A Clustering Expert System using Particle Swarm Optimization and K-means++ for Journal Recommendation to Publish the Papers

Abstract

In this paper, an android expert system for recommending the suitable journal for publishing the researchers' papers has been presented. In so doing, the expectations of different journals for accepting an article and also the requests of papers' writers for choosing the journals have been examined. Language, quality, waiting time for judgment, waiting time for publication after acceptance, field, length of paper and cost are the system inputs and its output is the degree of suitability of journals for publishing a certain paper. This system includes a database of different journals and their parameters. It uses particle swarm optimization method and K-means++ algorithm for assessing and clustering the journals database and determines an index for every cluster of journals. The process for matching the paper with a cluster's index is done through fuzzy induction system. After choosing the most similar cluster, the paper is compared with all the journals of the cluster in the same way and the result including the most similar journals is presented. This system has been tested for journals and papers in the computer field indexed in Elsevier.