CN108090810A - A kind of Products Show system based on big data - Google Patents

Abstract

The invention discloses a product recommendation system based on big data, which includes a server and a client. The server includes a source data collection module, a data processing module, a recommendation information generation module and a database, a source data collection module, a data processing module, and a recommendation information generation module. The modules are all connected to the database; the input end of the source data acquisition module is connected to the output end of the client, the output end is connected to the data processing module, the output end of the data processing module is connected to the input end of the recommended information generation module, and the output of the recommended information generation module is The data processing module uses the preference acquisition technology to process the data input by the source data acquisition module to obtain the user's search preference; the recommendation information generation module generates recommendation information according to the user search preference output by the data processing module, and The recommended information is returned to the client for display. This system realizes the recommendation of high-quality information based on user preference data.

Description

A product recommendation system based on big data

technical field

The invention relates to the technical field of data processing, in particular to a product recommendation system based on big data.

Background technique

In this era of information explosion, consumers are often at a loss when faced with numerous choices, unknown fields, and overloaded information; however, at the same time, content producers (such as merchants) are also struggling to find suitable users and find the most convenient channels, and the best tool to solve these two types of contradictions is the recommendation system.

The recommendation system originated from the search system, and there are a lot of similar technologies on the underlying system. However, in terms of corresponding user needs and application scenarios, the recommendation system is further away from the user: on the one hand, when the user's needs are specific and clear, Search; when the user's needs are unclear or difficult to express, make demand recommendations. On the other hand, when users need to find recognized and popular content in a certain field, search; when users need to find personalized content, make recommendations. In many scenarios, it is difficult to convert the user's personalized needs into short and clear query words, such as "I want to find a nearby restaurant that suits my taste and is not expensive at noon today", which is very common but very common. Difficult to use query words to express clearly. The recommendation system can just fill this gap, and dig out the personalized needs in depth according to the historical behavior of users, so as to realize its usefulness.

At present, recommendation systems generally include content recommendation algorithms and collaborative filtering algorithms. Among them, the content recommendation algorithm mainly analyzes the content information generated by users, digs out the interests and hobbies of users, and the connections between users, and finally completes the purpose of recommending products to users. However, both the content recommendation algorithm and the collaborative filtering algorithm have the problem of poor recommendation effect.

Contents of the invention

The purpose of the present invention is to provide a product recommendation system based on big data to improve the quality of recommended information.

In order to achieve the above object, the technical solution adopted in the present invention is: including a server and a client, wherein the server includes a source data acquisition module, a data processing module, a recommended information generation module and a database, a source data acquisition module, a data processing module, and a recommended information The generating modules are all connected to the database;

The input end of the source data acquisition module is connected to the output end of the client, the output end is connected to the data processing module, the output end of the data processing module is connected to the input end of the recommended information generation module, and the output end of the recommended information generation module is connected to the client;

Wherein, the data processing module uses the preference acquisition technology to process the data input by the source data acquisition module to obtain the user's search preference;

The recommendation information generation module generates recommendation information according to the user search preferences output by the data processing module, and returns the recommendation information to the client for display.

Among them, the client includes a user-defined resource display unit, a user interest list display unit, a hot commodity recommendation information display unit, and an automatic commodity recommendation information display unit;

The user-defined resource display unit is used to display the resource information added by the user;

The user interest list display unit is used to display the user interest list;

The hot commodity recommendation information display unit is used to display the hot commodity recommendation information;

The product automatic recommendation information display unit is used to display product recommendation information.

Wherein, the source data collection module includes a user scoring data collection unit, a user feedback data collection unit, a user network data collection unit, and a user demographic characteristic parameter collection unit;

The user rating data collection unit is used to collect user rating data on commodities;

The user feedback data collection unit is used to collect user feedback data on commodity use;

The user network data acquisition unit is used to collect usage data generated on the user network;

The user demographic feature parameter collection unit is used to collect the statistical number of users who purchase commodities.

Wherein, the data processing module includes a user preference analysis unit, a social network structure analysis unit and a contextual user preference analysis unit;

The user preference analysis unit is used to analyze user preference according to the data collected by the source data collection module, and obtain user preference data;

The social network structure analysis unit is used to analyze the social network structure according to the data collected by the source data collection module, and obtain the social network structure data;

The contextual user preference analysis unit is configured to analyze the contextual user preference data according to the data collected by the source data collection module.

Wherein, the recommendation information generation module includes a matrix decomposition-based recommendation information generation unit, an implicit feedback-based recommendation information generation unit, a socialized recommendation information generation unit, and a group recommendation information generation unit;

The recommendation information generation unit based on matrix decomposition is used to process user preference data in a matrix decomposition manner to generate recommendation information;

The recommendation information generation unit based on implicit feedback is used to generate recommendation information according to user feedback data;

The social recommendation information generation unit is used to generate recommendation information according to the social network structure data;

The group recommendation information generation unit is used to generate combined recommendation information based on the recommendation information generated by the matrix decomposition-based recommendation information generation unit, the implicit feedback-based recommendation information generation unit, and the socialized recommendation information generation unit.

