2018 Fiscal Year Final Research Report
Research on Information Elicitation Technologies for Estimating Latent Information Needs
| Project/Area Number |
26700009
|
|---|
| Research Category |
Grant-in-Aid for Young Scientists (A)
|
| Allocation Type | Partial Multi-year Fund |
|---|
| Research Field |
Multimedia database
|
|---|
| Research Institution | Kyoto University |
|---|
Principal Investigator |
Makoto P. Kato 京都大学, 国際高等教育院, 特定講師 (00646911)
|
|---|
| Project Period (FY) |
2014-04-01 – 2019-03-31
|
| Keywords | 対話型検索 / クエリ推薦 / 再検索 / インターリービング / エンティティ検索 / ランキング学習 |
|---|
| Outline of Final Research Achievements |
In this research project, I conducted research on information elicitation technologies, which estimate users' information needs, preferences, and opinions by observing responses to a certain stimulus. The first research topic was "elicitation of information needs by question-answering" and included conversational relevance feedback, optimization of search result presentation with query suggestions, and a study on effective questions in refinding search tasks. The second research topic was "elicitation of preferences", including estimation of impressions triggered by pictures and ranking preference estimation in a question-answering Web service. The third topic was "reliable elicitation of opinions based on multiple questions" and focused on opinion elicitation from social media, opinion elicitation by aggregating numerical information and reviews on the Web, and reliable elicitation of opinions from news media.
|
| Free Research Field |
情報検索
|
| Academic Significance and Societal Importance of the Research Achievements |
本研究は，これまで受動的にデータを獲得してきたデータ分析に基づく研究とは異なり，ユーザに対して積極的にデータの提供を要求し，より適した働きかけ方についての検討・実践を行ったところに学術的意義があると考えられる．近年，Google HomeやAmazon Alexaなどの対話型情報アクセス技術が大きな注目を集めており，本研究課題ではこれらの基礎的な研究を行っていたと言える．これに加え，ユーザの反応に基づいてWebサービスを改善していく技術についても，さまざまな企業が近年の研究成果を導入し始めており，特にインターリービング技術の実践的研究では実利用における有用な知見を得られたと考えられる．
|
