An intelligent product selection management method and system based on machine learning

Abstract

The embodiments of this application belong to the field of artificial intelligence and relate to an intelligent product selection management method based on machine learning, including: generating a customer preference portrait for each customer based on historical purchase data and browsing data; performing attribute embedding and knowledge graph embedding on the product information of each product to obtain a product portrait for each product; constructing a customer-product interaction graph according to historical purchase data and browsing data; performing representation learning on the customer-product interaction graph to obtain product node features for each product node; generating a comprehensive product feature for each product according to the product portrait of each product and the product node features of its corresponding product node; combining the customer preference portrait of the target customer and the comprehensive product features of each product respectively to obtain combined features and inputting them into a matching model to obtain the degree of interest between the target customer and each product; determining a target product among each product according to the obtained degree of interest for product recommendation. This application improves the accuracy of product selection and recommendation.

Description

Technical Field

[0001] This application relates to the field of artificial intelligence technology, and particularly to an intelligent product selection management method, device, computer device, and storage medium based on machine learning. Background Art

[0002] In e-commerce platforms, the number of products is huge and the types are diverse. Consumers often face problems of information overload and decision-making difficulties when choosing products. At the same time, the purchase preferences and needs of different consumers vary greatly, and traditional product recommendation technologies face many technical challenges. How to accurately capture and understand the needs of consumers is a major problem. Secondly, the product attribute feature dimensions are numerous, and there may be complex correlation relationships between different attributes. How to build a comprehensive and accurate product portrait is another challenge. Moreover, in the context of a large amount of product and user data, how to efficiently construct the matching relationship between users and products also poses high requirements. Therefore, there is an urgent need for an intelligent product selection and recommendation method to improve the accuracy of product selection and recommendation. Summary of the Invention

[0003] The purpose of the embodiments of this application is to propose an intelligent product selection management method, device, computer device, and storage medium based on machine learning to solve the accuracy of product selection and recommendation.

[0004] To solve the above technical problems, the embodiments of this application provide an intelligent product selection management method based on machine learning, and adopt the following technical solutions:

[0005] Based on the historical purchase data and browsing data of each customer, generate the customer preference portrait of each customer through a preset portrait generation algorithm;

[0006] Obtain the product information of each product, and perform attribute embedding and knowledge graph embedding on the product information of each product to obtain the product portrait of each product;

[0007] According to the obtained historical purchase data and browsing data, construct a customer-product interaction graph, where the customer-product interaction graph includes customer nodes and product nodes;

[0008] Perform representation learning on the customer-product interaction graph to obtain the product node features of each product node;

[0009] According to the product portrait of each product and the product node features of its corresponding product node, generate the product comprehensive features of each product;

[0010] Combine the customer preference portrait of the target customer and the product comprehensive features of each product respectively to obtain combined features, and input each combined feature into a matching model to obtain the degree of interest between the target customer and each product;

[0011] Determine a target product among the various products according to the obtained interest degree, and perform product recommendation for the target customer according to the determined target product.

[0012] Further, the step of generating the customer preference portraits of the customers based on the historical purchase data and browsing data of the customers through a preset portrait generation algorithm includes:

[0013] Based on the historical purchase data of the customers, calculate the customer similarity between the customers through a collaborative filtering algorithm, and determine the similar customers of the target customer among the customers according to the obtained customer similarity, where the target customer comes from the customers;

[0014] Determine the purchase preference products of the similar customers according to the historical purchase data of the similar customers, and construct the first product set of the target customer according to the purchase preference products.

[0015] Based on the browsing data of the customers, calculate the product similarity between the products through a content-based recommendation algorithm.

[0016] Determine the browsing preference products similar to the currently browsed product of the target customer according to the obtained product similarity, and add the browsing preference products to the first product set to obtain the second product set of the target customer.

[0017] Generate the customer preference portrait of the target customer according to the second product set, historical purchase data and browsing data of the target customer, and obtain the customer preference portraits of the customers.

[0018] Further, after the step of obtaining the second product set of the target customer, it further includes:

[0019] Process the historical purchase data of the customers through an association rule mining algorithm to obtain product purchase association information.

[0020] Add the purchase association products to the second product set according to the product purchase association information.

[0021] Further, the step of performing attribute embedding and knowledge graph embedding on the product information of the products to obtain the product portraits of the products includes:

[0022] Preprocess the unstructured data in the product information of the products to obtain the structured data of the products.

[0023] Extract attribute information from the structured data of the products to obtain the structured attribute data of the products.

[0024] Map the structured attribute data of each of the products into an attribute embedding representation respectively;

[0025] Based on the attribute embedding representations of the products, determine the association relationships between the products through similarity calculation and clustering algorithms, and construct a product knowledge graph according to the obtained association relationships;

[0026] Fuse the attribute embedding representations and knowledge graph embedding representations of the products to obtain the product portraits of the products.

[0027] Further, the step of performing representation learning on the customer product interaction graph to obtain the product node features of each product node includes:

[0028] Generate a node sequence on the customer product interaction graph through a random walk algorithm;

[0029] Input the node sequence into a skip-gram model to generate the product node features of each product node in the customer product interaction graph by maximizing the node co-occurrence probability.

[0030] Further, the method further includes:

[0031] For each product node in the customer product interaction graph, determine the neighbor nodes of the product node;

[0032] Calculate the node similarity between the product node and each neighbor node;

[0033] Determine the attention weights of the neighbor nodes according to the obtained node similarity;

[0034] Aggregate the product node features of the neighbor nodes according to the obtained attention weights to obtain aggregated neighbor information;

[0035] Update the product node features of the product node according to the aggregated neighbor information.

[0036] Further, the step of generating the product comprehensive features of the products according to the product portraits of the products and the product node features of the corresponding product nodes includes:

[0037] For each product, extract the product embedding vector of the product from the product portrait of the product;

[0038] Merge the product embedding vector with the product node features of the product node corresponding to the product to obtain the product comprehensive features of the product.

[0039] To solve the above technical problems, an embodiment of the present application further provides an intelligent product selection management device based on machine learning, which adopts the following technical solutions:

[0040] A preference generation module, configured to generate customer preference portraits of the customers based on the historical purchase data and browsing data of each customer through a preset portrait generation algorithm;

[0041] A product generation module, configured to obtain product information of each product, and perform attribute embedding and knowledge graph embedding on the product information of each product to obtain product portraits of each product;

[0042] An interaction construction module, configured to construct a customer-product interaction graph according to the obtained historical purchase data and browsing data, where the customer-product interaction graph includes customer nodes and product nodes;

[0043] A feature generation module, configured to perform representation learning on the customer-product interaction graph to obtain product node features of each product node;

[0044] A comprehensive generation module, configured to generate comprehensive product features of each product according to the product portraits of each product and the product node features of the corresponding product nodes;

[0045] A combination calculation module, configured to combine the customer preference portrait of a target customer and the comprehensive product features of each product to obtain combined features respectively, and input each combined feature into a matching model to obtain the interest degrees between the target customer and each product;

[0046] A product recommendation module, configured to determine target products from each product according to the obtained interest degrees, and recommend products to the target customer according to the determined target products.

