Enhancing Visualization with Vizster in Knowledge Extraction Scenario

Abstract

To integrate visualization and data mining techniques like knowledge extraction toward a mixed-initiative knowledge discovery, it is needed to survey and reflect on the various ways which takes the best of human and machine capabilities. It categorizes the observed techniques in classes following a bottom-up bibliographic research approach, highlighting current trends, gaps and potential future directions for research work. In particular, it looks at strengths and weaknesses of data mining and information visualization (InfoVis), and for which purposes researchers in InfoVis use data mining techniques and also investigates how researchers in data mining employ InfoVis techniques. This information is further used to analyze the discovery process by comparing the analysis steps from the perspective of data mining and information visualization. The comparison sheds light on new perspectives on how mining and visualization can best employ human and machine skills.