Decision tree approach to rules extraction for human gait analysis

Authors:
Marcin Derlatka;Mikhail Ihnatouski
Affiliations:
Faculty of Mechanical Engineering, Bialystok Technical University, Bialystok, Poland;The Research Center of Resources Saving Problems, National Academy of Sciences of Belarus, Grodno, Belarus
Venue:
ICAISC'10 Proceedings of the 10th international conference on Artificial intelligence and soft computing: Part I
Year:
2010

Citing 3
Cited 0

Principles of data mining

Principles of data mining
Fuzzy logic based gait classification for hemiplegic patients

IDA'07 Proceedings of the 7th international conference on Intelligent data analysis
Kernel-Based method for automated walking patterns recognition using kinematics data

ICNC'06 Proceedings of the Second international conference on Advances in Natural Computation - Volume Part II

Quantified Score

Hi-index	0.00

Visualization

Abstract

The article presents the application of decision tree techniques to exploring barometric information obtained with instruments measuring the pressure of the human plantar onto contact surface while walking. The investigation was carried out on a group of 28 typical subjects as well as the subjects affected by Pes Planovalgus and Cerebral Palsy. The decision tree has been inducted by means of the vector of 255 values describing single stride with 51 samples per each of five zones of the human foot. The classification made by the resultant decision tree was correct for more then 94% strides. This allows to point the parameters which are the best discriminators between the investigated types of human gait.