E-commerce commodity recommendation system based on artificial intelligence

Abstract

The invention relates to the technical field of e-commerce, and discloses an e-commerce commodity recommendation system based on artificial intelligence, and the system comprises a user information collection module which is used for obtaining the personal religious belief information of a user; the data processing module is used for processing the religious belief information and generating religious preference data of the user; the recommendation engine module is used for avoiding recommendation of religious taboo commodities when the commodities are recommended based on the religious preference data of the user; the commodity library is used for storing commodity information on the e-commerce platform; and the output module is used for providing the recommended commodity list filtered according to the religious preference data to the user.

Description

Technical Field

[0001] The present invention relates to the technical field of e-commerce, and particularly to an e-commerce product recommendation system based on artificial intelligence. Background Art

[0002] With the rapid development of e-commerce, product recommendation systems have become an indispensable part of e-commerce platforms. Product recommendation systems recommend products based on information such as users' historical behaviors, purchase preferences, browsing records, etc., using algorithms to improve users' shopping experiences and the platform's sales conversion rates. Existing product recommendation systems are widely used in major e-commerce platforms such as Amazon, Taobao, JD.com, etc. By collecting and analyzing users' behavior data, they can recommend products that users may be interested in, increasing the exposure rate and sales volume of products.

[0003] Existing product recommendation systems mainly make recommendations based on the following several methods: Collaborative filtering algorithm: This algorithm analyzes users' historical purchase or browsing records and uses the behaviors of other users with similar interests as the basis for recommendations. Collaborative filtering is divided into user-based collaborative filtering and item-based collaborative filtering.

[0004] Content-based recommendation: This method makes recommendations based on the feature information of products (such as category, brand, price, etc.). The system matches the attributes of products with the preferences in users' historical behaviors.

[0005] Hybrid recommendation method: Combines collaborative filtering and content-based recommendation. Through the collaborative work of multiple recommendation algorithms, it provides more accurate product recommendations.

[0006] However, although these existing recommendation systems can effectively improve the accuracy of product recommendations and user satisfaction, they often ignore users' religious belief factors. In certain specific cultural or religious backgrounds, users' religious beliefs have an important impact on product selection. For example, some religious users may be averse to food or liquor products containing pork ingredients; while some religious users may tend to choose vegetarian products and avoid products containing animal ingredients. Without considering these religious belief taboos, existing product recommendation systems may recommend products that conflict with users' religious beliefs, leading to user dissatisfaction, aversion, and even triggering religious or cultural sensitive issues. Summary of the Invention

[0007] In view of the deficiencies in the prior art, the present invention provides an e-commerce product recommendation system and method based on artificial intelligence. By comprehensively collecting users' religious belief information, deeply analyzing and combining the religious taboo labels of products, it provides product recommendations that conform to users' religious beliefs, enhances the user shopping experience, and avoids the adverse effects caused by religious belief conflicts.

[0008] To achieve the above object, the present invention adopts the following technical solutions: An e-commerce product recommendation system based on artificial intelligence, the system includes: A user information collection module for obtaining users' personal religious belief information; A data processing module for processing the religious belief information and generating users' religious preference data; A recommendation engine module that avoids recommending religious taboo products when recommending products based on users' religious preference data; A product library for storing product information on the e-commerce platform; An output module for providing the user with a list of recommended products filtered according to the religious preference data.

[0009] Preferably, the user information collection module includes: A religious belief information collection unit filled in by the user when registering or logging in; Or a judgment unit for judging users' religious belief information through users' behavior data (such as historical purchase records, browsing records).

[0010] Preferably, the user information collection module obtains users' religious belief information through the information filled in by the user when registering or logging in, specifically including: When the user registers an account, fills in religious belief-related fields through a form, and the fields include religious name, frequency of religious activities, religious taboos, etc.; When the user logs in and their information has not been updated, directly obtain the user's religious belief information by verifying the user's historical filling data; If the user does not provide religious belief information, prompt the user to fill it in, or judge the user's religious belief information through the user's historical behavior data.

[0011] Preferably, judge the user's religious belief by obtaining the user's regional information, specifically including: Obtain the user's regional information, which can be obtained through the user's IP address, geographical location, address information provided by the user when registering, or other geolocation means; Based on the regional information, use the pre-set association data between regions and religious beliefs to judge the user's religious belief as the main religious belief in that region.