Compared with the prior art, the present invention has the following technical effects: the present invention mainly digs out the interests and hobbies of the users by analyzing the content information generated by the users and the connections between users, and can track and learn the interests of the users, Feature information such as preferences and personality, real-time and accurate discovery of user needs. Based on this information, the system judges the agricultural products that users want to buy most, and recommends them, and finally completes the purpose of recommending products to users. The information recommended by the present invention recommends high-quality information aiming at the personalized access request of consumers. At the same time, for data services, in response to consumers' diverse access requests, data services should be able to describe services in a flexible way and dynamically generate new data services that meet the needs.

Description of drawings

Below in conjunction with accompanying drawing, the specific embodiment of the present invention is described in detail:

Fig. 1 is a structural representation of a product recommendation system based on big data in the present invention;

Fig. 2 is the structural representation of server among the present invention;

Fig. 3 is a schematic structural diagram of the recommendation algorithm adopted in the present invention.

Detailed ways

In order to further illustrate the features of the present invention, please refer to the following detailed description and accompanying drawings of the present invention. The accompanying drawings are for reference and description only, and are not intended to limit the protection scope of the present invention.

As shown in Figure 1, this embodiment discloses a product recommendation system based on big data, including a server 10 and a client 20, wherein the server 10 includes a source data collection module 11, a data processing module 12, and a recommendation information generation module 13 And database 14, source data acquisition module 11, data processing module 12, recommended information generating module 13 are all connected with database 14;

The input end of the source data acquisition module 11 is connected with the output end of the client 20, the output end is connected with the data processing module 12, the output end of the data processing module 12 is connected with the input end of the recommendation information generation module 13, and the recommendation information generation module 13 The output end is connected with the client 20;

Wherein, the data processing module 12 uses the preference acquisition technology to process the data input by the source data acquisition module 11 to obtain the user's search preference;

The recommended information generating module 13 generates recommended information according to the user's search preferences output by the data processing module 12, and returns the recommended information to the client 20 for display.

Wherein, the client 20 includes a user-defined resource display unit, a user interest list display unit, a hot commodity recommendation information display unit, and an automatic commodity recommendation information display unit;

The user-defined resource display unit is used to display the resource information added by the user, such as some websites and products that the user has collected;

The user interest list display unit is used to display the user's interest list, such as nothing, clothing, beauty, etc.;

The hot product recommendation information display unit is used to display the hot product recommendation information, mainly to display the current hot products, such as newly launched products such as Apple mobile phones, and recent special sales products such as discounted products;

The product automatic recommendation information display unit is used to display product recommendation information.

In this embodiment, the client searches through the client 20, the source data acquisition module 11 in the server 10 collects the input information of the client in the client 20, and the data processing module 12 processes the information by using the preference acquisition technology to obtain the user's interest. preferences, and recommend information based on user preferences, which greatly improves the quality of information recommendation and improves customer satisfaction.

Further, as shown in FIG. 2 , the source data collection module 11 includes a user rating data collection unit, a user feedback data collection unit, a user network data collection unit, and a user demographic feature parameter collection unit;

The user rating data collection unit is used to collect user rating data on commodities, such as specific evaluation scores or evaluation star ratings, etc.;

The user feedback data collection unit is used to collect user feedback data on the use of the product, for example, the user feedback data from aspects such as price, performance, and appearance after using the product;

The user network data acquisition unit is used to collect the usage data generated by the user on the Internet, such as the web page data browsed by the user online, the product data browsed or collected by the user, etc.;

The user demographic feature parameter collection unit is used to collect the statistical number of users who purchase commodities.

Further, the data processing module 12 includes a user preference analysis unit, a social network structure analysis unit and a contextual user preference analysis unit;

The user preference analysis unit is used to analyze user preferences according to the data collected by the source data collection module 11, and obtain user preference data;

The social network structure analysis unit is used to analyze the social network structure according to the data collected by the source data collection module 11, and obtain the social network structure data;

The contextual user preference analysis unit is used to analyze the contextual user preference data according to the data collected by the source data collection module 11, where the contextual user preference data mainly refers to the correlation between users, and the user's data has been indirectly obtained according to the correlation between users. preference.

