Search

CN-121981791-A - Commodity matching method, system, equipment and program product based on big data

CN121981791ACN 121981791 ACN121981791 ACN 121981791ACN-121981791-A

Abstract

The invention belongs to the technical field of computers, and aims to provide a commodity matching method, a system, equipment and a program product based on big data. The invention discloses a commodity matching method based on big data, which comprises the steps of obtaining historical behavior data of a user, storing the historical behavior data of the user in a big data storage system, analyzing the historical behavior data of the user in the big data storage system through a deep learning technology, establishing a relation model between the user and a commodity, obtaining user demand information of a designated user, and carrying out demand prediction on the designated user based on the relation model by adopting a machine learning algorithm to obtain a matched commodity corresponding to the designated user. The invention can realize the deep analysis of the user data, can meet the personalized requirements of the user, is beneficial to improving the precision and personalized level of commodity recommendation, and has stronger timeliness of the recommendation result.

Inventors

  • LI PINGXIU

Assignees

  • 广东水利电力职业技术学院（广东省水利电力技工学校）

Dates

Publication Date
20260505
Application Date
20240322

Claims (10)

  1. 1. The commodity matching method based on big data is characterized by comprising the following steps: Acquiring user historical behavior data and storing the user historical behavior data in a big data storage system; analyzing the historical behavior data of the user in the big data storage system through a deep learning technology, and establishing a relationship model between the user and the commodity; And obtaining user demand information of the appointed user, and carrying out demand prediction on the appointed user based on the relation model by adopting a machine learning algorithm to obtain a matched commodity corresponding to the appointed user.
  2. 2. The method for matching commodity according to claim 1, wherein said user historical behavior data comprises historical commodity purchase records, browsing behavior data and commodity feature data.
  3. 3. The commodity matching method according to claim 1, wherein said analyzing said user historical behavior data in said big data storage system by a deep learning technique and creating a relationship model between a user and commodity comprises: Preprocessing the user history behavior data stored in the big data storage system to obtain preprocessed user history behavior data; Dividing the preprocessed user history behavior data into a training set and a testing set; And acquiring a pre-constructed initial relationship model, training the initial relationship model by using the training set, and evaluating the initial relationship model by using the testing set so as to obtain a trained relationship model.
  4. 4. The commodity matching method according to claim 3, wherein preprocessing the user historical behavior data in the big data storage system to obtain preprocessed user historical behavior data comprises: and sequentially performing data cleaning, feature extraction and normalization on the user historical behavior data in the big data storage system to obtain preprocessed user historical behavior data.
  5. 5. The commodity matching method according to claim 3, wherein said initial relationship model is a recurrent neural network model.
  6. 6. The commodity matching method based on big data according to claim 1, wherein said user demand information is bound with user identification information of said specified user, and correspondingly, a machine learning algorithm is adopted to predict demand of said specified user based on said relation model, so as to obtain a matching commodity corresponding to said specified user, and the method comprises the steps of: Extracting specified user historical behavior data matched with the user identification information from the big data storage system; Inputting the historical behavior data of the appointed user into the relation model to obtain a user characteristic vector corresponding to the appointed user; Extracting required commodity characteristic information in the user required information, and extracting all initial matching commodity information corresponding to the required commodity characteristic information from the big data storage system; obtaining the preference of the appointed user for all kinds of initial matching commodity information according to the user feature vector; And obtaining initial matching commodity information corresponding to the highest preference of the appointed user according to the preference of the appointed user for various initial matching commodity information, and outputting the initial matching commodity information corresponding to the highest preference of the appointed user as the matching commodity of the appointed user.
  7. 7. The commodity matching method according to claim 6, wherein said designated user For the initial matching commodity information The preference degree of (2) is as follows: ; In the formula, Representing the specified user Is used to determine the user characteristic vector of (a), Representing initially matched merchandise information Is a commodity feature vector of (a), Representing the first of the big data storage systems The user feature vector, e, of each user is a natural constant.
  8. 8. The big data based commodity matching system is characterized by being used for realizing the big data based commodity matching method according to any one of claims 1 to 7, and comprises the following components: the data collection module is used for obtaining user historical behavior data and storing the user historical behavior data in the big data storage system; the association model building module is in communication connection with the data collection module and is used for analyzing the historical behavior data of the user in the big data storage system through a deep learning technology and building a relationship model between the user and the commodity; And the personalized recommendation module is in communication connection with the association model building module and is used for acquiring user demand information of the appointed user, and carrying out demand prediction on the appointed user based on the relation model by adopting a machine learning algorithm to obtain a matched commodity corresponding to the appointed user.
  9. 9. An electronic device characterized by comprising: a memory for storing computer program instructions, and A processor for executing the computer program instructions to perform the operations of the big data based commodity matching method according to any one of claims 1 to 7.
  10. 10. A computer program product comprising a computer program or instructions, characterized in that the computer program or instructions, when executed by a computer, implement the big data based commodity matching method according to any of claims 1 to 7.

