NLP-Driven Document Representations for Text Categorization: Empirical Selection of NLP-Driven Document Representations for Text Categorization

Text Categorization is the task of assigning predefined labels to textual documents. Current research has been focused on using word based representations called bag-of-words (BOW) with strong statistical learners. Few studies have explored the use of more complex Natural Language Processing (NLP) driven representations based on phrases, proper names and word senses. None of these had definitive results on these features? benefits for text categorization problems. This book studies the use of NLP-driven document representations captured at many different levels of language processing, and shows that NLP-driven document representations improve text categorization. A methodology, called? Empirical Selection Methodology for NLP-driven document representations? is presented. Methodology helps to select document representations for each category in the categorization problem. The methodology should help Text Categorization researchers as well as researchers working on other classification problems, because it is generalizable, and can produce better instance representations for different learning problems.

Ozgur Yilmazel, Ph.D. Assistant Research Professor at School of Information Studies, Syracuse University and Chief Software Engineer in Center for Natural Language Processing (CNLP). He received his doctorate from Syracuse University Electrical Engineering in April 2006.