[0047] To solve the above technical problems, an embodiment of the present application further provides a computer device, where the computer device includes a memory and a processor, and computer-readable instructions are stored in the memory. When the processor executes the computer-readable instructions, the steps of the above-mentioned intelligent product selection management method based on machine learning are implemented.

[0048] To solve the above technical problems, an embodiment of the present application further provides a computer-readable storage medium, on which computer-readable instructions are stored. When the computer-readable instructions are executed by a processor, the steps of the above-mentioned intelligent product selection management method based on machine learning are implemented.

[0049] Compared with the prior art, the embodiments of the present application mainly have the following beneficial effects: generating customer preference portraits for each customer through historical purchase data and browsing data, performing attribute embedding and knowledge graph embedding on the product information of each product to generate high-quality product portraits; constructing a customer-product interaction graph and performing representation learning to further enhance the implicit correlation representation between product nodes; combining the product portraits of each product and the product node features of the corresponding product nodes, and transforming the complex relationship between customers and products into a multi-dimensional representation of product comprehensive features; inputting the customer preference portraits and product comprehensive features into a matching model to accurately calculate the degree of interest, thereby realizing personalized recommendation, improving the accuracy of product selection and recommendation, and increasing the conversion rate of product recommendation and customer satisfaction. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] To more clearly illustrate the solutions in the present application, the following will briefly introduce the drawings required for the description of the embodiments of the present application. Obviously, the following-described drawings are some embodiments of the present application. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0051] Figure 1 is an exemplary system architecture diagram to which the present application can be applied;

[0052] Figure 2 is a flowchart of an embodiment of a machine learning-based intelligent product selection management method according to the present application;

[0053] Figure 3 is a schematic structural diagram of an embodiment of a machine learning-based intelligent product selection management device according to the present application;

[0054] Figure 4 is a schematic structural diagram of an embodiment of a computer device according to the present application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0055] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those of ordinary skill in the technical field to which this application belongs; the terms used in the description of the embodiments of this application herein are only for the purpose of describing specific embodiments, and are not intended to limit this application; the terms "including" and "having" and any variations thereof in the description and claims of this application and the above drawings are intended to cover non-exclusive inclusion. The terms "first", "second", etc. in the description and claims of this application or the above drawings are used to distinguish different objects, rather than to describe a specific order.

[0056] References to "embodiments" in this specification mean that particular features, structures, or characteristics described in connection with the embodiments can be included in at least one embodiment of the present application. The phrase appears in various places in the specification and does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment mutually exclusive of other embodiments. Those skilled in the art will understand explicitly and implicitly that the embodiments described herein can be combined with other embodiments.

[0057] To enable those skilled in the art to better understand the solutions of this application, the technical solutions in the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings.

[0058] As Figure 1 shown, the system architecture 100 may include a terminal device 101, a network 102, and a server 103. The terminal device 101 may be a laptop computer 1011, a tablet computer 1012, or a mobile phone 1013. The network 102 is a medium for providing a communication link between the terminal device 101 and the server 103. The network 102 may include various connection types, such as wired, wireless communication links, or fiber optic cables, etc.

[0059] The user can use the terminal device 101 to interact with the server 103 through the network 102 to receive or send messages, etc. Various communication client applications may be installed on the terminal device 101, such as a web browser application, a shopping application, a search application, an instant messaging tool, an email client, a social platform software, etc.

[0060] The terminal device 101 may be various electronic devices with a display screen and supporting web browsing. In addition to the laptop computer 1011, the tablet computer 1012, or the mobile phone 1013, the terminal device 101 may also be an e-book reader, an MP3 player (Moving Picture Experts Group Audio Layer III), an MP4 (Moving Picture Experts Group Audio Layer IV) player, a laptop portable computer, and a desktop computer, etc.

[0061] The server 103 may be a server providing various services, such as a background server supporting the pages displayed on the terminal device 101.

[0062] It should be noted that a machine learning-based intelligent product selection management method provided in the embodiments of this application is generally executed by the server. Correspondingly, a machine learning-based intelligent product selection management device is generally provided in the server.

[0063] It should be understood that Figure 1 the number of terminal devices, networks, and servers in Figure 1 is merely illustrative. According to implementation requirements, there can be any number of terminal devices, networks, and servers.

[0064] Continuing to refer to Figure 2 , a flowchart of an embodiment of an intelligent product selection management method based on machine learning according to the present application is shown. The intelligent product selection management method based on machine learning includes the following steps:

[0065] Step S201, based on the historical purchase data and browsing data of each customer, generate a customer preference portrait for each customer through a preset portrait generation algorithm.

[0066] In this embodiment, an electronic device (such as the server shown in Figure 1 ) on which an intelligent product selection management method based on machine learning runs can communicate with a terminal device through a wired connection method or a wireless connection method. It should be noted that the above wireless connection methods can include, but are not limited to, 3G / 4G / 5G connections, WiFi connections, Bluetooth connections, WiMAX connections, Zigbee connections, UWB (ultra wideband) connections, and other currently known or future-developed wireless connection methods.

[0067] Specifically, obtain the historical purchase data and browsing data of each customer. The historical purchase data can be the customer's past purchase records on the platform, and the browsing data can be the customer's behavior data of browsing products on the platform. These data can reflect the customer's preferences for different product categories, price ranges, brands, etc.

[0068] The historical purchase data and browsing data can be gradually processed according to a preset portrait generation algorithm to generate a customer preference portrait for each customer. The customer preference portrait is generated by analyzing the customer's historical purchase behavior and browsing behavior, aiming to capture the customer's consumption habits, interests, and needs. The portrait generation algorithm can include steps such as data cleaning, feature extraction, and behavior analysis, converting the historical data into a digital representation of the customer's preferences.

[0069] Step S202, obtain the product information of each product, and perform attribute embedding and knowledge graph embedding on the product information of each product to obtain a product portrait for each product.

[0070] Specifically, the product portrait is the characteristic information describing the product, which can include structured data such as the product's attributes, categories, brands, prices, etc., and unstructured data such as user-generated content (such as comments, tags, etc.).