Description

Commodity matching method, system, equipment and program product based on big data Technical Field The invention belongs to the technical field of computers, and particularly relates to a commodity matching method, a system, equipment and a program product based on big data. Background With the rapid development of electronic commerce, the variety of commodity types and the number of commodity are continuously increased, so that when a user faces a large number of commodity, the best matching commodity meeting the requirement of the user is difficult to find. Therefore, various prior art technologies which rely on basic recommendation algorithms and rule engines for commodity matching appear at present, but in the process of using the prior art, the inventor finds that at least the following problems exist in the prior art, namely the traditional commodity matching method often fails to fully mine user behavior data and cannot conduct deep personalized analysis, which leads to limitation of accuracy and user experience of commodity recommendation results, and part of commodity matching methods fail to timely adapt to changes of user interests and market trends, lead to lag of recommendation results and influence on acquisition of new and popular commodities by users. Disclosure of Invention The invention aims to solve the technical problems at least to a certain extent, and provides a commodity matching method, a system, equipment and a program product based on big data. In order to achieve the above purpose, the present invention adopts the following technical scheme: In a first aspect, the present invention provides a commodity matching method based on big data, including: Acquiring user historical behavior data and storing the user historical behavior data in a big data storage system; analyzing the historical behavior data of the user in the big data storage system through a deep learning technology, and establishing a relationship model between the user and the commodity; And obtaining user demand information of the appointed user, and carrying out demand prediction on the appointed user based on the relation model by adopting a machine learning algorithm to obtain a matched commodity corresponding to the appointed user. The invention can realize the deep analysis of the user data, can meet the personalized requirements of the user, is beneficial to improving the precision and personalized level of commodity recommendation, and has stronger timeliness of the recommendation result. In the implementation process, the method and the device utilize a big data technology, deep analysis is carried out on historical behavior information of a user such as historical purchasing records, browsing behaviors and commodity characteristics of the user through a deep learning technology so as to establish a correlation model between the user and the commodity, and meanwhile, a machine learning algorithm is adopted to accurately predict personalized demands of the user based on the correlation model so as to recommend commodity information which accords with preferences of the user. Therefore, the invention can be widely applied to the electronic commerce platform, is beneficial to meeting the personalized demands of users, further improves the shopping experience of the users, increases the shopping satisfaction of the users, and is beneficial to improving the sales of the electronic commerce platform. In one possible design, the user historical behavior data includes historical merchandise purchase records, browsing behavior data, and merchandise characterization data. In one possible design, the analysis of the user historical behavior data in the big data storage system by a deep learning technology and the establishment of a relation model between a user and a commodity comprises: Preprocessing the user history behavior data stored in the big data storage system to obtain preprocessed user history behavior data; Dividing the preprocessed user history behavior data into a training set and a testing set; And acquiring a pre-constructed initial relationship model, training the initial relationship model by using the training set, and evaluating the initial relationship model by using the testing set so as to obtain a trained relationship model. In one possible design, preprocessing the user historical behavior data in the big data storage system to obtain preprocessed user historical behavior data, including: and sequentially performing data cleaning, feature extraction and normalization on the user historical behavior data in the big data storage system to obtain preprocessed user historical behavior data. In one possible design, the initial relationship model employs a recurrent neural network model. In one possible design, the user requirement information is bound with user identification information of the appointed user, correspondingly, a machine learning algorithm is adopted to predict the requirement of the