Prof. Tsvi Kuflik (The University of Haifa)
Title: Small groups modeling in cultural heritage – automatic detection of social behavior of pairs of museum visitors
In many cases, visitors come to a museum in small groups. In these cases, the visitors’ social context has an impact on their museum visit experience. Knowing the social context may allow a system to provide socially-aware services to the visitors. Evidence of the social context can be gained from observing/monitoring the visitors’ social behavior. However, automatic identification of a social context requires on the one hand identifying typical social-behavior patterns, and on the other using relevant sensors that measure various signals and reason about them to detect the visitors’ social behavior. The talk will present such typical social-behavior patterns of visitor pairs, identified by observations, and then, the instrumentation, detection process, reasoning, and analysis of measured signals that enables to detect the visitors’ social behavior. Simple sensors’ data, such as proximity to other visitors, proximity to museum points-of-interest, and visitor orientation were used to detect social synchronization, attention to the social companion, and interest in museum exhibits. The presented approach may allow future research to offer adaptive services to museum visitors based on their social context, to support their group visit experience better.
Tsvi Kuflik is a professor of Information Systems Dept. at The University of Haifa, Israel. Over the past eighteen years, the focus of his work was on intelligent user interfaces and ubiquitous user modeling applied to cultural heritage. In the course of his work, a “Living Lab” has been developed, where multiple studies about the application of novel ICT for enhancing the museum visit experience are conducted in a realistic setting at the Hecht museum. Tsvi got B.Sc. and M.Sc. In computer science and Ph.D. In information systems from Ben-Gurion University of the Negev, Israel. Tsvi is a distinguished ACM scientist and a senior IEEE member.
Yoshifumi Seki (Gunosy Inc.)
Title: From startup to established company: News recommendation algorithm design based on user experience and business strategy
Recommender systems have an important role in various web services, such as e-commerce, video streaming, news media. In web services, user experiences and business value is constantly changing with the growth of its service. Of course, recommender systems need to change with these changes, however there are very few examples of how recommender systems change with the growth of the web service. In this talk, I introduce the case about Gunosy, which is a news delivery application in Japan, and its company, Gunosy inc. I’m a co-developer of the service and co-founder of the company. This service was released in 2011 and has grown rapidly, and the company was listed in Tokyo Stock Market in 2015. Now, the company runs four big news services. From that experience, I introduce that how the design of recommender systems has changed in each phase, such as the period of budding in the early stages, the period of rapid growth, and the period of large-scale service. Moreover, I will also describe the issues that we are currently working on. Especially in the context of Multi-Domain, we will also introduce what kind of algorithms have been designed and improved while providing multiple services as the service grows.
Yoshifumi Seki is co-founder of Gunosy Inc. He co-developed Gunosy in 2011 when he was a postgraduate student and eventually co-founded the company in 2012. This company was listed in Tokyo Stock Exchange Market in 2015. In parallel, he received his Ph.D. in Engineering from the University of Tokyo in 2017. He’s been responsible for news delivery algorithms at the company and has deep expertise in web-mining focusing on recommendation systems and application of machine learning. Currently, he is engaged in research and development focusing on recommender systems, user behavior analysis.
Prof. Takahiro Hara (Osaka University)
Title: User activity prediction based on cross-domain approaches in different service domains
In our daily life, a variety of services are available in both on-line and off-line (real) worlds. To provide effective personalized services, service providers have been trying to develop digital marketing techniques by analyzing big data such as service usage logs. However, these techniques have a strong limitation that they cannot share user activity prediction models developed in each service domain because they cannot share user IDs and raw data of the users’ service usage. In this talk, we will present some recent achievements of our research project under the JST CREST program. Our project aims to solve the above problem by developing cross-domain approaches for predicting user activities in different service domains. Concretely, we model users (personas) and items (e.g., products in an e-commerce domain and POIs in a location service domain) from various data sources, and then, develop user activity prediction models which bridge different service domains.
Takahiro Hara received the BE, ME, and Dr.E degrees in information systems engineering from Osaka University, Osaka, Japan, in 1995, 1997, and 2000, respectively. Currently, he is a full professor in the Department of Multimedia Engineering, Osaka University. His research interests include database systems, distributed systems, mobile computing, and social computing. He has published more than 500 Journal and international conference papers in the areas of databases, mobile computing, distributed systems, social computing, and wireless networking. He served as a General Chair of IEEE International Symposium on Reliable Distributed Systems (SRDS’14) and International Conference on Mobile and Ubiquitous Systems: Computing, Networking and Services (Mobiquitous’16, and 21). He served as a Program Chair of a number of international conferences including IEEE International Conferences on Mobile Data Management (MDM’06, 10 and 18), Mobiquitous’13, and IEEE SRDS’12. He served and is serving as a Program Committee Member of more than 200 international conferences including top-ranked ones such as VLDB, WWW and CIKM. He is a distinguished scientist of ACM and a senior member of IEEE.