中央研究院 資訊科學研究所

Rapid increase of scientific documents causes difficulty in acquiring up-to-date information even by experts. Through deepening the text and document analysis technologies and automatic construction of knowledge bases necessary for document understanding, this project aims to develop foundations of content understanding of large scale technical documents, integration of acquired knowledge, and semantic similarity at structural levels of contents and documents. In collaboration with the experts in Bio-science, Material Science, Neuroscience, law, and Artificial Intelligence, we aim to develop an integrated environment for content-based document retrieval and summarization, knowledge discovery by aggregating the contents from large scale documents, survey generation by inter-document relation analysis.

For achieving those aims, we set up 9 basic tasks in 4 technical categories as shown in the Figure. We conduct all of those tasks and an integrated system through complementary collaboration of 7 teams including 6 major collaborating groups.

Dr. Yuji Matsumoto is currently a Professor of Information Science, Nara Institute of Science and Technology, Japan. He received his M.S. and
Ph.D. degrees in information science from Kyoto University in 1979 and in 1989. He joined Machine Inference Section of Electrotechnical Laboratory in 1979. He has then experienced an academic visitor at Imperial College of Science and Technology, a deputy chief of First Laboratory at ICOT, and an associate professor at Kyoto University. He became an ACL fellow at 2011, and was the president of AFNLP during 2013-2014. He has won many best paper awards at various international conferences. His main research interests are natural language understanding and machine learning, and is the principal investigator for the CREST project "Knowledge Discovery through Structural Document Understanding".