사용자의 의미장소 추출 및 인식을 위한 스마트폰 라이프로그 마이닝
Smartphone Life-log Mining for Extracting and Recognizing User’s Meaningful Locations
0
조회
0
다운로드
0
인용
0
참고문헌
키워드
초록
최근 사용자의 장소를 기반으로 하는 다양한 어플리케이션이 증가하고 있지만, 장소정보를 측정하기 위한 GPS 센서는 실외 환경에서 비교적 높은 오차 때문에 정확한 장소를 인식하기 힘들고 실내위치는 파악하기 어렵다. GPS를 대신하여 Wi-Fi정보를 이용하는 WPS(Wi-Fi Positioning System)가 제안되었지만 높은 전력 소모량과 오차율을 보이는 문제가 있다. 본 논문에서는 사용자의 장소 변화 시퀀스를 마이닝하여 자동으로 현재 장소를 인식하는 방법을 제안한다. 이 방법은 클러스터링된 사용자의 GPS데이터에서 의미장소를 추출한 후, 시계열 확률 모델인 동적 베이지안 네트워크를 이용하여 장소들간의 사용자 경로를 모델링하고 현재 장소를 인식한다. 제안하는 방법은 GPS센서와 불확실성이 존재하는 모바일 환경에서 사용자가 의미장소를 재방문하였을 때 학습된 모델을 이용하여 현재 위치를 인식한다. 제안하는 장소인식 모델의 유용성을 입증하기 위하여 한달간 5명의 사용자로부터 획득된 모바일 데이터를 기반으로 실험한 결과 88.67%의 정확도를 확인하였다.
Abstract
Recently, various applications that require user’s location information are rapidly increasing. However, the GPS sensor in smartphone is unable to accurately determine the exact location since error rate is considerably high depending on the external environment. WPS (Wi-Fi Positioning System) collects the location information in stead of GPS, but scanning for surrounding Wi-Fi signals results in high energy consumption and high error rate. In this paper, we propose a method to recognize user’s location by modelling the location trajectories, in which user’s GPS data are clustered to extract semantic locations. User’s trajectories are modelled by DBN (dynamic Bayesian network), a time-series probabilistic model, which makes use of the various contexts acquired from built-in smartphone sensors. The model facilitates to recognize current location efficiently when a user revisits a location even if GPS sensor signals and mobile environment are noisy. In order to show the usefulness of the proposed method, we have conducted experiments using mobile data collected from five users in a month, resulting in location recognition accuracy of 88.67%.
참고문헌 (32)
- Z. Zhuang, "Improving energy efficiency of location sensing on smartphones", Proc. of the 8th Int. Conf. Mobile Systems, Applications, and Services, pp.315~330 (2010)
- J. Paek, "Energy-efficient rate-adaptive GPS-based positioning for smartphones", Proc. of the 8th Int. Conf. on Mobile Systems, Applications, and Services, pp.299~314 (2010)
- K. Akeyama, "Trajectory estimation improvement based on time-series constraint of GPS Doppler and INS in urban areas", Proc. of IEEE, 2012, 700~705
- D. Dearman, "An exploration of location error estimation", Proc. of UbiComp 2007: Ubiquitous Computing, pp.181~198 (2007)
- J. Hightower, "Location systems for ubiquitous computing", IEEE Computer, 34(8), pp.57~66 (2001)
- S. Bell, "WiFi-based enhanced positioning systems: Accuracy through mapping, calibration, and classification", Proc. of the 2nd ACM SIGSPATIAL Int. Workshop on Indoor Spatial Awareness, pp.3~9 (2010)
- Y. Wang, "A framework of energy efficient mobile sensing for automatic user state recognition", Proc. of the 7th Int. Conf. on Mobile systems, Applications, and Services, pp.179~192 (2009)
- R. Stumptner, "Using GPS trajectories to create a dynamic network of significant locations as an abstraction of road maps", Computer Aided Systems Theory, pp.161~168 (2012)
- W. H. Liao, "GRID:A fully location-aware routing protocol for mobile ad hoc networks", Telecommunication Systems, 18, pp.37~60 (2001)