[0071] The generation of product portraits is achieved through attribute embedding and knowledge graph embedding. Attribute embedding maps product attribute data (such as price, brand, etc.) into a low-dimensional space to obtain its feature representation; knowledge graph embedding constructs a knowledge graph by leveraging the association relationships between products (e.g., similar products, complementary products, etc.) to further enhance the representation of product features. In the knowledge graph, nodes can be products and edges can be the relationships between products. The product attribute data and the association relationships between products come from product information.

[0072] Step S203: Based on the obtained historical purchase data and browsing data, construct a customer-product interaction graph, which includes customer nodes and product nodes.

[0073] Specifically, according to the historical purchase data and browsing data of each customer, a customer-product interaction graph can be constructed. The customer-product interaction graph describes the relationship between customers and products. In this graph, nodes represent customers or products, and edges represent the interactions between customers and products (such as purchases, browsing, etc.). The graph structure helps to discover the implicit potential relationships between customers and products.

[0074] The customer-product interaction graph is an interaction network between users and products. For example, if a customer purchases a certain product, then in the customer-product interaction graph, connect the customer node and the product node to form an edge.

[0075] Step S204: Perform representation learning on the customer-product interaction graph to obtain the product node features of each product node.

[0076] Specifically, representation learning is to train each node (here, product nodes) in the customer-product interaction graph to have a low-dimensional vector representation, and these vectors can fully express the features of the nodes and the relationships between the nodes. Graph representation learning methods can include DeepWalk, node2vec, etc.

[0077] The vector representing the product node is called the product node feature. The product node feature of each product node is an embedded vector obtained through graph learning methods, and these feature vectors can capture the relationships and similarities with other product nodes.

[0078] Step S205: Generate the product comprehensive features of each product according to the product portraits of each product and the product node features of their corresponding product nodes.

[0079] Specifically, the product comprehensive features combine the attributes of the product (such as brand, price, type, etc., which can come from the product portrait of the product) with the product node features obtained from graph representation learning to form a multi-dimensional feature representation of the product. This feature representation can reflect both the basic attributes of the product and the similarities or relationships with other products.

[0080] In step S206, the customer preference portrait of the target customer and the comprehensive product features of each product are respectively combined to obtain combined features, and the combined features are input into the matching model to obtain the degree of interest between the target customer and each product.

[0081] Specifically, the customer preference portrait of the target customer and the comprehensive product features of each product are combined (such as splicing, weighting, etc.) to form a complete representation to comprehensively reflect the relationship between the target customer and each product. The combined features obtained after combination are input into the trained matching model (such as a neural network model, a support vector machine, etc.), and the matching model calculates the degree of interest between the target customer and each product. The higher the degree of interest, the more likely the customer is to be interested in the product.

[0082] In step S207, target products are determined among the products according to the obtained degree of interest, and product recommendations are made to the target customer according to the determined target products.

[0083] Specifically, according to the degree of interest of the target customer in each product, products with higher degrees of interest are selected as target products. These target products may be the satisfaction points of the customer's potential needs. Making product recommendations to the target customer according to the target products helps to improve the purchase conversion rate of the customer and the accuracy of product selection and product recommendation.

[0084] In this embodiment, customer preference portraits of each customer are generated through historical purchase data and browsing data, and attribute embedding and knowledge graph embedding are performed on the product information of each product to generate high-quality product portraits; a customer-product interaction graph is constructed and representation learning is performed, further enhancing the implicit correlation representation between product nodes; the product portraits of each product and the product node features of the corresponding product nodes are combined, and the complex relationship between the customer and the product is transformed into comprehensive product features represented in multiple dimensions; the customer preference portrait and the comprehensive product features are input into the matching model to accurately calculate the degree of interest, thereby realizing personalized recommendation, improving the accuracy of product selection and recommendation, and increasing the conversion rate of product recommendation and customer satisfaction.

[0085] Further, the above step S201 may include: based on the historical purchase data of each customer, calculating the customer similarity between each customer through a collaborative filtering algorithm, and determining the similar customers of the target customer among each customer according to the obtained customer similarity, where the target customer comes from each customer; determining the preferred products for purchase of each similar customer according to the historical purchase data of each similar customer, and constructing the first product set of the target customer according to each preferred product for purchase; based on the browsing data of each customer, calculating the product similarity between each product through a content-based recommendation algorithm; determining the browsing preferred products similar to the currently browsed product of the target customer according to the obtained product similarity, and adding each browsing preferred product to the first product set to obtain the second product set of the target customer; generating the customer preference portrait of the target customer according to the second product set, historical purchase data, and browsing data of the target customer, and obtaining the customer preference portraits of each customer.

[0086] Specifically, through the collaborative filtering algorithm, according to the historical purchase data of the target customer and other customers, the customer similarity between the target customer and other customers is calculated. The calculation of customer similarity can adopt user-based collaborative filtering, and similar customers are found by comparing purchase behaviors (such as purchased products, purchase frequencies, etc.).

[0087] The collaborative filtering algorithm is a recommendation algorithm that mainly makes recommendations by analyzing the similarity between users or the similarity between items. The user-based collaborative filtering method predicts the products that a user may be interested in according to the similarity between the user and other users.

[0088] According to the calculated customer similarity, find other customers whose purchase behaviors are most similar to that of the target customer. The purchase preferences of these similar customers can recommend suitable products for the target customer. Extract the products they prefer to purchase from the purchase history data of the similar customers, that is, the preferred products for purchase, and determine the extracted preferred products for purchase as the first product set of the target customer. The preferred products for purchase are inferred based on the purchase preferences of the similar customers and can reflect the potential interests of the target customer.

[0089] Based on the browsing data of the target customer, adopt a content-based recommendation algorithm to calculate the similarity between the currently browsed product of the customer and other products. This step focuses on the attributes of the product itself, such as category, brand, price, etc. Based on the product similarity, select the products with higher similarity from other products similar to the currently browsed product of the target customer as the browsing preferred products, and add them to the first product set to obtain the second product set. These products reflect the current interests and needs of the target customer.

[0090] Integration of the historical purchase data, browsing data of the target customers, and the second product set generates a customer preference profile of the target customers. It describes the degree of interest and purchase propensity of the target customers for products in different categories. In one embodiment, the customer preference profile may further include the customer information of the target customers.

[0091] In this embodiment, by combining the historical purchase data and browsing data, recommendations can be generated based on the past behaviors of customers, and the content to be recommended can be adjusted according to the current browsing preferences, expanding the diversity of recommendations and facilitating more personalized and accurate product recommendations; the real-time browsing data can timely adapt to the changes in customer interests and provide more accurate recommendations.

[0092] Further, after the step of obtaining the second product set of the target customers, it may further include: processing the historical purchase data of each customer through an association rule mining algorithm to obtain product purchase association information; adding the purchased associated products to the second product set according to the product purchase association information.

