T-Patterns Revisited: Mining for Temporal Patterns in Sensor Data
TLDR
This work extends the T-Pattern algorithm, which was previously applied for detection of sequential patterns in behavioural sciences, to find patterns in temporal event data, with a statistical model to obtain a fast and robust algorithm.Abstract:
The trend to use large amounts of simple sensors as opposed to a few complex sensors to monitor places and systems creates a need for temporal pattern mining algorithms to work on such data. The methods that try to discover re-usable and interpretable patterns in temporal event data have several shortcomings. We contrast several recent approaches to the problem, and extend the T-Pattern algorithm, which was previously applied for detection of sequential patterns in behavioural sciences. The temporal complexity of the T-pattern approach is prohibitive in the scenarios we consider. We remedy this with a statistical model to obtain a fast and robust algorithm to find patterns in temporal data. We test our algorithm on a recent database collected with passive infrared sensors with millions of events.read more
Citations
More filters
Journal ArticleDOI
Critical infrastructure protection
TL;DR: The vulnerabilities and threats facing modern critical infrastructures with special emphasis on industrial control systems are explored, and a number of protection measures are described.
Journal ArticleDOI
Learning routines over long-term sensor data using topic models
TL;DR: In this work, topic models are employed to learn the latent structure and dynamics of sensor network data and have shown the ability to find routines of activity over sensor networkData in office environments.
Journal ArticleDOI
Improving Activity Recognitionby Segmental Pattern Mining
Umut Avci,Andrea Passerini +1 more
TL;DR: A novel approach for introducing long-range interactions based on sequential pattern mining and enriching sensor-based representations with the mined patterns allows improving results over sequential and segmental labeling algorithms in most of the cases.
Proceedings ArticleDOI
Building an occupancy model from sensor networks in office environments
TL;DR: The work presented here aims to answer the question: Using just binary occupancy sensors is it possible to build a behaviour occupancy model over long-term logged data and shows the power of the LDA model in extracting relevant patterns from sensor network data.
Journal ArticleDOI
Spatio-Temporal Constrained Human Trajectory Generation from the PIR Motion Detector Sensor Network Data: A Geometric Algebra Approach.
TL;DR: A geometric algebra (GA)-based approach is developed to generate all possible human trajectories from the PIR sensor network data and can effectively extract all possible trajectories of the human motion, which makes it more accurate.
References
More filters
Journal ArticleDOI
Observation of nurse-patient interaction in oncology: review of assessment instruments.
TL;DR: A systematic search of the literature revealed a variety of methods and instruments applicable to studies recording nurse-patient interaction that offered valuable information on observational research in general, on procedures relating to informed consent and observational arrangements in nursing practice.
Journal ArticleDOI
The “Wireless Sensor Networks for City-Wide Ambient Intelligence (WISE-WAI)” Project
Paolo Casari,Aldo Castellani,Angelo Cenedese,Claudio Lora,Michele Rossi,Luca Schenato,Michele Zorzi +6 more
TL;DR: This paper gives a detailed technical overview of some of the activities carried out in the context of the “Wireless Sensor networks for city-Wide Ambient Intelligence (WISE-WAI)” project, funded by the Cassa di Risparmio di Padova e Rovigo Foundation, Italy.
Journal ArticleDOI
Energy efficient strategies for object tracking in sensor networks: A data mining approach
Vincent S. Tseng,Kawuu W. Lin +1 more
TL;DR: This paper proposes a novel data mining algorithm named TMP-Mine with a special data structure named T MP-Tree for efficiently discovering the temporal movement patterns of objects in sensor networks and proposes novel location prediction strategies that utilize the discovered temporalmovement patterns so as to reduce the prediction errors for energy savings.
Journal ArticleDOI
Energy-efficient real-time object tracking in multi-level sensor networks by mining and predicting movement patterns
TL;DR: This paper proposes a novel strategy named multi-level object tracking strategy (MLOT) for energy-efficient and real-time tracking of the moving objects in sensor networks by mining the movement log by conducting hierarchical clustering to form a hierarchical model of the sensor nodes.
Proceedings ArticleDOI
A new constraint for mining sets in sequences
TL;DR: A new constraint based on a new interestingness measure combining the cohesion and the frequency of a pattern is introduced, for a dataset consisting of a single sequence and a similar constraint for datasets consisting of multiple sequences.