作者: Kirsten Kirkegaard Moe , Birger Larsen , Jeanette M. Jensen
DOI:
关键词: Point (typography) 、 Human–computer interaction 、 Element (category theory) 、 Reading (process) 、 Context (language use) 、 Eye tracking 、 Relevant information 、 Computer science 、 Relevance feedback
摘要: Our goal in this study was to explore the potentials of extracting features from eye-tracking data that have potential improve performance implicit relevance feedback. We view type as an example searcher’ immediate context and containing useful clues indications interaction between searcher IR system. In particular, we explored if could qualitatively identify feedback, how such correlate with document elements assessed relevant or non-relevant. The results point so-called thorough reading one most promising for identifying information input feedback – particular when it is related total time has looked element.