Studies on Text Data Analysis Based on Fuzzy Clustering

点击翻译按钮获取中文摘要

With the goal being to extract meaningful information from non-structural text data, several techniques for intuitively understanding intrinsic association rules were developed considering sample classification, variable selection and visualization based on linear fuzzy clustering (or local principal component analysis). Useful text-keyword maps are constructed through such techniques as automated keyword selection or core document identification by noise rejection in document grouping processes.