| |||
|
ThoughtTreasureTM ThoughtTreasure is a commonsense knowledge base and architecture for natural language processing that uses multiple representations including logic, finite automata, grids, and scripts. The ThoughtTreasure knowledge base consists of: QA system From England Статья об этой системе Контекстный поиск Торговая марка RCOтм компании "Гарант-Парк-Интернет" объединяет продукты и решения, которые предназначены для внедрения в базы данных и информационно-поисковые системы и позволяют задействовать широкий арсенал лингвистических и аналитических средств для решения прикладных задач, требующих компьютерной обработки документов на естественном языке. Демо ресурс Этот ресурс посвящен демонстрации технологий и перспективных решений, которые разработаны подразделением RCO Research Group компании "Гарант-Парк-Интернет" в рамках направления RCO по созданию средств автоматического анализа и поиска текстовой информации. Edinburgh Question-Answering Reading Group (Fall 2002) Goal: To position the group to develop a Question-Answering system in Spring 2003; to understand the current state of the art in QA. NLP for Answer Extraction in Technical Domains In this paper we argue that questionanswering (QA) over technical domains is distinctly different from TREC-based QA or Web-based QA and it cannot benefit from data-intensive approaches. Technical questions arise in situations where concrete problems require specific answers and explanations. Finding a justification of the answer in the context of the document is essential if we have to solve a real world problem. Recently there has been an increase of research on text-based QA, triggered by the Text REtrieval Conference (TREC) Question Answering Track (Voorhees, 2001). In these modern approaches the knowledge base is replaced by collections of documents (mainly large corpora of newspaper articles) thereby eliminating the major problem of early QA systems. |