- J. H. Kang, "Extracting places from traces of locations", Proc. of the 2nd ACM Int. workshop on Wireless Mobile Applications and Services on WLAN Hotspots, pp.110~118 (2004)
- J. Hightower, "Learning and recognizing the places we go", Proc. of UbiComp 2005: Ubiquitous Computing, pp.903~903 (2005)
- C. Zhou, "Discovering personally meaningful places:An interactive clustering approach", ACM Trans. on Information Systems, 25(3), pp.1~31 (2007)
- P. Nurmi, "Identifying meaningful places: The Non-parametric way", IEEE Pervasive Computing, pp.111~127 (2008)
- A. J. Dou, "Data clustering on a network of mobile smartphones", Proc. of IEEE, 2011, 118~127
- N. Brouwers, "Dwelling in the canyons: Dwelling detection in urban environments using GPS, Wi-Fi, and geolocation", Pervasive and Mobile Computing, pp.1~16 (2012)
- A. W. Parvez, "Data set property based 'K' in VDBSCAN clustering algorithm", World of Computer Science and Information TechNology Journal, 2(3), pp.115~119 (2012)
- D. Ashbrook, "Using GPS to learn significant locations and predict movement across multiple users", Personal and Ubiquitous Computing, 7, pp.275~286 (2003)
- G. Yavas, "A data mining approach for location prediction in mobile environments", Data & KNowledge Engineering, 54, pp.121~146 (2005)
- S. Gambs, "Next place prediction using mobility Markov chains", Proc. of the 1st Workshop on Measurement, Privacy, and Mobility, pp.1~6 (2012)
- C. Zhou, "Discovering personally meaningful places:An interactive clustering approach", ACM Trans. on Information Systems, 25(3), pp.1~31 (2007)
- G. Hamerly, "Learning the k in k-means", Proc. of Conf. on Neural Information Processing System, pp.281~288 (2004)
- M. Kim, "Extracting a mobility model from real user traces", Proc. of IEEE Infocom, pp.1~13 (2006)
- S. Scellato, "Nextplace: A spatio-temporal prediction framework for pervasive systems", Proc. of Int. Conf. on Pervasive Computing, pp.152~169 (2011)
- J. -K. Min, "A hybrid context-aware wearable system with eVolutionary optimization and selective inference of dynamic Bayesian networks", Lecture Notes in Computer Science, 6678, pp.444~451 (2011)
- N. Friedman, "Learning the structure of dynamic probabilistic networks", Proc. of Conf. on Uncertainty in Artificial Intelligence, pp.139~147 (1998)
- T. Choudhury, "Boosting and structure learning in dynamic Bayesian networks for audio-visual speaker detection", Proc. of Int. Conf. on Pattern Recognition, pp.789~794 (2002)
- G. D. Kleither, "Propagating imprecise probabilities in Bayesian networks", Artificial Intelligence, 88(1), pp.143~161 (1996)
- G. F. Cooper, "A Bayesian method for the induction of probabilistic networks from data", Machine Learning, 9, pp.309~347 (1992)
- 오근현, "표현형 기반 종분화 진화 신경망을 이용한 다양한 로봇 행동 패턴 생성", 정보과학회 컴퓨팅의 실제 논문지, 17(5), pp.332~336 (2011), 한국정보과학회
- K. Murao, "A context-aware system that changes sensor combinations considering energy consumption", Proc. of Int. Conf. on Pervasive Computing, pp.197~212 (2008)
- J.-C. Ying, "Semantic trajectory mining for location prediction", ACM SIGSPATIAL GIS, pp.34~43 (2011)
- 이영설, "결정트리와 베이지안 네트워크를 이용한 효율적인 안드로이드폰 에너지 관리 프레임워크", 정보과학회논문지 : 시스템 및 이론, 39(3), pp.209~215 (2012), 한국정보과학회