A recommendation method based on active learning to solve commodity cold start problem

Abstract

The invention discloses an active learning-based recommendation method for solving a commodity cold-start problem. The method comprises the steps of 1, building a scoring model of a user for a commodity, and performing pre-training on the model through historical scoring data of the user for the commodity and attribute characteristics of the commodity; 2, for a new commodity, estimating whether different users perform scoring on the commodity or not and how many points the users give by using the scoring model in the step 1; 3, selecting the users to perform scoring on the new commodity according to a result in the step 2 so as to obtain scoring data of the new commodity; 4, performing re-training on the scoring model in the step 1 by utilizing the scoring data of the new commodity; and 5, predicting scores of unselected users for the new commodity by utilizing the re-trained scoring model, and performing commodity recommendation according to the scores. According to the method, user experience of all users is considered simultaneously, so that the fairness of a selection policy is ensured to a certain extent; finite user resources are fully utilized; and the commodity is effectively recommended to the user.

Description

technical field [0001] The invention relates to the field of recommendation systems, in particular to a recommendation method based on active learning to solve the cold start problem of commodities. Background technique [0002] The rapid development of Internet multimedia has produced a large amount of information. On the one hand, it meets the needs of users for information, but on the other hand, it is difficult for users to obtain useful content from a large amount of information (information overload), so it also reduces the user's demand for information. usage efficiency. Recommender systems are a useful approach to address the problem of information overload. It predicts the user's information needs by analyzing data such as the user's historical behavior, thereby directly recommending the information that the user may need to the user. [0003] At present, the recommendation system has been widely used in the recommendation applications of commodities, movies, musi...

AI Technical Summary

Problems solved by technology

The content-based recommendation algorithm uses similar information of product attributes to make recommendations, but products with similar attributes may have large quality differences, resulting in wrong recommendations

For example, the screenwriters of the movie Taken3 and the movie Taken are the same as many actors, so they are similar in terms of attributes, but users on the IMDB website have high scores for Taken, but not high scores for Taken3, so Take3 is recommended It is probably a wrong recommendation for users who like Taken

Images