퍼스널 컴퓨팅 환경에서 다이어리 생성을 위한 시계열 군집화 방법

Abstract

People try to keep a diary to perpetuate their memories and experiences even in PC environment. However, as like a life-log system, we need to consider several issues related to activity capturing and data management. Moreover, the PC usage logs are composed of multivariate, categorical, and temporal events. Our goal, in this paper, is to generate a PC diary based on the abstract and meaningful information of a user’s PC activities. Therefore, we propose a temporal clustering method to differentiate user activities and extract representative logs from the data stream. After implementing a real-time process of extracting PC usage logs, we investigated our clustering method and demonstrated a process of diary generation.