| Project/Area Number |
15300031
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (B)
|
| Allocation Type | Single-year Grants |
|---|
| Section | 一般 |
|---|
| Research Field |
Media informatics/Database
|
|---|
| Research Institution | Ochanomizu University |
|---|
Principal Investigator |
MASUNAGA Yoshifumi Ochanomizu University, Department of Information Science, Professor, 理学部, 教授 (70006261)
|
|---|
| Co-Investigator(Kenkyū-buntansha) |
TACHI Kaoru Ochanomizu University, Institute for Gender Studies, Professor, ジェンダー研究センター, 教授 (50155082)
OYAMA Naoko Ochanomizu University, Institute for Gender Studies, Guest Researcher, ジェンダー研究センター, COE客員研究員 (00194639)
KITSUREGAWA Masaru University of Tokyo, Institute for Industrial Science, Professor, 生産技術研究所, 教授 (40161509)
FUJISHIRO Issei Tohoku University, Institute for Fluid Science, Professor, 流体科学研究所, 教授 (00181347)
|
|---|
| Project Period (FY) |
2003 – 2005
|
| Project Status |
Completed (Fiscal Year 2005)
|
| Budget Amount *help |
¥16,500,000 (Direct Cost: ¥16,500,000)
Fiscal Year 2005: ¥4,100,000 (Direct Cost: ¥4,100,000)
Fiscal Year 2004: ¥6,000,000 (Direct Cost: ¥6,000,000)
Fiscal Year 2003: ¥6,400,000 (Direct Cost: ¥6,400,000)
|
|---|
| Keywords | database / Web community / data mining / gender / portal site / social science / research methodology / WWW / リンク解析 / Companion- / 共時的 / 通時的 / 可視化 / ウェブ / コミュニティ / companion- / ブラウザ |
|---|
| Research Abstract |
Web mining is one of the most important research issues in modern information technology. Among Web mining techniques, the Web structure mining is essential to find Web communities. A variety of mining algorithms and tools have been developed for finding Web communities. In this research, we paid attention to the HITS-based Web structure mining tool "Companion", which was able to provide diachronic as well as synchronic analysis of Web communities. The capability of diachronic analysis was essential, particularly when a mining tool was applied to investigate a Web community-like the gender-related community-that changes shape dramatically over time according to socio-cultural and political changes. In this research we first observed how changes in the real world communities, i.e. organizations their members, could be reflected on the Web as changes in Web communities. Second, we analyzed gender-related Web communities to discover new trends or movements that were hard to identify witho
… More
ut a mining tool capable of diachronic analysis. Last, we investigated a case that calls for the deciphering capability of miners. Based on the above results, we constructed a "portal site for gender studies" in the Institute for Gender Studies of Ochanomizu University.
By analyzing the Web communities constructed around the keyword "gender" on the Web, we have attempted to gain new conceptions of societal gender phenomena, available for observation only on the Web. Using Companion- as a Web mining tool, the research examines the development of genderrelated communities extracted through the search term "ender." This method enabled us to examine the gender communities subjectively as well as objectively, and in fact, to perceive a heretofore unrealized phenomenon. Plus, since the developmental process of the Web communities was examined thoroughly by gender specialists, the present analysis provided a positive example in terms of the meaningful nature of its findings. In other words, we were able to verify that the texturized displays, which seem nonsense at first glance, become a vital and challenging space to be deciphered (=analyzed) through the eyes of specialists of a particular field. The research concretely demonstrates the effectiveness and impact of collaborating with researchers of specific fields who are familiar with the reality of the contents in facilitating the development of appropriate Web mining tools. This kind of dialog with researchers can only make the analysis tool more sophisticated. If the functions of the mining tools improve, there is certainly the possibility of finding even more interesting types of results from their analyses. Simultaneously, we have cautioned that if particular tendencies or characteristics of the analysis tool are not seriously considered, the examination research may be flawed. Reflection upon the "analysis tool" itself is always necessary for such examinations. Less
|
