Mining scientific trends based on topics in conference call for papers

Abstract

Ever since analyzing scientific topics and evolution of technology have become vital for researchers, academics, funding institutes and research administration departments, there is a crucial need to mine scientific trends to fill this appetite more rigorously. In this paper, we procured a novel Call for Papers (CFPs) dataset in order to analyze scientific evolution and prestige of conferences that set scientific trends using scientific publications indexed in DBLP. Using ACM CSS, 1.3 million publications that appear in 146 data mining conferences are mapped into different thematic areas by matching the terms that appear in publication titles with ACM CSS. In recent years, an attempt termed as Topic Detection and Tracking (TDT) [1] is made to find the solution for the problem of "well-awareness" on this dynamic data. As conference ranking has been made by different forums on the basis of mixed indicators1. ERA2 ranks Australia's higher education research institutions. The major contributions of this paper are as follows: (i) compilation of CFPs dataset, (ii) identification of topics and keywords from CFP corpus, and (iii) measure the impact of these extracted hot topics from CFPs.