Modified K-means algorithm for emotional intelligence mining

Data mining algorithms are widely used in many fields of science and technology. Clustering algorithms are the popular algorithms of data mining. K-means is the most popular and simplest of all clustering algorithm and used in various fields. The basic K-means algorithm has some problems, such as, it may produce empty clusters, immense computational complexity and poor quality of cluster. This paper focusing on minimizing the problems of K-means algorithm by some modifications in it. Post modification, K-means algorithm applied on large sensitive Emotional Intelligence (EI) data set, rather than simple numerical data set, to get meaningful clusters. Samples of 200 person´s emotional intelligence data collected from the survey. Then applied modified algorithm on EI data to create the clusters of persons based on emotions. This cluster analysis will be useful in the organization for team formation, team leader selection and decision making purpose. From the experiment, it was found that modified K-means, worked efficiently than basic K-means on EI data.