Proper

Abstract

The commercialization of various smart devices capable of collecting lifelogs has raised expectations regarding the utilization of lifelogs in health care services. Lifelogs sufficiently cover all of the features of Big Data and present time series properties. Due to this, there are difficulties with respect to the integration, processing, and analysis of lifelogs. In consideration of this, this study proposed a life balance service based on Big Data based feature extractions. This regarded the analysis of lifelog features and the pursuit of life balance predictions through collaborative filtering and user log data through dimensional reductions and similarity calculations using the features. In addition, performance evaluations regarding the accuracy of predictions and user satisfaction were undertaken to validate the effectiveness of the proposed method. As a result, the proposed method presented a 16.5% higher accuracy in predicting life balance than existing methods while also presenting excellent levels of user satisfaction. In light of this, the provision of life balance services through the proposed is expected to enable efficient life habit correcting health services using the log data of users.