Project selection method based on self-adaptive active learning

Abstract

The invention discloses a project selection method based on self-adaptive active learning. The method comprises the steps of calculating the uncertainty of a candidate project; calculating the representativeness of the candidate project; and self-adaptively and dynamically selecting the project with highest information content according to the uncertainty and representativeness. By adopting the project selection method, the uncertainty and representativeness of the project can be comprehensively considered, and the project with the highest information content can be selected.

Description

technical field [0001] The invention relates to the technical field of recommendation systems, in particular to an item selection method based on adaptive active learning. Background technique [0002] In a collaborative filtering recommendation system, the key to solving the user cold start problem is how to quickly establish a new user's interest preference model. When the user initially uses the system, the method based on the active learning scoring guide actively selects some items for user evaluation to effectively obtain the user's personalized preference information. There are two considerations for selecting items to rate users: (1) users can obtain more rating data for users by rating items, and the more rating information, the more effective the recommendation system is; (2) not all rating information is equivalent, Some ratings can represent users' personalized information, while others cannot, so different active learning scoring guidance methods will bring dif...

AI Technical Summary

Problems solved by technology

For example, keep selecting popular items for user evaluation. Although more user rating data can be obtained, it is not very helpful for the system to obtain personalized preference information of users, because most users like popular items.

Images