[0093] Specifically, association rule mining is a technique in data mining used to discover relationships between different things. It discovers the relationships or "rules" between "item sets" from a large amount of data. The most common algorithms are the Apriori algorithm and the FP-growth algorithm, which can mine rules such as "people who buy A usually also buy B". This algorithm can be applied to scenarios such as retail and recommendation systems to discover patterns and relationships hidden in the data.

[0094] By analyzing the historical purchase data of all customers and using the association rule mining algorithm, potential purchase relationships between products are discovered to obtain product purchase association information. For example, if a large number of customers also buy "mobile phones" when they buy "headphones", then there is a certain purchase association between "headphones" and "mobile phones".

[0095] The product purchase association information can be a rule such as "when A is purchased, B is often purchased". Each rule can be accompanied by a certain support (indicating the frequency of the rule appearing in the purchase data) and confidence (indicating the probability when the rule holds), and this information helps to evaluate the strength and reliability of the association rule.

[0096] According to the obtained product purchase association information, add the associated products to the second product set of the target customer. For example, if the target customer browses or has purchased a certain product, the system will recommend some associated products according to the association rules of this product in the historical purchase data. For example, if the customer has browsed a smart watch and the purchase history shows a strong association between "smart watch" and "sports shoes", then add the sports shoes to the second product set, thereby updating the second product set to form a richer product set. This can include not only the current products that the customer is interested in, but also the related products that the customer may be interested in, further improving the accuracy and diversity of the recommendation.

[0097] In this embodiment, by processing the historical purchase data of each customer through the association rule mining algorithm and introducing the product purchase association information, it is possible to expand on the existing product recommendations, recommend some potentially relevant but unviewed or unpurchased products to the target customer, improve the diversity and coverage of the recommendation results, and enhance the accuracy of the recommendation.

[0098] Further, the steps of performing attribute embedding and knowledge graph embedding on the product information of each product to obtain the product portrait of each product may include: preprocessing the unstructured data in the product information of each product to obtain the structured data of each product; extracting attribute information from the structured data of each product to obtain the structured attribute data of each product; mapping the structured attribute data of each product into an attribute embedding representation respectively; based on the attribute embedding representation of each product, determining the association relationship between each product through similarity calculation and clustering algorithm, and constructing a product knowledge graph according to the obtained association relationship; fusing the attribute embedding representation and the knowledge graph embedding representation of each product to obtain the product portrait of each product.

[0099] Specifically, the product information has unstructured data (for example, product descriptions, user reviews, pictures, etc.). First, perform preprocessing operations such as text processing and image processing to convert this information into structured data. For example, the text in the product description can extract keywords, sentiment analysis, and entity recognition through natural language processing technology, and the pictures can extract features through computer vision algorithms and be converted into structured data.

[0100] Extract the attribute information of the product from the structured data of each product to obtain the structured attribute data of each product, such as brand, category, price, size, etc.

[0101] Map the structured attribute data of each product into an attribute embedding representation. For example, for information such as the brand, type, and size of a product, convert them into vector representations through an embedding layer. The purpose of embedding is to transform the discrete attributes of a product into a low-dimensional representation with semantics, capture the semantic relationships between attributes, and facilitate subsequent calculations and analyses. This process is usually learned through a neural network to obtain better representation capabilities.

[0102] Use a similarity calculation method to calculate the similarity between different products based on the attribute embedding representations of the products. For example, cosine similarity, Manhattan distance, etc. can be used as metrics to calculate the similarity between product vectors, so as to determine which products have similarity and correlation relationships in terms of attributes.

[0103] Clustering algorithms can also be used to group each product. For example, through K-means clustering, products with similar characteristics can be divided into the same category, and the products in each category have correlation relationships.

[0104] Based on the obtained correlation relationships, a product knowledge graph can be constructed, which contains information such as the similarity relationships and correlation relationships between products. A knowledge graph is a knowledge base represented by a graphical structure, where nodes represent entities (such as products) and edges represent the relationships between entities. In a product recommendation system, the knowledge graph builds an association network between products by integrating various product relationships (such as similarity, purchase association, etc.) to help the recommendation system better understand the relationships between products.

[0105] After constructing the knowledge graph, fuse the attribute embedding representation of the product with the knowledge graph embedding representation of the product in the knowledge graph. The fusion methods can be concatenation, weighted average, etc. Through fusion, a product portrait of each product is obtained, which contains the basic attribute information of the product and also combines the relationship and similarity information between it and other products.

[0106] In this embodiment, attribute information is extracted from product information, the correlation relationships between products are determined based on the attribute embedding representations of each product, and a product knowledge graph is constructed according to the correlation relationships, revealing the potential semantic relationships and connections between products; the attribute embedding representations and knowledge graph embedding representations of each product are comprehensively embedded to generate a comprehensive and accurate product portrait, which not only includes the basic attributes of the product, but also combines the similarity and correlation information between products, can better understand the complex relationships between products, and is beneficial to improving the relevance and accuracy of subsequent product recommendations.

[0107] Further, step S204 may include: generating a node sequence on the customer-product interaction graph through a random walk algorithm; inputting the node sequence into a skip-gram model to generate product node features of each product node in the customer-product interaction graph by maximizing the node co-occurrence probability.

[0108] Specifically, based on the interaction data such as the purchase history data and browsing records of each customer, a product interaction graph is constructed. In this graph, the nodes include customer nodes and product nodes.

[0109] The edges represent the interaction behaviors of customers with products (such as purchase or browsing). The edges between products reflect the degree of their association in customer interactions.

[0110] Perform a random walk on the customer-product interaction graph through the random walk algorithm to generate a node sequence. The random walk algorithm is a graph algorithm. Each node selects an adjacent node with a certain probability as the next node to visit. This method simulates the process of "wandering" in the graph and is widely used in fields such as graph embedding and social network analysis. In a recommendation system, random walk can be used to generate a node sequence of a product interaction graph, thereby revealing the potential associations between products. In this application, the algorithm starts from a certain product node and randomly selects an adjacent node (i.e., a related product) as the next node to visit. By generating multiple node sequences through multiple iterations, the model can capture the mutual relationships and similarities between products.

[0111] The node sequence refers to a series of nodes (product nodes in this application) generated through the random walk algorithm. The arrangement order of these nodes reflects their relationships and interaction frequencies in the graph. Through the node sequence, the potential associations between different products can be revealed.

