Browsing by Author "Liu, Sun-Wei"

Now showing 1 - 2 of 2

Long-Term User Location Prediction Using Deep Learning and Periodic Pattern Mining
(2017-01-01)
Wong, Mun Hou
;
Tseng, Vincent S.
;
Tseng, Jerry C. C.
;
Liu, Sun-Wei
;
Tsai, Cheng-Hung
In recent years, with the advances in mobile communication and growing popularity of the fourth-generation mobile network along with the enhancement in location positioning techniques, mobile devices have generated extensive spatial trajectory data, which represent the mobility of moving objects. New services are emerged to serve mobile users based on their predicted locations. Most of the existing studies on location prediction were focused on predicting the next location of a user, which is regarded as short-term next location prediction. While more advanced location-based services could be enabled for the users if long-term location prediction could be achieved, the existing methods constrained in next-location prediction are not applicable for long-term prediction scenario. In this paper, we propose a novel prediction framework named LSTM-PPM that utilises deep learning and periodic pattern mining for long-term prediction of user locations. Our framework devises the ideology from natural language model and uses multi-step recursive strategy to perform long-term prediction. Furthermore, the periodic pattern mining technique is utilized to reduce the accumulated loss in the multi-step strategy. Through empirical evaluation on a real-life trajectory dataset, our proposed approach is shown to provide effective performance in long-term location prediction. To the best of our knowledge, this is the first work addressing the research topic on long-term user location prediction.
Mining Life Patterns from Wearable Sensors Data for Elderly Anomaly Detection
(2017-01-01)
Tsai, Cheng-Hung
;
Chu, Cheng-Hao
;
Liu, Sun-Wei
;
Hsieh, Sun-Yuan
;
Tseng, Vincent S.
Life patterns can represent an individual's life style and they can help people understand their daily behavior as well as the regular habits. Discovery of life patterns has a manifold of application scenarios, which can be embedded into location based recommender systems, precise advertising, computer-aided scheduling, kind care/alert systems. In this paper, we propose an approach for life style mining with applications on elderly anomaly detection. Although there existed already studies for discovering life styles, they were mostly based on traditional single-sensor environment. Consequently, it cannot completely represent an individual's lifestyle due to the lack of sufficient information and related applications like anomaly detection cannot reach high accuracy. To deal with above-mentioned problems, our approach can mine an individual's life pattern from wearable-devices-based environment with multiple sensors. When the life patterns are applied to elderly anomaly detection, multiple-sensors-based elderly's conditions, such as physical condition and locations, are taken into considerations at the same time. For experimental evaluations, we design a data simulator to generate sensors data of elderly's daily life, based on which the effectiveness of our proposed framework is verified.