2015 Fiscal Year Final Research Report
An Automatic Coding System for Answers to Open-ended Questions in social surveys
| Project/Area Number |
25380640
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (C)
|
| Allocation Type | Multi-year Fund |
|---|
| Section | 一般 |
|---|
| Research Field |
Sociology
|
|---|
| Research Institution | Keiai University |
|---|
Principal Investigator |
|
|---|
| Co-Investigator(Kenkyū-buntansha) |
TAKI HIROFUMI 法政大学, 社会学部, 講師 (20634033)
TANABE SHUNSUKE 早稲田大学, 文学学術院, 准教授 (30451876)
|
|---|
| Research Collaborator |
RI I 東京工業大学, 大学院理工学研究科, 博士課程在学
|
|---|
| Project Period (FY) |
2013-04-01 – 2016-03-31
|
| Keywords | 社会調査 / 自由回答 / 職業／産業コーディング自動化システム / SSM/ISCO/ISIC / Web公開システム / 機械学習 / 自然言語処理 / 確信度 |
|---|
| Outline of Final Research Achievements |
Based on the National/International standard of occupation/industry code (SSM/ISCO/ISIC), the system can assign three candidate codes to an answer of an open-ended question in a social survey. And a three-grade confidence level was assigned to the first-ranked predicted code by using classification scores. Many functions are implemented in this system, for example, an ISCO/ISIC code can be assigned to the SSM code automatically, also maintenance function contained. This system is now released to the public. Researcher can use it through the official website of CSRDA.
A new automatic coding system for social surveys is in development, which contains an open-ended question.
|
| Free Research Field |
自然言語処理、機械学習、社会調査法
|
