Friend-space: Cluster-based users similar post friend recommendation technique in social networks

Abstract

Online social networking is a way of access and share the information with user friends. All the social networking sites like Facebook, twitter are provided the services. Parameters like life-style, interest, education, similarity or common things, mutual friends are considered for friend recommendation in social networks. In this paper we propose the best clusters based friend recommendation technique name as Friend-Space. For development of Friend-Space application four algorithms are implement and use in system. K-Means, Apriori, Ranking and Recommendation algorithms are developed for Friend-Space Application. Friend-space Cluster based application gives better performance in case of execution time for K-Means algorithm. In this paper we achieve the parameters like efficiency, effectiveness and execution time for number of executions. Recommendation algorithm implements for display the final output.

An improved Apriori–based algorithm for friends recommendation in microblog

“…According to the based type of model, collaborative filtering algorithm can be divided into 2 directions. The first one is memory based and the second is model based . Based on these ideas, Arain et al construct intelligent travel information platform based on location base services to predict user travel behavior from user‐generated GPS traces …”

“…The first one is memory based and the second is model based. 5,6,23 Based on these ideas, Arain et al construct intelligent travel information platform based on location base services to predict user travel behavior from user-generated GPS traces. [28][29][30]…”

“…Apriori algorithm is a classical algorithm for mining association rules for Boolean items that the core of the algorithm is to find frequent item sets by increasing the number of candidate item set. After the method was put forward, the researchers did a lot of supplement, some of these works made a hit in some areas . In spite of this, there is an improved need for this method, especially in computational efficiency and time.…”

“…[2][3][4] To recommend proper friends for various users, friend recommendation algorithms are in urgent need. Though friend recommendation technology has been widely spread, new ideas, methods, and concepts are emerging from different perspectives such as cluster-based users similar post friend recommendation technique, 5 real-time recommendation for microblogs, 6 using seed sets to find journalists, 7 semantic recommender for microblog users, 8 and copy adjustable incentive scheme in community-based socially aware networking. 9 Besides, some other critical methods are proposed as well.…”

“…After the method was put forward, the researchers did a lot of supplement, some of these works made a hit in some areas. 1,2,5,6,17,[21][22][23][24][25] In spite of this, there is an improved need for this method, especially in computational efficiency and time.…”