Gray forecasting model-based recommending method and system

Abstract

The invention discloses a gray forecasting model-based recommending method and a gray forecasting model-based recommending system, and relates to the field of recommending systems. The method comprises the following steps of: establishing an inter-project similarity matrix; selecting a predetermined number of projects having high similarity with unrated projects of a target user from the rated projects of the target user according to the similarity matrix; ordering the scores of the selected projects according to the order of the similarity, from high to low, of the unrated projects to obtain a forecasting sequence; establishing a gray forecasting model according to the forecasting sequence; and forecasting the scores, forecast by the target user, of the unrated projects according to the gray forecasting model, and recommending the corresponding unrated project to the target user according to the forecasting score. By using the recommending method and the recommending system, very high recommending precision can still be ensured when user rating data is extremely sparse and high correlation exists among the data, and the recommending method and the recommending system have a very wide application prospect.

Description

technical field [0001] The invention relates to the technical field of recommendation systems, in particular to a recommendation method and system based on a gray prediction model. Background technique [0002] With the vigorous development of Web 2.0 technology and interactive media, people have gradually become active participants from passive web browsers in the past, and people's participation is constantly increasing. However, while the Internet brings us convenience, it also brings us into the era of information explosion. Information overload makes it difficult for people to obtain valuable content from the colorful network world, and makes a large amount of little-known information become "dark information" in the network. [0003] The traditional search engine based on information filtering technology lacks the consideration of personalization, and still cannot solve the problem of information overload well. In recent years, the recommendation system, as a very po...

AI Technical Summary

Problems solved by technology

[0007] The technical problem to be solved by the present invention is: how to provide a recommendation method and system based on a gray prediction model, so as to overcome the existing recommendation method and system that will lead to recommendation accuracy when the user rating data is extremely sparse and there is a serious correlation between the data. lower disadvantage

Images