Hyun Jung La  Moon Kwon Kim  Han Ter Jung  Soo Dong Kim

10.5626/JOK.2018.45.4.348

audio context Semantic analysis preference tendency analysis activity pattern analysis

Recently, there has been a trend for collecting various rich personal contexts, such as social network services (SNS), user created contents (UCC), and digital diaries. Because of this trend, much more attention to semantic analysis with personal contexts is being paid. The semantic analysis allows users to analyze and understand diverse aspects of their lives such as their lifestyle and quality of life which are not easily recognized by them. Hence, in this paper, we propose a process to infer semantics from personal contexts, more specifically, audio contexts. This paper is focused on proposing detailed algorithms for analyzing user’s preference tendencies and activity patterns. To evaluate the proposed methods, we apply them to developing a system, called Smart Diary System, which is used to analyze a user’s preference tendency and activity pattern from audio diaries, and we present experiment results with the system. We expect to use the proposed process and algorithms in various application domains such as personal secretary service, recommendation services, and advertising services.