2007 Fiscal Year Final Research Report Summary
Automatic Classification of Legel Information
| Project/Area Number |
18500194
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (C)
|
| Allocation Type | Single-year Grants |
|---|
| Section | 一般 |
|---|
| Research Field |
情報図書館学・人文社会情報学
|
|---|
| Research Institution | Osaka University |
|---|
Principal Investigator |
YORO Shin-ichi Osaka University, Graduate School of Law and Politics, Associate Professor (30240831)
|
|---|
| Project Period (FY) |
2006 – 2007
|
| Keywords | Legal Information / Machine Learning / Automatic Classification / SVM |
|---|
| Research Abstract |
We can obtain a lot of legal information by retrieving the Internet, but usually it contains a lot of needless information. If we can automatically classify information into legal information and other information, we can efficiently retrieve legal information. We tried to automatically classify it by using Support Vector Machine Algorithm.
In 2006, we developed the experiment system of automatic document classification to support our research. This system learns the features of the documents that have already been classified into some categories. And then it can classify a document based on the learning result.
In 2007, we experimented on an automatic classification of legal documents. We tried to classify documents into law information and other information, tried to classify legal documents into the decision sentences and the statutes, and tried to classify decision sentences into those of the criminal trials and those of the civil trials. In all cases, we could automatically classify documents with high accuracy. This result will leads to the development of the search engine that specializes in legal information.
|
Research Products
(8 results)
-
-
-
-
-
-
-
[Book] 法情報学(第2版補訂版)2006
Author(s)
加賀山茂, 松浦好治, 門昇, 田中規久雄, 養老真一
Total Pages
298
Publisher
有斐閣
Description
「研究成果報告書概要(和文)」より
-
[Book] Legal Informatics2006
Author(s)
Shigeru, Kagayama, Yoshiharu, Matsuura, Noboru, Kado, Kikuo, Tanaka, Shin-ichi, Yoro
Total Pages
298
Publisher
Yuhikaku
Description
「研究成果報告書概要(欧文)」より
