Improving the classification accuracy of streaming data using SAX similarity features

Authors:
Pekka Siirtola;Heli Koskimäki;Ville Huikari;Perttu Laurinen;Juha Röning
Affiliations:
Computer Science and Engineering Laboratory, P.O. BOX 4500, FI-90014, University of Oulu, Finland;Computer Science and Engineering Laboratory, P.O. BOX 4500, FI-90014, University of Oulu, Finland;Computer Science and Engineering Laboratory, P.O. BOX 4500, FI-90014, University of Oulu, Finland;Computer Science and Engineering Laboratory, P.O. BOX 4500, FI-90014, University of Oulu, Finland;Computer Science and Engineering Laboratory, P.O. BOX 4500, FI-90014, University of Oulu, Finland
Venue:
Pattern Recognition Letters
Year:
2011

Citing 25
Cited 2

An introduction to symbolic dynamics and coding

An introduction to symbolic dynamics and coding
Models and issues in data stream systems

Proceedings of the twenty-first ACM SIGMOD-SIGACT-SIGART symposium on Principles of database systems
What Shall We Teach Our Pants?

ISWC '00 Proceedings of the 4th IEEE International Symposium on Wearable Computers
Optimizing time series discretization for knowledge discovery

Proceedings of the eleventh ACM SIGKDD international conference on Knowledge discovery in data mining
Symbolic time series analysis for anomaly detection: a comparative evaluation

Signal Processing
Gesture spotting using wrist worn microphone and 3-axis accelerometer

Proceedings of the 2005 joint conference on Smart objects and ambient intelligence: innovative context-aware services: usages and technologies
New Time Series Data Representation ESAX for Financial Applications

ICDEW '06 Proceedings of the 22nd International Conference on Data Engineering Workshops
Activity Recognition of Assembly Tasks Using Body-Worn Microphones and Accelerometers

IEEE Transactions on Pattern Analysis and Machine Intelligence
Symbolic time series analysis via wavelet-based partitioning

Signal Processing - Special section: Distributed source coding
Experiencing SAX: a novel symbolic representation of time series

Data Mining and Knowledge Discovery
Feature Extraction from Sensor Data Streams for Real-Time Human Behaviour Recognition

PKDD 2007 Proceedings of the 11th European conference on Principles and Practice of Knowledge Discovery in Databases
Using acceleration measurements for activity recognition: An effective learning algorithm for constructing neural classifiers

Pattern Recognition Letters
Gestures are strings: efficient online gesture spotting and classification using string matching

Proceedings of the ICST 2nd international conference on Body area networks
Privacy Preserving Pattern Discovery in Distributed Time Series

ICDEW '07 Proceedings of the 2007 IEEE 23rd International Conference on Data Engineering Workshop
Using rhythm awareness in long-term activity recognition

ISWC '08 Proceedings of the 2008 12th IEEE International Symposium on Wearable Computers
Activity recognition using a wrist-worn inertial measurement unit: A case study for industrial assembly lines

MED '09 Proceedings of the 2009 17th Mediterranean Conference on Control and Automation
Activity recognition from accelerometer data

IAAI'05 Proceedings of the 17th conference on Innovative applications of artificial intelligence - Volume 3
Mining an optimal prototype from a periodic time series: an evolutionary computation-based approach

CEC'09 Proceedings of the Eleventh conference on Congress on Evolutionary Computation
Recognition of hand movements using wearable accelerometers

Journal of Ambient Intelligence and Smart Environments
SE-155 DBSA: a device-based software architecture for data mining

Proceedings of the 2010 ACM Symposium on Applied Computing
Discriminative temporal smoothing for activity recognition from wearable sensors

UCS'07 Proceedings of the 4th international conference on Ubiquitous computing systems
Child-activity recognition from multi-sensor data

Proceedings of the 7th International Conference on Methods and Techniques in Behavioral Research
Feature selection and activity recognition from wearable sensors

UCS'06 Proceedings of the Third international conference on Ubiquitous Computing Systems
Finding Unusual Medical Time-Series Subsequences: Algorithms and Applications

IEEE Transactions on Information Technology in Biomedicine
Detection of Daily Activities and Sports With Wearable Sensors in Controlled and Uncontrolled Conditions

IEEE Transactions on Information Technology in Biomedicine

SAPHE: simple accelerometer based wireless pairing with heuristic trees

Proceedings of the 10th International Conference on Advances in Mobile Computing & Multimedia
Time series visualization based on shape features

Knowledge-Based Systems

Quantified Score

Hi-index	0.12

Visualization

Abstract

The classification accuracy of time series is highly dependent on the quality of used features. In this study, features of new type, called SAX (Symbolic Aggregate approXimation) similarity features, are presented. SAX similarity features are a combination of the traditional statistical number-based and the template-based classification. SAX similarity features are obtained from the data of the time window by first transforming the time series into a discrete presentation using SAX. Then the similarity between this SAX presentation and predefined SAX templates are calculated, and these similarity values are considered as SAX similarity features. The functioning of these features was tested using five different activity data sets collected using wearable inertial sensors and five different classifiers. The results show that the recognition rates calculated using SAX similarity features together with traditional features are much better than those obtained employing traditional features only. In 20 tested cases out of 23, the improvement is statistically significant according to the paired t-test.