2015 Fiscal Year Final Research Report
Analyzing Microblog Articles based on Text Understanding
Project/Area Number |
24500296
|
Research Category |
Grant-in-Aid for Scientific Research (C)
|
Allocation Type | Multi-year Fund |
Section | 一般 |
Research Field |
Library and information science/Humanistic social informatics
|
Research Institution | Okayama Prefectural University |
Principal Investigator |
|
Co-Investigator(Kenkyū-buntansha) |
TAJIMA Yasuhiro 岡山県立大学, 情報工学部, 准教授 (00334467)
|
Project Period (FY) |
2012-04-01 – 2016-03-31
|
Keywords | ウエブマイニング / テキストマイニング / 自然言語解析 |
Outline of Final Research Achievements |
This work has two major results. One result relates to parsing micro-blog sentences. Since micro-blog articles often lacks sentence boundary markers, we first developed sentence boundary detector by using machine learning technique applied to word/character sequences. Then, we developed dependency analyzer including base phrase (so-called bunsetsu) chunker. Experimental results show our method outperforms existing software by 10 points. The other result is on trend analysis for micro-blog articles. We have developed a method for choosing an article which best describes a given burst word. The method applies sentence extraction to articles within the automatically identified burst period (for the given word).
|
Free Research Field |
自然言語処理
|