[0012] Preferably, the user information collection module is used to obtain multi-dimensional data of the user, including the user's purchase record of goods, browsing content, interest tags, etc.; The data processing module is used to integrate the multi-dimensional data of the user with the regional religious custom data, specifically including: If the user's purchased goods, browsing content or interest tags conflict with the religious customs of the region, it is determined through a judgment mechanism that the user's religious belief is different from the main religious belief of the region; If the user shows interest in certain goods or activities related to a specific religion, the system adjusts the judgment of the user's religious belief according to the behavioral information.

[0013] Preferably, the user information collection module is used to obtain the content published by the user on a third-party platform (such as a social media platform, a news website, etc.) by linking to a third-party software interface, and the content includes religious festivals, prayer activities, religious gatherings, religious articles or videos shared by the user, etc.; The data processing module is used to analyze the published content and identify the religious-related information therein, including keywords, tags, sentiment tendencies, etc.; The judgment module is used to judge the user's religious belief according to the analysis result and assign a religious belief label to the user; The behavior analysis module is used to analyze the interaction behavior of the user on the third-party platform, including liking, commenting on, and sharing religious-related content, so as to further adjust the judgment result; The user feedback module is used to feedback the judgment result to the user, and the user can confirm or modify their religious belief information.

[0014] Preferably, the commodity library contains religious taboo tags of commodities, and the tags are preset by the e-commerce platform according to commodity attributes or classification information.

[0015] An e-commerce commodity recommendation method based on artificial intelligence includes the following steps: Obtain the religious belief information of the user; Generate the religious preference data of the user according to the religious belief information; When recommending commodities, based on the religious preference data of the user, screen out the commodities that conform to the religious belief and exclude the commodities that conflict with the religious belief; Recommend the screened commodities to the user.

[0016] Compared with the prior art, the present invention has the following beneficial effects: 1. Improve the user experience: fully consider the user's religious belief factor, avoid recommending commodities that conflict with the user's religious belief, enhance the user's trust and satisfaction with the recommendation system, and improve the shopping experience.

[0017] 2. Improve recommendation accuracy: By collecting users' religious belief information through multiple channels and dimensions and analyzing and processing it using advanced algorithms, it is possible to more accurately grasp users' needs and improve the accuracy of product recommendations.

[0018] 3. Enhance system adaptability: Provide multiple ways to collect users' religious belief information, as well as flexible data processing and recommendation algorithms, enabling the system to adapt to different user groups and complex and changing user behaviors, with stronger adaptability and scalability. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following will briefly introduce the drawings required in the embodiments. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0020] Figure 1 It is a schematic diagram of the overall process of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0021] The present invention will be further described in detail below with reference to the drawings.

[0022] The following description is used to disclose the present invention so that those skilled in the art can implement the present invention. The preferred embodiments described below are only examples, and those skilled in the art can think of other obvious variations. The basic principles defined in the following description can be used in other implementation schemes, variation schemes, improvement schemes, equivalent schemes, and other technical schemes without departing from the spirit and scope of the present invention. Embodiment

[0023] Please refer to Figure 1 , an e-commerce product recommendation system based on artificial intelligence, the system includes: A user information collection module for obtaining multi-dimensional data of users, including users' purchase product records, browsing content, interest tags, etc.; A data processing module for processing religious belief information, generating users' religious preference data, and integrating users' multi-dimensional data with regional religious custom data; A recommendation engine module that avoids recommending religious taboo products when recommending products based on users' religious preference data; A product library for storing product information on the e-commerce platform; An output module for providing a list of recommended products filtered according to religious preference data to users.

[0024] User Information Collection Module: Responsible for obtaining the user's personal religious belief information, including but not limited to the religious belief information collection unit filled in by the user during registration or login, and the judgment unit that determines the user's religious belief information through the user's behavior data (such as historical purchase records, browsing records). The collection methods are as follows: 1. Registration and Login Collection Method: When the user registers an account, they can fill in relevant fields related to religious beliefs through a form, such as religious name, frequency of religious activities, religious taboos, etc. When the user logs in, if the information has not been updated, the system directly obtains the religious belief information by verifying the user's historical filling data; if the user has not provided relevant information, the system prompts the user to fill it in or determines the user's religious belief information by relying on the user's historical behavior data.

[0025] 2. Region Information Inference Method: By obtaining the user's region information (which can be obtained through IP address, geographical location, address information provided by the user during registration, or other geolocation means), based on the pre-set association data between regions and religious beliefs, the user's religious belief is initially determined as the main religious belief in that region. At the same time, this module also obtains the user's multi-dimensional data, such as purchase records of goods, browsing content, interest tags, etc., and integrates them with the regional religious habit data. If the user's purchase, browsing behavior, or interest tags conflict with the regional religious habits, the system determines through a judgment mechanism that the user's religious belief may be different from the main religious belief in that region; if the user shows interest in specific religious-related goods or activities, the system adjusts the judgment of the user's religious belief accordingly.

