現在の私は、聴覚障害を持つ学生、ろう・難聴学生を対象に手話を併用した授業を行 い、また、ろう・難聴の学生と一緒に、障害当事者ならではの視点を活かした研究に主 として取り組んでいます。特に、障害に関係なく、誰もが誰かの情報を支援する ISeee Project（Information Support of Everyone, by Everyone, for Everyone） においては、筑波大学の森嶋厚行先生や岩手大学の張先生と一緒に、クラウドソーシ ングとアクセシビリティを結び続け、新たな支援のあり方を提示するような研究をして います。その様な中で、学会発表や学会活動において（広義の）データベースとアクセシ ビリティを結び続ける活動を継続的に行ってきたことを評価していただいたことが、 今回の受賞につながったのではないかと自負しております。
It is a great honor for me to receive the DBSJ Young Researcher’s Achievement and Contribution Award. Actually, this is the second award from DBSJ as I have also received the DBSJ Kambayashi Young Researcher Award in the past. Both are for young persons, so I am happy I am still considered young (although, my wife would definitely disagree here :-) ).
With this occasion, I want to express my deep gratitude to my great mentors who influenced me so much both in the professional and private lives: Prof. Katsumi Tanaka, Prof. Masatoshi Yoshikawa and Prof. Yukiko Kawai. Very special thanks go also to all the faculty members of Kyoto University’s Tanaka lab (which is now past but will always remain eternal), who helped me so much and who were my great friends at the same time. I also send my thanks and best wishes to all the members of Prof. Tanaka’s group at NICT in the Keihanna area that I had pleasure to meet and collaborate with. It was so much fun to work with you all!
The time has come for me to venture into new places. If we stay too long at the same good place, we will never know what the world looks like outside there. “Don’t you think so?” I am thus on my learning path now. I encourage you to look for chances of working at foreign institutions or staying for some time abroad to gain experience. This will probably change you, just like the experience in Japan has shaped me. Also, travelling abroad will help you make useful life-time connections. I hope that I myself can be such a connection for some of you. So please don’t hesitate to contact me if you need any help or want to collaborate, or just want to say hi.
My research has started with Web Archives. Then after the great advice of Prof. Tanaka I extended my focus to broader topics related to the temporal and history-related aspects of IR and NLP areas. I have also stayed open for other research areas like social media mining, text readability assessment or citation recommendation, and even crime analysis & prediction! I think that curiosity and imagination were always driving my research. I tended to first come up with a novel and exciting research task, and then to find a way to solve it (or not..). But anyway, I think this was a good approach, at least I enjoyed it much over the course of time, and I think I could propose something quite new in the end. It is a great benefit of being in academia that you can do what you like or dream about, and that you can choose to work with the people who you really like. So precious!
With this I am going to conclude. I wish the DBSJ community to keep doing the fantastic work it does. Everyone, please stay safe. And the last thing I shall say is: “I’ll be back!”.
■３−１■ 日本データベース学会上林奨励賞を受賞して ～How to Find a Good Research topic: My Experience in Data Privacy Research～ 曹 洋（京都大学 特定助教）
I would like to express my sincere gratitude to the Database Society of Japan (DBSJ) for awarding me the Kambayashi Young Researcher Award. Taking this opportunity, I would like to share some thoughts on "how to find a research topic.” I hope to inspire my peers and students who are just starting their research careers.
First of all, I would like to introduce myself briefly. I received my Ph.D. from Kyoto University in 2017 under the supervision of Professor Masatoshi Yoshikawa. Then I did postdoctoral research at Emory University with Prof. Li Xiong's group for a year as a postdoctoral researcher, and in 2018 I returned to Yoshikawa-Ma lab at Kyoto University as a program-specific assistant professor until now. My main research areas are data privacy and security. In particular, I am interested in differential privacy, federated learning, blockchain, and data market. Below, I summarized three keys to "how to find a good research topic" through my journey. (Please note that the following is a personal opinion and may not apply to everyone.)
- Skepticism In 2014, I started my Ph.D. research at Kyoto University. After several months of discussion with Prof. Yoshikawa, I decided on my research direction: data privacy. What are the specific problems I can study? How? I had no idea. Although I had already heard that "in research, a good question is much better than a good answer,” I don't know what a good question in data privacy is. After a few months of literature survey, I found many papers that mentioned that a fundamental problem in data privacy research is how to define privacy. And they all pointed to differential privacy, which is a mathematical definition of data privacy. Since differential privacy has already well-established, is there any need to study how to define privacy? Maybe differential privacy is not good enough in some cases? This sounds like a severe accusation because many privacy-preserving algorithms are implemented based on the definition of differential privacy. But I don't want to believe it blindly. I started exploring when differential privacy doesn't provide the guarantees it defines. Ultimately, we found that the privacy guarantee of differential privacy on temporally correlated data is not as exact as it describes. In short, we proved that the guarantees provided by differential privacy might diminish over time. This result is published in IEEE ICDE 2017 and selected as the top-6 best papers.
- Half new, half old When I finished my Ph.D. thesis, my question was: what's my next research topic? At the end of 2016, I was a visiting student in Prof. Xiong's research group at Emory University. I learned many different topics in data privacy since their group mainly focused on data privacy. In 2017, I started my postdoctoral research at Emory. The first question I thought about was how to expand my knowledge and research area. Should I start over in a new direction? Or should I continue with my previous doctoral work on differential privacy? The former’s advantage is that I can learn something new, but the disadvantage is that there is a learning curve and time cost because I have to study many new things. The advantage of the latter is that, after a few years of doctoral study, I have accumulated my expertise on differential privacy, and it is relatively easier to publish papers in familiar research problems; the disadvantage is that I risk myself being stagnant in a small area. In my case, I chose half new and half old: location privacy, which is a field very related to differential privacy, but different from it. In this way, I could learn something new, and at the same time, use the experience I already had.
- Finding the cold in the hot When I became a faculty member at Kyoto University in 2018, my question was still: What is my next research topic? I thought about the following questions: Should I follow the trend of AI? We all know that AI is changing our world and is an exploding "hot" topic. One answer is no, because you should keep your own identity and uniqueness. One answer is yes, because you shouldn't ignore a field that is changing the world! Ultimately, the answer I gave myself was: keep your uniqueness in this impactful field. As you can imagine, my students and I are now working on a half-new, half-old topic: differentially private machine learning. It's not (yet) a hot area in AI, but I'm convinced that we can create extraordinary value by shaping a different future.