Survey of Document’s Clustering Methods by Means of Learning Algorithms

The growing rise of databases almost in every area of human activity has caused the need for new powerful tools to change the suitable knowledge increase. In order to satisfy this need, the researches of various fields such as machine learning, pattern identification, analysis of statistical data, data visualization, neural networks, econometrics, information retrieving, information extraction, etc. have explored some methods and ideas. Text mining uses unstructured textual information, studying it in order to discover the structure and hidden lateral meanings in the text. Documents’ clustering via unsupervised machine learning methods has an expanded function in different areas of natural processing languages such as automatic multi-text summarization, information retrieving, etc. The current paper aims to introduce some useful functions of this area, clustering the documents with the approach of decreasing noise redundancy as well as unrelated data. Dimension reduction is a method of erasing such features

Keywords:

document clustering , dimension reduction , unsupervised machine learning , pattern extraction