2017 Fiscal Year Final Research Report
Contents diversity analysis for consumer generated media contents
Project/Area Number |
15K00451
|
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 | Kyushu University |
Principal Investigator |
Ito Eisuke 九州大学, 情報基盤研究開発センター, 准教授 (90294991)
|
Project Period (FY) |
2015-04-01 – 2018-03-31
|
Keywords | コンテンツ / CGM / 多様性 / 類似度 / 回帰分析 / 深層学習 |
Outline of Final Research Achievements |
This research aimed to (1) quantitative evaluation of content diversity trends in CGM (Consumer Generated Media), and (2) establishment of content selection model of users. Targeted CGMs are movies, novels and academic papers. In (1), I proposed three metrics for quantitative contents diversity. Using sum of similarity metric, I found that diversity are decreasing for movies and novels, but academic papers keep diversity. In (2), I found that pageview distribution is close to the log-normal distribution, but the distribution differs for each genre. As a result of regression analysis to estimate pageviews of contents, estimation accuracy was good, but establishment of the user's content selection model was incomplete.
|
Free Research Field |
情報検索 情報統合 テキストマイニング
|