Agent based mobile recommender system

Abstract

Social networking websites (SNS) are attaining a lot of recognition by researchers as they prove to be a reservoir of huge amount of data. These sites have revolutionized the world of technology by allowing their users to form a virtual world of their own. The main notion behind building the social networking websites is to bring people from all walks of life together, having some link with each other either on the base of their philosophies or ideas about life, or belonging to same occupational class, or sharing same thoughts and views, in a form of virtual community. SNSs provide a medium for people to remain in contact with their beloved ones, get global as well as local updates, and share their feelings with others. Additionally, these sites have also been used effectively for marketing and advertising almost every kind of products. Companies announce their pages while giving demonstrations of their items. Thus, SNS have basically emerged most common recommender engines. However, most of the times, unnecessary and irrelevant information are displayed on these social interaction pages which causes high data traffic and distract of user concentration. This issue becomes severe in case a user visits multiple social websites frequently. Thus, a personalized system is needed that could provide a single platform to make recommendations based on user's friends circle on different SNS sites. Different multi agent system architectures have been proposed to efficiently carryout different search and recommendation procedure but, little attention was given to the user interface. In this research, we solve this issue by the design and development of adaptable intelligent agent based interface. The objective is to intelligently present the personalize recommendations help a user to get what he needs from his social Web site in a blink of eye without wasting time in dealing with a complex search procedures.

Distributed recommender system using multi-agent for social networks

“…In a MAS environment, all agents have to communicate and coordinate with each other. Because of their ability of parallel processing, MAS is utilized in social recommender systems [62,63].…”

“…This system uses content-based similarity analysis of web documents. In [63], a mobile application that uses multi-agent system is presented in order to provide personalized and quick recommendations to the social network users.…”