[0112] Input the node sequence into the Skip-gram model. The Skip-gram model is a model based on word embedding (Word2Vec) in natural language processing and is used to learn the vector representation of vocabulary. In a recommendation system, the Skip-gram model is used to learn node (product) features. Skip-gram generates an embedding representation for each node by maximizing the probability of adjacent node co-occurrence. In a graph embedding task, it can capture the semantic relationships between nodes.

[0113] In this application, the goal of the Skip-gram model is to learn the embedding vector (i.e., product node feature) of each product node by maximizing the node co-occurrence probability. This vector can be used to represent the semantic relationship of the product in the interaction graph. The model continuously adjusts the embedding vector so that nodes that frequently appear together in the sequence have similar vector representations, reflecting their similarities in the interaction graph.

[0114] The generated product node features not only reflect the static information of the product (such as category, price, etc.), but also contain the dynamic association information learned through customer behavior and interaction relationships.

[0115] In this embodiment, by using the random walk algorithm and the Skip-gram model, the feature representation of product nodes can capture the association between products more accurately. This kind of feature learned through customer behavior and interaction data can more accurately reflect the potential similarity of products, thereby enhancing the accuracy of the recommendation system's understanding and prediction of customer preferences.

[0116] Further, the above intelligent product selection management method based on machine learning may further include: for each product node in the customer product interaction graph, determining the neighbor nodes of the product node; calculating the node similarity between the product node and each neighbor node; determining the attention weights of each neighbor node according to the obtained node similarity; aggregating the product node features of each neighbor node according to the obtained attention weights to obtain aggregated neighbor information; and updating the product node features of the product node according to the aggregated neighbor information.

[0117] Specifically, for each product node in the customer product interaction graph, find the nodes directly connected to it in the customer product interaction graph, and these nodes are the neighbor nodes of the target node. Neighbor nodes usually refer to those products that have a strong connection with the target product in the customer's purchase, browsing or other interaction behaviors. For example, a certain product A may be neighbors with products B, C, D, etc. because they are often purchased or browsed by the same batch of customers.

[0118] Calculate the similarity for each pair of product nodes and their neighbor nodes, such as through cosine similarity or Jaccard similarity based on customer interaction. The higher the similarity, the closer the association between these two products in customer interaction.

[0119] According to the calculated node similarity, assign an attention weight to each neighbor node. The attention weight reflects the importance of the neighbor node in the update of the target product node features. Neighbor nodes with high similarity will have a higher weight, and neighbor nodes with low similarity will have a lower weight.

[0120] Use the attention weights to perform weighted summation on the features of the neighbor nodes to obtain a new feature representation, that is, aggregated neighbor information. This weighted feature can better capture the influence of neighbor nodes in the recommendation, thereby providing more accurate product recommendations.

[0121] Update the product node features of the product node according to the aggregated neighbor information. After the update, the product node features not only reflect its own information but also incorporate the information of its related neighbor nodes, which can help the system more accurately understand and recommend products that the user may be interested in.

[0122] In this embodiment, determine the neighbor nodes of the product node, calculate the node similarity between the product node and each neighbor node, determine the attention weights of each neighbor node according to the node similarity, and by introducing the features of the neighbor nodes and aggregating these features weighted according to the node similarity and the attention mechanism, it is beneficial to deeply explore the potential relationships between products. The product node features of the product are more comprehensively represented, can more accurately reflect the potential interests of users, capture the deep-level relationships between products, and are beneficial to providing more personalized and accurate product recommendations.

[0123] Further, the above step S205 may include: for each product, extract the product embedding vector of the product from the product portrait of the product; merge the product embedding vector with the product node features of the product node corresponding to the product to obtain the comprehensive product features of the product.

[0124] Specifically, for each product, extract the product embedding vector of the product from the previously generated product portrait. It is a low-dimensional vector representation generated by embedding the various information of the product, capturing various characteristics of the product, such as the static characteristics of the product (such as brand, category, price), and may also incorporate the user's behavior data (such as purchase history, browsing records).

[0125] Then obtain the product node features of the product node corresponding to the product, and merge the product embedding vector with the product node features. Such a combination is to achieve a comprehensive integration of information. The product node features may include the basic information and historical behavior data of the product, and the product embedding vector captures the potential relationships between the product and other products and users in a more complex way. When merging, these two types of information can be combined by concatenation, weighting, etc. to form a more multi-dimensional comprehensive feature representation, that is, the comprehensive product features of the product.

[0126] The comprehensive product features are the final representation used by the product in the recommendation system, can more accurately describe each product, combine static attributes and dynamically generated behavior features, can provide richer information for the subsequent recommendation model, and help the system more accurately recommend to users.

[0127] In this embodiment, by combining the product embedding vector of the product with the product node features, the generated comprehensive product features can more comprehensively reflect the multi-dimensional information of the product. Such fused features enable the recommendation system to more accurately understand the potential relationship between the product and the user, capture both the explicit attributes and implicit relationships of the product, enhance the depth of understanding of the product, provide more accurate personalized recommendations, and improve the accuracy of recommendations and user satisfaction.

[0128] The embodiments of this application can acquire and process relevant data based on artificial intelligence technology. Among them, artificial intelligence (AI) uses digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use knowledge to obtain the best results in theory, methods, technologies, and application systems.

[0129] Artificial intelligence basic technologies generally include technologies such as sensors, dedicated artificial intelligence chips, cloud computing, distributed storage, big data processing technologies, operation / interaction systems, and mechatronics. Artificial intelligence software technologies mainly include several major directions such as computer vision technology, robotics, biometric technology, speech processing technology, natural language processing technology, and machine learning / deep learning.

[0130] Those of ordinary skill in the art can understand that all or part of the processes in the methods of the above embodiments can be completed by instructing relevant hardware through computer-readable instructions. These computer-readable instructions can be stored in a computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above methods. Among them, the aforementioned storage medium can be a non-volatile storage medium such as a magnetic disk, an optical disk, a read-only memory (ROM), or a random access memory (RAM), etc.

[0131] It should be understood that although the steps in the flowchart of the accompanying drawings are shown in sequence according to the arrows, these steps do not necessarily have to be executed in the order indicated by the arrows. Unless there is a clear indication in this article, the execution of these steps has no strict order limit, and they can be executed in other orders. Moreover, at least some of the steps in the flowchart of the accompanying drawings may include multiple sub-steps or multiple stages. These sub-steps or stages do not necessarily have to be executed at the same time, but can be executed at different times. Their execution order does not necessarily have to be sequential, but can be executed alternately or in turn with at least a part of other steps or sub-steps or stages of other steps.

[0132] For further reference Figure 3 to the above Figure 2For the implementation of the method described above, this application provides an embodiment of an intelligent product selection management device based on machine learning. This device embodiment corresponds to Figure 2 the method embodiment shown above, and this device can be specifically applied to various electronic devices.

