Evaluation of H- and G-indices of Scientific Authors using Modified K-Means Clustering Algorithm

—In this paper I proposed modified K-means\udalgorithm as the means to assess scientific authors\udperformance by using their h,g-indices values. K-means\udsuffers from poor computational scaling and efficiency as\udthe number of clusters has to be supplied by the user. In\udthis work, I introduce a modification of K-means\udalgorithm that efficiently searches the data to cluster\udpoints by compute the sum of squares within each cluster\udwhich makes the program to select the most promising\udsubset of classes for clustering. The proposed algorithm\udwas tested on IRIS and ZOO data sets as well as on our\udlocal dataset comprising of h- and g-indices, which are\udthe prominent markers for scientific excellence of authors\udpublishing papers in various national and international\udjournals. Results from analyses reveal that the modified\udk-means algorithm is much faster and outperforms the\udconventional algorithm in terms of clustering\udperformance, measured by the data discrepancy factor