Improved Social Network User Recommendation System—The Machine Learning Approach

[摘要] In this paper, we propose a recommendation system for social media users which makes recommendations on the basis of the user profile information and the contents posted by users-the bio-aware algorithm. Text mining techniques are used to pre-process the words before feeding it to an LDA model which handles the topic and feature extraction. In this way, we determine the similarity between users based on their interests. We further trained a machine learning model which is able to identify and score the top interests of a particular social media user. Other users who share similar scores are shown as recommendations to each other.

[关键词] social network;people interaction;information searching;information extraction;bio-aware approach