[0026] 3. Third-Party Platform Data Acquisition Method: By linking to the third-party software interface, obtain the content published by the user on the third-party platform (such as social media platforms, news websites, etc.), such as religious festival sharing, prayer activity records, religious gathering information, religious articles or videos, etc. The data processing method for the data obtained from the third-party platform: The data processing module analyzes this content to identify religious-related information, including keywords, tags, sentiment tendencies, etc. The judgment module determines the user's religious belief based on the analysis results and assigns a religious belief label. The behavior analysis module further analyzes the user's interaction behavior on the third-party platform, such as liking, commenting, and sharing religious-related content, to adjust the judgment results. Finally, the user feedback module feeds back the judgment results to the user, and the user can confirm or modify them.

[0027] Data Processing Module: Used to integrate the user's multi-dimensional data with the regional religious habit data, specifically including: If the user's purchase of goods, browsing content, or interest tags conflict with the religious habits of the region, then through the judgment mechanism, it is determined that the user's religious belief is different from the main religious belief in that region; If a user shows interest in certain goods or activities related to a specific religion, the system adjusts the user's religious belief judgment based on this behavioral information; Process the collected religious belief information to generate the user's religious preference data. During the processing, not only consider the religious belief information directly provided by the user, but also comprehensively analyze the relevant data obtained through various means to ensure that the generated religious preference data accurately reflects the impact of the user's religious belief on commodity selection.

[0028] Among them, the data processing and fusion algorithm method of the data processing module is as follows: When fusing the user's multi-dimensional data with the regional religious custom data, a weighted fusion method is adopted. First, assign weights to different types of data. For example, the weight of the user's purchase record of goods is set to 0.4, the weight of the browsing content is set to 0.3, the weight of the interest label is set to 0.2, and the weight of the regional religious custom data is set to 0.1. Then, for each data dimension, quantify it according to its correlation with religious belief. For example, for the purchase record of goods, if a commodity related to a specific religion (such as an item of a certain religion) is purchased, the score in this dimension increases; if a commodity that conflicts with religious taboos is purchased, the score decreases. Finally, multiply the data scores of each dimension by the corresponding weights and add them up to obtain a comprehensive score, and judge the user's religious belief tendency according to this score. The specific formula is as follows: Comprehensive score = 0.4 × score of purchase record of goods + 0.3 × score of browsing content + 0.2 × score of interest label + 0.1 × score of regional religious custom data.

[0029] Recommendation engine module: Based on the user's religious preference data, avoid recommending religious taboo goods when recommending goods. This module adopts an algorithm that combines rules and machine learning. Based on the pre-set religious taboo rules, conduct a preliminary screening of the goods in the commodity library. The screening method is as follows: If the user is a religious person who has an aversion to food or liquor products containing pork ingredients, directly exclude the goods containing pork ingredients or liquor. Then, use machine learning algorithms (such as logistic regression, decision tree, etc.) to sort the remaining goods according to the user's religious preference data and other relevant information (such as purchase history, browsing records, etc.). When training the model, use the user's feedback (such as behaviors like purchase, click, ignore, etc.) on the recommended goods as labels, and continuously optimize the model parameters to improve the accuracy of the recommendation. The specific steps are as follows: 1. Data preprocessing: Encode and standardize the user's religious preference data and other relevant features so that they can be used as inputs for machine learning algorithms.

[0030] 2. Model training: Use historical user data for model training, and adjust the model parameters to enable the model to accurately predict the user's preference degree for goods.

[0031] 3. Product Sorting: Based on the trained model, score and sort the preliminarily screened products, and recommend the products with higher scores to users.

[0032] Product Library: Used to store product information on the e-commerce platform. The product library contains religious taboo labels for products, which are preset by the e-commerce platform according to product attributes or classification information, and the merchant applications are regularly reviewed and processed by the product information management team, and are updated in real time in combination with the industry information monitoring mechanism.

[0033] Output Module: Display the product list screened by the recommendation engine module to users in a simple and beautiful interface. During the display process, in addition to the basic information of the products, such as pictures, names, prices, etc., a brief description related to religious beliefs of the products can also be provided according to user needs, such as which religious belief requirements the product meets, to help users better understand the recommended content.