Further, the recommendation information generation module 13 includes a matrix decomposition-based recommendation information generation unit, an implicit feedback-based recommendation information generation unit, a socialized recommendation information generation unit, and a group recommendation information generation unit;

The recommendation information generation unit based on matrix decomposition is used to process user preference data in a matrix decomposition manner to generate recommendation information;

The recommendation information generation unit based on implicit feedback is used to generate recommendation information according to user feedback data;

The social recommendation information generation unit is used to generate recommendation information according to the social network structure data;

The group recommendation information generation unit is used to generate combined recommendation information based on the recommendation information generated by the matrix decomposition-based recommendation information generation unit, the implicit feedback-based recommendation information generation unit, and the socialized recommendation information generation unit.

Further, the server 10 also includes an evaluation module 15, which is mainly used to evaluate the quality of the current recommendation information, which includes a real-time evaluation unit, an accuracy evaluation unit, a diversity evaluation unit and a novelty evaluation unit.

It should be noted that, as shown in FIG. 3 , the recommendation information generating module 13 in this embodiment is different from the content recommendation algorithm or the collaborative recommendation algorithm in the prior art, but uses the content recommendation algorithm and the collaborative recommendation algorithm comprehensively, and at the same time It avoids the defects of content recommendation algorithm and collaborative recommendation algorithm. By analyzing the content information generated by users, it can dig out the user's interests and hobbies, as well as the connection between users, and finally complete the purpose of product recommendation for users.

A product recommendation system based on big data disclosed by the present invention has the following beneficial effects:

(1) By tracking and learning user interests, preferences, personality and other characteristic information, real-time and accurate discovery of user needs and adjustments can be made. Based on this information, the system can determine the agricultural products that users want to buy most, and realize high-quality information recommendation.

(2) For data services, the system can describe services in a flexible way and dynamically generate new data services that meet the needs of consumers in response to diversified access requests.

(3) The service application of this system combines applications related to data characteristics to realize information query, analysis and visualization.

The above descriptions are only preferred embodiments of the present invention, and are not intended to limit the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present invention shall be included in the protection of the present invention. within range.

Claims (5)

1. A product recommendation system based on big data, characterized in that it includes a server (10) and a client (20), wherein the server (10) includes a source data acquisition module (11), a data processing module (12), The recommended information generation module (13) and the database (14), the source data acquisition module (11), the data processing module (12), and the recommended information generation module (13) are all connected to the database (14); The input end of the source data collection module (11) is connected to the output end of the client (20), the output end is connected to the data processing module (12), the output end of the data processing module (12) is connected to the recommendation information generation module (13) The input end is connected, and the output end of the recommended information generation module (13) is connected to the client (20); Among them, the data processing module (12) uses preference acquisition technology to process the data input by the source data acquisition module (11) to obtain the user's search preference; The recommended information generation module (13) generates recommended information according to the user search preferences output by the data processing module (12), and returns the recommended information to the client (20) for display. 2. The product recommendation system based on big data according to claim 1, characterized in that: the client (20) includes a user-defined resource display unit, a user interest list display unit, a hot commodity recommendation information display unit, and Commodity automatic recommendation information display unit; The user-defined resource display unit is used to display the resource information added by the user; The user interest list display unit is used to display the user interest list; The hot commodity recommendation information display unit is used to display the hot commodity recommendation information; The product automatic recommendation information display unit is used to display product recommendation information. 3. The product recommendation system based on big data according to claim 1, characterized in that: the source data collection module (11) includes a user rating data collection unit, a user feedback data collection unit, a user network data collection unit and user Demographic characteristic parameter acquisition unit; The user rating data collection unit is used to collect user rating data on commodities; The user feedback data collection unit is used to collect user feedback data on commodity use; The user network data acquisition unit is used to collect usage data generated on the user network; The user demographic feature parameter collection unit is used to collect the statistical number of users who purchase commodities. 4. The product recommendation system based on big data according to claim 3, characterized in that: the data processing module (12) includes a user preference analysis unit, a social network structure analysis unit and a contextual user preference analysis unit; The user preference analysis unit is configured to analyze user preference according to the data collected by the source data collection module (11), and obtain user preference data; The social network structure analysis unit is used to analyze the social network structure according to the data collected by the source data collection module (11), and obtain the social network structure data; The contextual user preference analysis unit is configured to analyze the contextual user preference data according to the data collected by the source data collection module (11). 5. The product recommendation system based on big data according to claim 1, characterized in that: the recommendation information generation module (13) includes a matrix decomposition-based recommendation information generation unit, an implicit feedback-based recommendation information generation unit, A social recommendation information generation unit and a group recommendation information generation unit; The recommendation information generation unit based on matrix decomposition is used to process user preference data in a matrix decomposition manner to generate recommendation information; The recommendation information generation unit based on implicit feedback is used to generate recommendation information according to user feedback data; The social recommendation information generation unit is used to generate recommendation information according to the social network structure data; The group recommendation information generation unit is used to generate combined recommendation information based on the recommendation information generated by the matrix decomposition-based recommendation information generation unit, the implicit feedback-based recommendation information generation unit, and the socialized recommendation information generation unit.