A study on group recommender systems

Abstract

In recent years the most popular recommender systems are none other than the personalized tourist attraction recommender systems on social networks which use the personal profiles willingly provided by users on the social network and then recommend tourist attractions that the users may like based on their life experiences and trajectories With such recommender systems users can save time used to plan their travels ahead of time We have found that most of these services focus on the individual when making recommendations However most people have company when they travel and are not alone e g going picnics with family and watching movies with friends Furthermore looking at existing social websites (i e Facebook、Meetup、Foursquare) it has become a trend for check-in data to also include the accompanying members This motivates the studies on group recommendation which aims to recommend tourist attractions for a group of users In this paper we propose an algorithm named HMUR (Hybrid Method with Users Rating) to analyze the group information and user ratings and make group recommendations We conduct an experiments and the results show that the proposed algorithm is effective in making group recommendations and outperforms baseline methods significantly

Date of Award	2016 Aug 18
Original language	English
Supervisor	Chiang Lee (Supervisor)