Trend analysis of machine learning -A text mining and document clustering methodology

Abstract

The Machine Learning is certificated as one of the most important technologies in today`s world. There are several various researches applying Machine Learning to improve its operation efficiency in many different aspects. Based on the Social Science Citation Index (SSCI) database, this research is using text mining technology which collecting the homogeneous glossaries in the articles, conducting to the literature cluster analysis. To select the term frequency index which generated by various glossaries aggregation from each article as well as an input variable for Self-Organization map (SOM) network, following by utilizing the network neuron automatic clustering function, dividing into 10 application domains of machine learning, finally proceeding the trend analysis coordinated with the articles by published year, discovering the historical vein and collecting the results by each research area, and further forecasting the future possible tendency. © 2009 IEEE.