Human action recognition from multi-sensor stream data by genetic programming

Authors:
Feng Xie;Andy Song;Vic Ciesielski
Affiliations:
RMIT University, Melbourne, VIC, Australia;RMIT University, Melbourne, VIC, Australia;RMIT University, Melbourne, VIC, Australia
Venue:
EvoApplications'13 Proceedings of the 16th European conference on Applications of Evolutionary Computation
Year:
2013

Citing 10
Cited 0

Instance-Based Learning Algorithms

Machine Learning
C4.5: programs for machine learning

C4.5: programs for machine learning
Fast training of support vector machines using sequential minimal optimization

Advances in kernel methods
HOT SAX: Efficiently Finding the Most Unusual Time Series Subsequence

ICDM '05 Proceedings of the Fifth IEEE International Conference on Data Mining
Improvements to Platt's SMO Algorithm for SVM Classifier Design

Neural Computation
An analysis of adaptive windowing for time series forecasting in dynamic environments: further tests of the DyFor GP model

Proceedings of the 10th annual conference on Genetic and evolutionary computation
Recognizing multi-user activities using wearable sensors in a smart home

Pervasive and Mobile Computing
Estimating continuous distributions in Bayesian classifiers

UAI'95 Proceedings of the Eleventh conference on Uncertainty in artificial intelligence
Application of genetic programming for multicategory patternclassification

IEEE Transactions on Evolutionary Computation
A novel approach to design classifiers using genetic programming

IEEE Transactions on Evolutionary Computation

Quantified Score

Hi-index	0.00

Visualization

Abstract

This paper presents an approach to recognition of human actions such as sitting, standing, walking or running by analysing the data produced by the sensors of a smart phone. The data comes as streams of parallel time series from 21 sensors. We have used genetic programming to evolve detectors for a number of actions and compared the detection accuracy of the evolved detectors with detectors built from the classical machine learning methods including Decision Trees, Naïve Bayes, Nearest Neighbour and Support Vector Machines. The evolved detectors were considerably more accurate. We conclude that the proposed GP method can capture complex interaction of variables in parallel time series without using predefined features.