[0133] As Figure 3 shown, an intelligent product selection management device 300 based on machine learning described in this embodiment includes: a preference generation module 301, a product generation module 302, an interaction construction module 303, a feature generation module 304, a comprehensive generation module 305, a combination calculation module 306, and a product recommendation module 307, where:

[0134] The preference generation module 301 is configured to generate a customer preference portrait for each customer based on the historical purchase data and browsing data of each customer through a preset portrait generation algorithm.

[0135] The product generation module 302 is configured to obtain the product information of each product, and perform attribute embedding and knowledge graph embedding on the product information of each product to obtain a product portrait of each product.

[0136] The interaction construction module 303 is configured to construct a customer-product interaction graph according to the obtained historical purchase data and browsing data. The customer-product interaction graph includes customer nodes and product nodes.

[0137] The feature generation module 304 is configured to perform representation learning on the customer-product interaction graph to obtain product node features of each product node.

[0138] The comprehensive generation module 305 is configured to generate comprehensive product features of each product according to the product portrait of each product and the product node features of its corresponding product node.

[0139] The combination calculation module 306 is configured to combine the customer preference portrait of the target customer and the comprehensive product features of each product respectively to obtain combined features, and input each combined feature into a matching model to obtain the interest degree between the target customer and each product.

[0140] The product recommendation module 307 is configured to determine a target product among each product according to the obtained interest degree, and perform product recommendation for the target customer according to the determined target product.

[0141] In this embodiment, customer preference portraits of each customer are generated through historical purchase data and browsing data. Product information of each product is subjected to attribute embedding and knowledge graph embedding to generate high-quality product portraits. A customer-product interaction graph is constructed and representation learning is performed, further enhancing the implicit correlation representation between product nodes. The product portraits of each product and the product node features of the corresponding product nodes are combined to transform the complex relationship between customers and products into the comprehensive product features represented in multiple dimensions. The customer preference portraits and the comprehensive product features are input into a matching model to accurately calculate the degree of interest, thereby realizing personalized recommendation, improving the accuracy of product selection and recommendation, and increasing the conversion rate of product recommendation and customer satisfaction.

[0142] In some optional implementation manners of this embodiment, the preference generation module 301 may include: a customer determination sub-module, a purchase determination sub-module, a similarity calculation sub-module, a browsing determination sub-module, and a preference generation sub-module, where:

[0143] The customer determination sub-module is configured to calculate the customer similarity between each customer based on the historical purchase data of each customer through a collaborative filtering algorithm, and determine the similar customers of the target customer among each customer according to the obtained customer similarity. The target customer comes from each customer.

[0144] The purchase determination sub-module is configured to determine the purchase preference products of each similar customer according to the historical purchase data of each similar customer, and construct the first product set of the target customer according to each purchase preference product.

[0145] The similarity calculation sub-module is configured to calculate the product similarity between each product based on the browsing data of each customer through a content-based recommendation algorithm.

[0146] The browsing determination sub-module is configured to determine the browsing preference products similar to the current browsing product of the target customer according to the obtained product similarity, and add each browsing preference product to the first product set to obtain the second product set of the target customer.

[0147] The preference generation sub-module is configured to generate the customer preference portrait of the target customer according to the second product set, historical purchase data, and browsing data of the target customer, and obtain the customer preference portraits of each customer.

[0148] In this embodiment, by combining historical purchase data and browsing data, recommendations can be generated based on the past behaviors of customers, and the content to be recommended can be adjusted according to the current browsing preferences, expanding the diversity of recommendations and facilitating the realization of more personalized and accurate product recommendations. The real-time browsing data can timely adapt to the changes in customer interests and provide more accurate recommendations.

[0149] In some alternative implementation manners of this embodiment, the preference generation module 301 may further include: an association determination sub-module and an association addition sub-module, where:

[0150] The association determination sub-module is configured to process the historical purchase data of each customer through an association rule mining algorithm to obtain product purchase association information.

[0151] The association addition sub-module is configured to add purchase-associated products to the second product set according to the product purchase association information.

[0152] In this embodiment, by processing the historical purchase data of each customer through an association rule mining algorithm and introducing product purchase association information, it is possible to expand on the existing product recommendations, recommend some potentially relevant but unviewed or unpurchased products to target customers, enhance the diversity and coverage of the recommendation results, and improve the accuracy of the recommendation.

[0153] In some alternative implementation manners of this embodiment, the product generation module 302 may include: a preprocessing sub-module, an attribute extraction sub-module, an attribute mapping sub-module, a relationship determination sub-module, and an embedding fusion sub-module, where:

[0154] The preprocessing sub-module is configured to preprocess the unstructured data in the product information of each product to obtain the structured data of each product.

[0155] The attribute extraction sub-module is configured to extract attribute information from the structured data of each product to obtain the structured attribute data of each product.

[0156] The attribute mapping sub-module is configured to map the structured attribute data of each product into an attribute embedding representation respectively.

[0157] The relationship determination sub-module is configured to determine the association relationship between each product based on the attribute embedding representation of each product through similarity calculation and a clustering algorithm, and construct a product knowledge graph according to the obtained association relationship.

[0158] The embedding fusion sub-module is configured to fuse the attribute embedding representation of each product and the knowledge graph embedding representation to obtain the product portrait of each product.

[0159] In this embodiment, attribute information is extracted from the product information, and the association relationship between each product is determined based on the attribute embedding representation of each product. A product knowledge graph is constructed according to the association relationship, revealing the potential semantic relationships and connections between products; the attribute embedding representation of each product and the knowledge graph embedding representation are comprehensively embedded to generate a comprehensive and accurate product portrait, which not only includes the basic attributes of the product but also combines the similarity and association information between products, can better understand the complex relationships between products, and is conducive to improving the relevance and accuracy of subsequent product recommendations.

[0160] In some alternative implementation manners of this embodiment, the feature generation module 304 may include: a sequence generation sub-module and a feature determination sub-module, where:

[0161] The sequence generation sub-module is configured to generate a node sequence on the customer-product interaction graph through a random walk algorithm.

[0162] The feature determination sub-module is configured to input the node sequence into a skip-gram model to generate product node features of each product node in the customer-product interaction graph by maximizing the node co-occurrence probability.

[0163] In this embodiment, by using the random walk algorithm and the Skip-gram model, the feature representation of product nodes can more accurately capture the correlation between products. Such features learned from customer behavior and interaction data can more accurately reflect the potential similarity of products, thereby enhancing the accuracy of the recommendation system's understanding and prediction of customer preferences.

