A study on text linkage for document browsing and writing assistance

2017 Fiscal Year Final Research Report

• Project/Area Number: 15H02754
• Research Category: Grant-in-Aid for Scientific Research (B)
• Allocation Type: Single-year Grants
• Section: 一般
• Research Field: Intelligent informatics
• Research Institution: National Institute of Informatics
• Principal Investigator: Aizawa Akiko 国立情報学研究所, コンテンツ科学研究系, 教授 (90222447)
• Project Period (FY): 2015-04-01 – 2018-03-31
• Keywords: 言語横断情報推薦 / エンティティリンキング / 専門用語翻訳 / 意味の分散表現 / 英語論文執筆支援

Outline of Final Research Achievements

In this research, we studied text processing techniques to assist users in understanding and writing documents. First, we developed a technique for technical term linking where technical terms are extracted from text and then linked automatically to corresponding external knowledge base entries, such as Wikipedia articles, considering variations of the notations and semantic ambiguity. Next, we worked on methods for document structure analysis and question answering. Then, we analyzed the structure of scientific papers and constructed a text corpus. Finally, we developed a system for cross-lingual retrieval and paper recommendation and showed the usefulness of the proposed technical term linking.

Free Research Field

情報学／知能情報学