MARST: Multi-Agent Recommender System for e-Tourism Using Reputation Based Collaborative Filtering

摘要

This paper presents a Multi-Agent Recommender system for e-Tourism (MARST) for recommending tourism services to the users. This system uses Reputation based Collaborative Filtering (RbCF) algorithm that augments reputation to existing Collaborative approach for generating relevant recommendations and to handle cold-start new user problem in tourism domain. The structure of a tourist product is more complex than a book or a movie and hence user profile modeling for these systems is much harder than most of other applications domains like books or movies. Moreover the frequency of activities and rating in tourism domain is also much smaller than in most of the other domains. This increases the complexity in designing and development of Recommender Systems in tourism domain. An attempt has been made in this paper to generate relevant services for a user in tourism domain using reputation based collaborative filtering. Most of the existing Recommender systems focus on one service at a time, whereas the proposed system incorporates three services (hotels, places to visit and restaurants) at a single place to ease the searching of information at one place only. The prototype of MARST has been designed and developed using various JAVA technologies and its performance was evaluated using precision, recall and F_1 metrics.