[0164] In some alternative implementation manners of this embodiment, the feature generation module 304 may further include: a neighbor determination sub-module, a similarity calculation sub-module, a weight determination sub-module, an aggregation determination sub-module, and a feature update sub-module, where:

[0165] The neighbor determination sub-module is configured to determine neighbor nodes of a product node for each product node in the customer-product interaction graph.

[0166] The similarity calculation sub-module is configured to calculate the node similarity between a product node and each neighbor node.

[0167] The weight determination sub-module is configured to determine the attention weight of each neighbor node according to the obtained node similarity.

[0168] The aggregation determination sub-module is configured to aggregate the product node features of each neighbor node according to the obtained attention weight to obtain aggregated neighbor information.

[0169] The feature update sub-module is configured to update the product node features of a product node according to the aggregated neighbor information.

[0170] In this embodiment, determining the neighbor nodes of a product node, calculating the node similarity between a product node and each neighbor node, determining the attention weight of each neighbor node according to the node similarity, introducing the features of neighbor nodes, and weighted aggregating these features according to the node similarity and the attention mechanism is beneficial to deeply explore the potential relationship between products. The product node feature representation of products is more comprehensive, can more accurately reflect the potential interests of users, capture the deep relationship between products, and is beneficial to providing more personalized and accurate product recommendations.

[0171] In some alternative implementation manners of this embodiment, the comprehensive generation module 305 may include: a vector extraction sub-module and a merging sub-module, where:

[0172] The vector extraction sub-module is configured to extract the product embedding vector of each product from the product portrait of the product.

[0173] The merging sub-module is configured to merge the product embedding vector with the product node features of the product node corresponding to the product to obtain the comprehensive product features of the product.

[0174] In this embodiment, by combining the product embedding vector of the product with the product node features, the generated comprehensive product features can more comprehensively reflect the multi-dimensional information of the product. Such fused features enable the recommendation system to more accurately understand the potential relationship between the product and the user, can simultaneously capture the explicit attributes and implicit relationships of the product, improve the depth of understanding of the product, provide more accurate personalized recommendations, and enhance the accuracy of the recommendation and user satisfaction.

[0175] To solve the above technical problems, an embodiment of the present application also provides a computer device. Specifically, please refer to Figure 4 , Figure 4 which is the basic structural block diagram of the computer device in this embodiment.

[0176] The computer device 4 includes a memory 41, a processor 42, and a network interface 43 that are communicatively connected to each other through a system bus. It should be noted that only the computer device 4 with the memory 41, the processor 42, and the network interface 43 is shown in the figure. However, it should be understood that it is not required to implement all the shown components, and more or fewer components can be alternatively implemented. Among them, those skilled in the art of the present technology can understand that a computer device here is a device that can automatically perform numerical calculations and / or information processing according to pre-set or stored instructions, and its hardware includes but is not limited to a microprocessor, an application specific integrated circuit (ASIC), a field-programmable gate array (FPGA), a digital signal processor (DSP), an embedded device, etc.

[0177] The computer device may be a desktop computer, a notebook, a palm computer, a cloud server, and other computing devices. The computer device can perform human-computer interaction with the user through a keyboard, a mouse, a remote control, a touchpad, a voice control device, or other means.

[0178] The memory 41 includes at least one type of readable storage medium, which includes flash memory, hard disk, multimedia card, card-type memory (such as SD or DX memory, etc.), random access memory (RAM), static random access memory (SRAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), programmable read-only memory (PROM), magnetic memory, magnetic disk, optical disc, etc. In some embodiments, the memory 41 may be an internal storage unit of the computer device 4, such as the hard disk or memory of the computer device 4. In other embodiments, the memory 41 may also be an external storage device of the computer device 4, such as a plug-in hard disk, Smart Media Card (SMC), Secure Digital (SD) card, FlashCard, etc. equipped on the computer device 4. Of course, the memory 41 may also include both the internal storage unit and the external storage device of the computer device 4. In this embodiment, the memory 41 is generally used to store the operating system and various application software installed on the computer device 4, such as computer-readable instructions of an intelligent product selection management method based on machine learning. In addition, the memory 41 may also be used to temporarily store various data that have been output or will be output.

[0179] In some embodiments, the processor 42 may be a central processing unit (CPU), a controller, a microcontroller, a microprocessor, or other data processing chips. The processor 42 is generally used to control the overall operation of the computer device 4. In this embodiment, the processor 42 is used to run the computer-readable instructions stored in the memory 41 or process data, such as running the computer-readable instructions of the intelligent product selection management method based on machine learning.

[0180] The network interface 43 may include a wireless network interface or a wired network interface, and the network interface 43 is generally used to establish a communication connection between the computer device 4 and other electronic devices.

[0181] The computer device provided in this embodiment can execute the above-mentioned intelligent product selection management method based on machine learning. Here, the intelligent product selection management method based on machine learning may be the intelligent product selection management method based on machine learning in the above-mentioned various embodiments.

[0182] In this embodiment, customer preference portraits of each customer are generated through historical purchase data and browsing data, and attribute embedding and knowledge graph embedding are performed on the product information of each product to generate high-quality product portraits; a customer-product interaction graph is constructed and representation learning is carried out to further enhance the implicit correlation representation between product nodes; the product portraits of each product and the product node features of the corresponding product nodes are combined to transform the complex relationship between customers and products into a multi-dimensional product comprehensive feature; the customer preference portrait and the product comprehensive feature are input into a matching model to accurately calculate the degree of interest, so as to realize personalized recommendation, improve the accuracy of product selection and recommendation, and increase the conversion rate of product recommendation and customer satisfaction.

[0183] The present application also provides another implementation manner, that is, to provide a computer-readable storage medium storing computer-readable instructions that can be executed by at least one processor, so that the at least one processor executes the steps of a machine learning-based intelligent product selection management method as described above.

[0184] In this embodiment, customer preference portraits of each customer are generated through historical purchase data and browsing data, and attribute embedding and knowledge graph embedding are performed on the product information of each product to generate high-quality product portraits; a customer-product interaction graph is constructed and representation learning is carried out to further enhance the implicit correlation representation between product nodes; the product portraits of each product and the product node features of the corresponding product nodes are combined to transform the complex relationship between customers and products into a multi-dimensional product comprehensive feature; the customer preference portrait and the product comprehensive feature are input into a matching model to accurately calculate the degree of interest, so as to realize personalized recommendation, improve the accuracy of product selection and recommendation, and increase the conversion rate of product recommendation and customer satisfaction.

[0185] The present application provides a machine learning-based intelligent product selection management system, which is also a recommendation system, including a computer device and a computer-readable storage medium, and a machine learning-based intelligent product selection management device is provided in the computer device.

