Trust model based food quality evaluation method

Abstract

At present, with the accelerated pace of social life, the requirement of people on diet quality is higher and higher, and the difficulty in finding foods with satisfied quality by users is higher and higher. Most food recommendation systems all serve specific groups, so that user interests are often ignored. The invention, for personalized demands of users on foods in the recommendation systems, proposes a trust model based food quality evaluation method. The method comprises: analyzing user behavior data to analyze trust degree among the users from three perspectives of interest depth, span and interest timeliness; then analyzing trust degree of the foods by utilizing evaluation information of the users to the foods and the like; and finally evaluating the food quality by combining the trust degree among the users and the trust degree among the foods. Experimental results in a real recommendation system data set show that the method can improve the accuracy of personalized recommendation.

Description

technical field [0001] The invention belongs to the field of food quality safety, and in particular relates to a food quality evaluation method based on a trust degree model. Background technique [0002] With the popularization of the Internet and the extensive and in-depth application of e-commerce, people enjoy the convenience of online shopping but also fall into the dilemma of information overload. It is difficult for users to find the food they need in a large amount of food information. Although more and more recommender systems are playing the role of salespersons on e-commerce platforms, recommending products to users, helping users find what they need, and completing the purchase process smoothly. However, there are many types of food in China, and the classification is complex. Each food-related website has its own set of classification rules, resulting in a large amount of heterogeneous food-related information on the Internet. Thereby, it is more and more diffi...

AI Technical Summary

Problems solved by technology

To sum up, domestic researchers mainly focus on the two-dimensional code system and RFID and other technical methods to study food safety issues, and do not involve the establishment of a food quality recommendation model, so that it is impossible to effectively realize the intelligent analysis and diagnosis of food quality

Since there are many factors that affect food quality in food production, and the logical relationship between these factors is complex, it is very difficult to establish an effective detection or supervision system, and traditional methods such as neural networks are also difficult to achieve.

Images