DEPARTMENT RESEARCH ACADEMICS PEOPLE NEWS
Rice Computer Science		SEARCH:
Rice University Department of Computer Science presents Hisham Al-Mubaid University of Houston-Clear Lake Machine Learning for Natural Language Processing Abstract Most of the efficient natural language processing (NLP) systems and methods are designed based on machine learning approaches. We have devised a new learning architecture and applied it to a number of NLP problems including detection and corrections of context-based spelling errors, word prediction, and text classification. The learning architecture is based on a logic learning classification technique called Lsquare. Lsquare is a classification technique based on learning logic rules; this technique has achieved a competitive performance in a number of classification-based NLP and data mining problems. The learning method has the following new properties: (1) A new way of encoding the representation of words. The features of words are extracted such that, for a given occurrence of a word w, the structure of the neighborhood of w is recorded in connection with other occurrences of w and their neighborhoods. (2) A new way of learning how correct word occurrences can be recognized from prior, correct text. This step uses the encoding of (1) and the data-mining algorithm Lsquare. It acquires knowledge (trained classifiers) that contains insight into the context-based syntactic [and to some extent semantic] representation of text words. (3) A new way of employing the knowledge of (2) in the target NLP task. The learning method will be explained along with a number of examples applications from NLP domain. Tuesday, May 20 at 2:00 p.m. in DH 1049 About Hisham Al-Mubaid Hisham Al-Mubaid has received his master's and Ph.D. degrees in computer science from the University of Texas at Dallas in 1996 and 2000, respectively. He is now an assistant professor in department of computer science at University of Houston - Clear Lake. His research interests include natural language processing and machine learning, operating and distributed systems. Copyright 2000 Rice University site problems: www@cs.rice.edu --- --- --- --- --- --- --- --- --- --- --- --- --- --- --- --- --- --- --- --- --- --- --- --- --- --- --- --- --- --- --- --- --- --- --- --- --- --- ---