■論文概要
Due to the advent of Web services such as MOOC (e.g., Coursera, iTunes U) and presentation sharing websites (e.g., SlideShare, Prezi), they provide the archives of oral presentations and materials in public speeches and lectures owing to the features for searching, browsing, reusing, and sharing presentation contents. For example, famous Coursera and SlideShare provide an online educational presentation archive for self-learning and later review with presentation slides or video recordings of lectures. Thereby, people can be readers of presentation contents even if they are not in the same place as presenters. Although these tools and Web services make it easy for creating and sharing presentation contents are widely used, criticisms have pointed out their problems from the viewpoint of understandability about relevant information, structural information, and contextual information of presentation contents. The goal of this research is to solve the problems of traditional presentation tools focusing on searching, generating, and grasping presentation contents in e-learning, and then to develop efficient and useful applications for improving understanding of presentation contents, enhancing user interactions through presentation contents.

In this doctoral dissertation, in order to achieve these goals, the fundamental approach of this research is to explore explicit and implicit semantics of presentation contents. This approach enables to determine semantic relationships, extract expression styles, and present presentation context in presentation contents. Specifically, our approaches not only for a structural and semantic analysis of presentation contents, but also focus on support for retrieval, generation, grasping overviews of presentation contents. This research consists of five themes: support for presentation content retrieval to readers or searchers, (1) scene combination for slides with recorded videos, and (2) semantic slide ranking and snippet generation; support for presentation content generation to presenters or authors, (3) outline generation for presentation slides; support for grasping overviews of presentation contents to readers, (4) dynamic word clouds of presentation slides, and (5) iPoster: a collaborative browsing platform for slides.

■学位を取得して
学位取得は一朝一夕になることではなく、本当に長い道のりでした。振り返れば、2008年5月から日本へ留学し、兵庫県立大学研究生となって、日本語や専門分野の勉強に沒頭して日夜を送っており、３ヶ月後に同大学環境人間学研究科（博士前期課程）入試に合格しました。それから5年間という長い時間を経て2014年3月に博士（環境人間学）学位を授与されました。その博士前期・後期課程の5年間に得た経験と達成感は、私個人にとってはとても大きな財産でもあったが、指導教官のご指導や同研究室の皆様の支援を賜わって初めてできたと思います。無論、博士号取得自体は、一人前の研究者になるには程遠いと認識していますが、今日までの努力はそのゴールに近づくものであると確信し、根性をもって絶対に勝者になるよう、努力をすればいつか必ず報われるようになると確信し、コツコツやってきた数年間でした。今後もプロの研究者を目指してやって行こうという志をもっています。