[0034] Steps of the Recommendation Method: S1. Obtain the religious belief information of the user: Comprehensively collect the religious belief information of the user through various methods of the above-mentioned user information collection module; S2. Generate the religious preference data of the user according to the religious belief information: Use the data processing module to deeply analyze and process the collected information to generate data that accurately reflects the user's religious belief preferences for product selection; S3. When recommending products, based on the religious preference data of the user, screen out the products that conform to the religious belief and exclude the products that conflict with the religious belief: The recommendation engine module screens products by using the optimized algorithm based on the generated religious preference data and in combination with the religious taboo labels in the product library; S4. Recommend the screened products to the user: Display the product recommendation list that conforms to the user's religious belief to the user through the output module.

[0035] Those skilled in the art should understand that the embodiments of the present invention described above and shown in the drawings are only examples and do not limit the present invention. The object of the present invention has been fully and effectively achieved. The functions and structural principles of the present invention have been shown and described in the embodiments, and the embodiments of the present invention can have any deformation or modification without departing from the said principles.

Claims

1. An e-commerce product recommendation system based on artificial intelligence, characterized in that: The system includes: a user information collection module for obtaining the user's personal religious belief information; A data processing module, used to process the religious belief information and generate the user's religious preference data; The recommendation engine module avoids recommending religiously taboo products based on the user's religious preference data; Product library, used to store product information on the e-commerce platform; The output module is used to provide the recommended product list filtered according to the religious preference data to the user.

2. The artificial intelligence-based e-commerce product recommendation system according to claim 1, characterized in that: The user information collection module includes: The religious belief information collection unit filled in by the user when registering or logging in; Or a judgment unit that judges a user's religious belief information through user behavior data (such as historical purchase records, browsing records).

3. The artificial intelligence-based e-commerce product recommendation system according to claim 2, characterized in that: The user information collection module obtains the user's religious belief information through the information filled in by the user when registering or logging in, specifically including: When registering an account, users fill in fields related to religious beliefs through a form, including the name of the religion, frequency of religious activities, religious taboos, etc.; When a user logs in and his / her information is not updated, the user's religious belief information is directly obtained by verifying the user's historical data; If the user does not provide information about religious beliefs, the user is prompted to fill it out, or the user's religious belief information is determined based on the user's historical behavior data.

4. The artificial intelligence-based e-commerce product recommendation system according to claim 2, characterized in that: Determine the user's religious beliefs by obtaining the user's region information, including: Obtain the user's region information, which can be obtained through the user's IP address, geographic location, address information provided by the user during registration, or other geolocation methods; Based on the region information, the user's religious belief is determined to be the main religious belief in the region using pre-set region and religious belief association data.

5. The artificial intelligence-based e-commerce product recommendation system according to claim 4, characterized in that: The user information collection module is used to obtain multi-dimensional data of users, including the user's purchase records, browsing content, interest tags, etc.; The data processing module is used to integrate the user's multi-dimensional data with the regional religious custom data, including: If the user's purchases, browsing content, or interest tags conflict with the religious practices of the region, the user's religious beliefs are judged to be different from the main religious beliefs in the region through a judgment mechanism; If a user shows interest in certain products or activities related to a specific religion, the system adjusts the user's religious belief judgment based on this behavioral information.

6. The artificial intelligence-based e-commerce product recommendation system according to claim 1, characterized in that: The user information collection module is used to obtain the content posted by users on third-party platforms (such as social media platforms, news websites, etc.) by linking to third-party software interfaces. The content includes religious festivals, prayer activities, religious gatherings, religious articles or videos shared by users, etc.; A data processing module is used to analyze the published content and identify religious related information therein, including keywords, tags, emotional tendencies, etc.; A judgment module is used to judge the user's religious belief based on the analysis results and assign a religious belief label to the user; Behavior analysis module, used to analyze users’ interactive behaviors on third-party platforms, including liking, commenting, and sharing religious-related content, to further adjust the judgment results; The user feedback module is used to feed back the judgment results to the user, and the user can confirm or modify his / her religious belief information.

7. The artificial intelligence-based e-commerce product recommendation system according to claim 1, characterized in that: in, The product library includes religious taboo labels for products, and the labels are pre-set by the e-commerce platform according to product attributes or classification information.

8. An e-commerce product recommendation method based on artificial intelligence, characterized in that: The following steps are involved: Obtain information about the user's religious beliefs; generating religious preference data of users based on religious belief information; When recommending products, we filter out products that match religious beliefs based on the user’s religious preference data, and exclude products that conflict with religious beliefs; Recommend the filtered products to users.

Images