Securing reliability in extracting minor topics from the Web

2017 Fiscal Year Final Research Report

• Project/Area Number: 15K00426
• Research Category: Grant-in-Aid for Scientific Research (C)
• Allocation Type: Multi-year Fund
• Section: 一般
• Research Field: Web informatics, Service informatics
• Research Institution: Kyushu University
• Principal Investigator: Nakatoh Tetsuya 九州大学, 情報基盤研究開発センター, 助教 (20253502)
• Co-Investigator(Kenkyū-buntansha): 廣川 佐千男 九州大学, 情報基盤研究開発センター, 教授 (40126785) ; 池田 大輔 九州大学, 学内共同利用施設等, 助教授 (00294992) ; 山田 泰寛 島根大学, 総合理工学研究科, 助教 (50529609)
• Project Period (FY): 2015-10-21 – 2018-03-31
• Keywords: マイナートピック / 低頻度語 / 信頼性 / 学術情報

Outline of Final Research Achievements

Regarding extraction of minor information (1) Tourist information written on the blog (2) Academic paper information obtained from the website of academic paper information was analyzed. Regarding (1), attention was focused on the low frequency word extracted as a tourism resource candidate included in the sightseeing blog and its background was analyzed, showing that there are two types of appearance tendencies. As for (2), we showed the effectiveness by proposing a new evaluation index limited to the number of citations from a specific area meeting the research purpose, and an evaluation index not using the cited number. In addition, by emphasizing the papers located at positions connecting important research, we constructed a system that appropriately visualizes the flow of research. It became possible to search and analyze documents including minor articles.

Free Research Field

情報工学