[0186] Through the description of the above embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus a necessary general hardware platform. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation manner. Based on such an understanding, the technical solution of the present application, in essence, or the part that contributes to the prior art can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk), including several instructions for causing a terminal device (which can be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in various embodiments of the present application.

[0187] Obviously, the embodiments described above are only a part of the embodiments of this application, rather than all the embodiments. The preferred embodiments of this application are shown in the drawings, but they do not limit the patent scope of this application. This application can be implemented in many different forms. On the contrary, the purpose of providing these embodiments is to make the understanding of the disclosed content of this application more thorough and comprehensive. Although this application has been described in detail with reference to the foregoing embodiments, for those skilled in the art, they can still modify the technical solutions described in the foregoing specific embodiments, or perform equivalent replacements on some of the technical features. Any equivalent structure made by using the content of this application's specification and drawings, directly or indirectly applied in other related technical fields, is similarly within the scope of patent protection of this application.

Claims

1. An intelligent product selection management method based on machine learning, characterized in that: The steps include: Based on the historical purchase data and browsing data of each customer, a customer preference profile of each customer is generated by a preset profile generation algorithm; Obtain product information of each product, and embed attributes and knowledge graphs of the product information of each product to obtain product portraits of each product; Constructing a customer-product interaction graph based on the obtained historical purchase data and browsing data, wherein the customer-product interaction graph includes customer nodes and product nodes; Performing representation learning on the customer-product interaction graph to obtain product node features of each product node; Generate comprehensive product features of each product according to the product profile of each product and the product node features of its corresponding product node; Combining the customer preference profile of the target customer and the comprehensive product features of each product to obtain combined features, and inputting each combined feature into a matching model to obtain the interest degree between the target customer and each product; Determine a target product from among the products according to the obtained interest level, and recommend products to the target customer according to the determined target product; The step of performing representation learning on the customer-product interaction graph to obtain product node features of each product node comprises: Generate a node sequence on the customer-product interaction graph by a random walk algorithm; Inputting the node sequence into a skip-gram model to generate product node features of each product node in the customer-product interaction graph by maximizing the node co-occurrence probability; The method further comprises: For each product node in the customer-product interaction graph, determining a neighbor node of the product node; Calculating the node similarity between the product node and each neighbor node; Determine the attention weight of each neighbor node according to the obtained node similarity; Aggregating the product node features of each neighbor node according to the obtained attention weights to obtain aggregated neighbor information; The product node feature of the product node is updated according to the aggregated neighbor information.

2. According to claim 1, a method for intelligent product selection management based on machine learning is characterized in that: The step of generating the customer preference portrait of each customer by a preset portrait generation algorithm based on the historical purchase data and browsing data of each customer includes: Based on the historical purchase data of each customer, the customer similarity between the customers is calculated by a collaborative filtering algorithm, and similar customers of the target customer are determined from among the customers according to the obtained customer similarity, wherein the target customer comes from the customers; Determine the purchase preference products of each similar customer according to the historical purchase data of each similar customer, and construct the first product set of the target customer according to each purchase preference product; Based on the browsing data of each customer, calculating the product similarity between each product by a content-based recommendation algorithm; Determine browsing preference products similar to the current browsing product of the target customer according to the obtained product similarity, and add each browsing preference product to the first product set to obtain a second product set of the target customer; A customer preference profile of the target customer is generated based on the second product set, historical purchase data and browsing data of the target customer, and a customer preference profile of each customer is obtained.

3. The intelligent product selection management method based on machine learning according to claim 2 is characterized in that: After the step of obtaining the second product set of the target customer, the method further includes: Process the historical purchase data of each customer through an association rule mining algorithm to obtain product purchase association information; According to the product purchase association information, purchase-associated products are added to the second product set.

4. The intelligent product selection management method based on machine learning according to claim 1 is characterized in that: The step of embedding attributes and knowledge graphs into the product information of each product to obtain a product portrait of each product includes: Preprocessing the unstructured data in the product information of each product to obtain the structured data of each product; Extracting attribute information from the structured data of each product to obtain structured attribute data of each product; Mapping the structured attribute data of each product into attribute embedding representations respectively; Based on the attribute embedding representation of each product, determine the association relationship between the products through similarity calculation and clustering algorithm, and construct a product knowledge graph according to the obtained association relationship; The attribute embedding representation of each product and the knowledge graph embedding representation are fused to obtain the product portrait of each product.

5. The intelligent product selection management method based on machine learning according to claim 1 is characterized in that: The step of generating the comprehensive product features of each product according to the product portrait of each product and the product node features of its corresponding product node comprises: For each product, extract a product embedding vector of the product from the product portrait of the product; The product embedding vector is combined with the product node feature of the product node corresponding to the product to obtain the comprehensive product feature of the product.

6. An intelligent product selection management device based on machine learning, characterized in that: include: A preference generation module, used to generate a customer preference profile of each customer through a preset profile generation algorithm based on the historical purchase data and browsing data of each customer; A product generation module is used to obtain product information of each product, and embed attributes and knowledge graphs into the product information of each product to obtain a product profile of each product; An interaction construction module, used to construct a customer-product interaction graph based on the obtained historical purchase data and browsing data, wherein the customer-product interaction graph includes customer nodes and product nodes; A feature generation module, used for performing representation learning on the customer-product interaction graph to obtain product node features of each product node; A comprehensive generation module, used to generate comprehensive product features of each product according to the product profile of each product and the product node features of its corresponding product node; A combination calculation module, used to combine the customer preference profile of the target customer and the comprehensive product features of each product to obtain a combination feature, and input each combination feature into a matching model to obtain the interest degree between the target customer and each product; A product recommendation module, used to determine a target product from among the products according to the obtained interest level, and recommend products to the target customer according to the determined target product; The feature generation module is also used to generate a node sequence on the customer-product interaction graph by a random walk algorithm; input the node sequence into a skip-gram model to generate product node features of each product node in the customer-product interaction graph by maximizing the node co-occurrence probability; The feature generation module is also used to determine the neighbor nodes of each product node in the customer-product interaction graph; calculate the node similarity between the product node and each neighbor node; and determine the attention weight of each neighbor node according to the obtained node similarity; The product node features of the neighbor nodes are aggregated according to the obtained attention weights to obtain aggregated neighbor information; and the product node features of the product node are updated according to the aggregated neighbor information.

7. A computer device, characterized in that: It includes a memory and a processor, the memory stores computer-readable instructions, and the processor implements the steps of an intelligent product selection management method based on machine learning as described in any one of claims 1 to 5 when executing the computer-readable instructions.

8. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer-readable instructions, which, when executed by a processor, implement the steps of an intelligent product selection management method based on machine learning as described in any one of claims 1 to 5.

Images