Automatic Extension of Semantic Lexicons with a Bootstrapping Algorithm: Using Corpora to Learn Semantic Features

Where is the information I need in this huge pile of data? This question will be asked more and more in our information society and the only way to solve it on a large scale is to process the data automatically. Since most often this data is unstructured and available only in natural language, we need to understand the inner structure of natural languages and employ the tools of computer science to effectively extract the information we need. For this task to be successful a good lexicon is needed. Many fields of advanced natural language processing, such as information retrieval, word sense disambiguation or semantic web applications are based on lexical information. Because the manual creation of high quality lexical resources is very expensive and time consuming, there is a need for automatic or semi-automatic tools to create these. This book investigates and extends a bootstrapping approach which permits to extend high quality lexical resources with the help of very large corpora. It is directed towards researchers or technical staff interested in either automatically creating a new lexicon or extending their general or domain specific lexicons to new domains.