Zhou's Research Improves Data Mining Processes

Information granules, which are collections of entities that usually originate at the numeric level and are arranged together due to their similarity, functional or physical adjacency, indistinguishability, or coherency, are effective in revealing the structure of data. Therefore, it is a common practice in data mining to use information granules for classifying datasets. A new study by TSYS School computer scientist Yi Zhou and his colleagues from Jinan University and the University of Exeter utilizes the feature weighting of data to produce information granules with a high consistency rate. As a first step, Zhou and his co-authors use consistency rate and contribution scores to generate information granules. Next, they propose a granular two-stage classifier that divides the data into fuzzy and fixed points and then calculates the interval matching degree to assign data points to the most suitable cluster. Zhou and his colleagues ultimately compare their output with that of two other state-of-the-art granular models, finding that their process has sufficient performance to describe the relevant data structure. More specifically, their process reduces the time overhead required to obtain information granules by 51%, while also increasing the per unit quality of the granules by about 15%. The study is set to appear in a future issue of IEEE Transactions on Big Data.