IBM Research - Text Analysis and Language Engineering

Roy Byrd has a degree in music theory and composition from Yale and is ABD in theoretical linguistics from New York University. His long career at IBM has included development of programming languages and compilers and of operating systems, as well as research in relational databases and query languages, natural language processing, computational lexicology, information retrieval, and text mining. He currently manages the Text Analysis and Language Engineering project at IBM Research. The project is concerned with developing scaleable NLP techniques for identifying and extracting conceptual information from large document collections, organizing that information, and using it to build advanced knowledge management systems.

Roy is always willing to talk to anyone about music, linguistics, computers, bicycling, cross-country skiing, travel, and other interesting things.

Rie Kubota Ando received a Ph.D. in computer science at Cornell University in 2001, and joined IBM T.J. Watson Research Center. Her thesis was about applying subspace projection to the document representation problems. Rie began her IBM career in Japan after graduating from University of Tokyo with a B.S. in computer science. Her current research interests are in machine learning.

Bran Boguraev is a non-native speaker of English, who would have liked to pursue linguistics and etymology; instead, he graduated with degrees in electronics and computer science, picking up a few languages along the way. At the first opportunity of doing independent research, he chose to study computational linguistics; he now holds a Ph.D. in that subject conferred by the Board of Mathematics at the University of Cambridge. Even after leading a number of joint academic projects and industrial consortia, as well as managing the natural language programs at two research centers of excellence, he still is unable to explain to his mother what he does for living. (His best attempt to hint at the challenges of his work is by inviting people to consider why the plural of "passer by" is "passers by", but the plural of "forget me not" is not "forget us not"...)

He is the author/co-author of four books, and over sixty refereed articles, in the area of computational linguistics and natural language engineering; most recently, his work has focused on leveraging shallow syntax and semantics for deep content characterization. He likes reading books; his hobby is collecting books; and he exercises by energetically reading as many books as possible.

Frank Elio earned a B.A. in Psychology from Cornell University and an M.S. in Computer Science from Pace University, graduating with distinction. He started out programming on and supporting DEC “minicomputers. When he joined IBM Watson Research in 1984, he shifted gears to work on mainframes, but was dismayed to find them less interactive. When the PC explosion hit IBM, he was freed from JCL, half-duplex connections, and line editors. At Research, he’s worked on a wide variety of projects including Information Center consulting, systems management, management, application development and support, and other software development projects. Since joining the TALENT group, he’s mostly done work in glossary administration, summarization, and the semi-automatic generation of back of the book indexes.

Mary has a bachelor's degree in German and French from the University of Arizona, was a Fulbright Fellow at Christian Albrecht Universitaet in Kiel, Germany, and holds a Ph.D. in Germanic Languages from the University of Texas at Austin. Her academic interests were Germanic and Indo-European linguistics and religion, especially Germanic and Indo-European religious vocabulary. She began her IBM career in Austin as an applications programmer for engineering and manufacturing, but quickly moved into applied natural language processing when she became part of a multi-lingual group that produced one of the first commercial spelling checkers. At Thomas J. Watson Research Center, she has pursued research interests in computational lexicography, computational terminology, bilingual terminology management, machine translation, information retrieval, and text mining.

Mary has a well-worn passport, a respectable collection of ribbons won in local and international photographic slide competitions, a shelf of speech contest trophies, and a collection of ethnic footwear, artifacts of an interest in international folk dancing and ethnomusicology. She is the author of a family history book containing recent original research in German archives, and is a member of the board of directors of Westchester Disabled on the Move.

Paul has a bachelor's degree in zoology (that's right, zoology) from George Washington University, and has worked in research at such diverse organizations as Cornell University Medical College, MIT and Bayer Diagnostics, where he developed applications for the display and analysis of flow cytometric data. He began his IBM career as a contractor, working in digital libraries and educational research.

Youngja received a Ph.D. in Computer Science from Yonsei University in Seoul, Korea. Her research interests lie in natural langugae processing; in particular, unsupervised lexical knowledge acquisition, computational terminology, information extraction, information retrieval and the Semantic Web. Youngja is also interested in the genetic algorithm; especially applying the genetic algorithm to clustering problems.

In addition to the current members of the group, significant contributions were made to our research by these past memebers over the past several years.

Members