor, implements the artificial intelligence based user data recommendation method according to any one of claims 1 to 6.