Using k-means++ algorithm for researchers clustering

The Clustering of researchers based on publications is one of identifying community of researchers in a research environment. The researchers will know the relationships with other researchers regarding the similarity of topics and disciplines of publications based on the research community. The clustering will perform the extraction and analysis of the concept, topic detection and clustering of researchers. The attributes of data that can be obtained through the publications and characteristics of researchers on social networks that have been formed on the relations among researchers. The extraction and analysis of document, has two stages: extraction of keywords using keyphrase automatic rapid extraction (RAKE), and extraction concept of using latent semantic analysis (LSA). Clustering concept use k-means ++ algorithm. The last process, clustering of researchers is formed by feature extraction of social networking analysis,also use the k-means ++ algorithm. Applications for clustering researchers will be presented in the table containing information on researchers in each of these clusters.