DiSCourse Seminar with Yu Suzuki
17 March 2023, 12:00 (CET), hybrid
Digital Science Center, Innrain 15, 1st floor, Open Space Area or Big Blue Button
DiSCourse* - The Digital Science Seminar Series on
High-quality Social Media Analysis using Crowdsourcing and Machine Learning
Social media, such as Twitter and Facebook, is widely used for communications. The messages on social media contain customer reviews. However, these reviews do not always show clear opinions, it is difficult to identify whether the reviews include positive or negative opinions automatically.
In our research, we use both crowdsourcing and machine learning for sentiment analysis of social media. We constructed the crowdsourcing system and analyzed about 530,000 tweets that mention smartphones, automatic vacuums, and others. We used several machine learning techniques such as multi-task learning and anomaly detection. Using our system, we can construct large-scale test collections. These test collections will help construct novel machine-learning algorithms.
*featuring a distinguished guest: Yu Suzuki, Head of the Suzuki Lab at Gifu University
Yu Suzuki is an Associate Professor at the Department of Faculty Engineering of the Gifu University, Japan, and the Head of the Suzuki Lab. His research interests include social data mining, crowdsourcing, and databases. He is an Associate Editor of the IEICE Journal (Institute of Electronics, Information, and Communication Engineers) and a reviewer of many conferences including the Web Conference and the Conference on Information and Knowledge Management (CIKM).