A Study on Activity Predictive Modeling for Prompt and Delayed Services in Smart Space

Knowledge about position and activities of the participants is commonly used in location-based services and applications in smart environment, which need to know an approximated location of the users to provide a proper service. Assistive services provided in the smart space can be divided into two categories: prompt services and delayed services. Prompt services can be served instantly and require no preparation time. On the other hand, delayed services need some time to prepare and be ready. When users are moving in an environment doing tasks, knowledge of the next location or destination of those movements can be used to assist the system to give more accurate system responses for the prompt services that can be served right away when the user arrives in his/her destination. These services require the following knowledge to operate: 1) a predicted location of the users or a plausible destination, and 2) a predicted time of arrival. For these two requirements, a predictive temporal sequential pattern mining algorithm is proposed in this paper, which is a method aimed at predicting the next location of a moving object from its temporal and spatial context. The prediction uses previously extracted temporal sequential patterns, which represent behaviors of moving objects as sequences of locations frequently visited within a certain speed. A decision tree based classifier is trained from the temporal sequential patterns and used as a predictor for the next location that is most probable location to be visited within the movement sequence. Moreover, A preliminary study on a predictive model for delayed services is also discussed in this paper. Finally, a performance evaluation of the methods tested over a real dataset is presented.

[1]  Anna Monreale,et al.  WhereNext: a location predictor on trajectory pattern mining , 2009, KDD.

[2]  Jiawei Han,et al.  Frequent pattern mining: current status and future directions , 2007, Data Mining and Knowledge Discovery.

[3]  Ramakrishnan Srikant,et al.  Mining Sequential Patterns: Generalizations and Performance Improvements , 1996, EDBT.

[4]  Masayuki Numao,et al.  Theory and Practice of Computation , 2012, Proceedings in Information and Communications Technology.

[5]  Masayuki Numao,et al.  Time-Interval Clustering in Sequence Pattern Recognition as Tool for Behavior Modeling , 2012 .

[6]  Alberto Maria Segre,et al.  Programs for Machine Learning , 1994 .

[7]  Hayato Yamana,et al.  Generalized Sequential Pattern Mining with Item Intervals , 2006, J. Comput..

[8]  Yen-Liang Chen,et al.  Discovering fuzzy time-interval sequential patterns in sequence databases , 2005, IEEE Transactions on Systems, Man, and Cybernetics, Part B (Cybernetics).

[9]  Ramakrishnan Srikant,et al.  Mining sequential patterns , 1995, Proceedings of the Eleventh International Conference on Data Engineering.

[10]  Georges Gardarin,et al.  Advances in Database Technology — EDBT '96 , 1996, Lecture Notes in Computer Science.

[11]  Ming-Tat Ko,et al.  Discovering time-interval sequential patterns in sequence databases , 2003, Expert Syst. Appl..

[12]  Dino Pedreschi,et al.  Trajectory pattern mining , 2007, KDD '07.

[13]  Mohammed J. Zaki,et al.  SPADE: An Efficient Algorithm for Mining Frequent Sequences , 2004, Machine Learning.

[14]  Jianyong Wang,et al.  Mining sequential patterns by pattern-growth: the PrefixSpan approach , 2004, IEEE Transactions on Knowledge and Data Engineering.

[15]  Alaaeldin M. Hafez,et al.  Association mining of dependency between time series , 2001, SPIE Defense + Commercial Sensing.

[16]  J. Ross Quinlan,et al.  C4.5: Programs for Machine Learning , 1992 .

[17]  Yen-Liang Chen,et al.  On mining multi-time-interval sequential patterns , 2009, Data Knowl. Eng..

[18]  J. MacQueen.  Some methods for classification and analysis of multivariate observations , 1967 .