2009 Fiscal Year Final Research Report
The feature selection focused on a latent semantic in a document set
| Project/Area Number |
20860085
|
|---|
| Research Category |
Grant-in-Aid for Young Scientists (Start-up)
|
| Allocation Type | Single-year Grants |
|---|
| Research Field |
System engineering
|
|---|
| Research Institution | Tokyo Metropolitan College of Industrial Technology |
|---|
Principal Investigator |
YOKOI Takeru Tokyo Metropolitan College of Industrial Technology, ものづくり工学科, 助教 (40469573)
|
|---|
| Project Period (FY) |
2008 – 2009
|
| Keywords | テキストマイニング / トピック抽出 / 特徴量抽出 / 自然言語処理 |
|---|
| Research Abstract |
The novel feature selection, i.e.an index words selection, based on a latent semantic (topic), for a large document set has been proposed in this study. The combination of topics among divided document sets was especially focused to obtain topics from a large document set, so that our proposal could extract a similar topic with the topic from a large document set with an extraction directly. In addition, the index words selected based on the above topics are confirmed to be efficient by the application of an information filtering with a user's interest.
|
