IMPROVED FRIEND RECOMMENDATION SYSTEM FOR SOCIAL NETWORKING SITE THROUGH FP-GROWTH AND ANT COLONY OPTIMIZATION ALGORITHM

Recommender systems have established to be of high-quality useful resource in dealing with the issue of information overload by using enhancing the person experience through best recommendations, with the speedy improvement of clever metropolis services, social network plays a higher role in giant areas with smart technology. Social Networking allow customers to create keywords such as tags in social tagging system to describe sources that are of activity to them, assisting to organize and share these assets with different users in the network, friend advice is important and inevitable in social network. however, many recommendation techniques of social networks are not necessarily steady with user's interests. in order to avoid the randomness and unreliability of friend recommendation in social community we proposed improved friend recommendation system for social networking website thru FP-growth and ant colony optimization algorithm, the outcomes of our test are weighted to adjusted for a higher accurate result, which ultimately forms recommendation lists for the target users. Finally, the experimental results on the Delicious dataset for friend suggestion show that the effectiveness and achieved accurate outcomes and a strong recommendation.