Abstract

Classification algorithms require training data initially labelled by classes to build a model and then to be able to classify the new data. The amount and diversity of training data affect the classification quality and usually the larger the training set, the better the accuracy of classification. In many applications only small amounts of training data are available. This article presents a new time series classification algorithm for problems with small training sets. The algorithm was tested on hand gesture recordings in tasks of person identification and gesture recognition. The algorithm provides significantly better classification accuracy than other machine learning algorithms. For 22 different hand gestures performed by 10 people and the training set size equal to 5 gesture execution records per class, the error rate for the newly proposed algorithm is from 37% to 75% lower than for the other compared algorithms. When the training set consists of only one sample per class the new algorithm reaches from 45% to 95% lower error rate. Conducted experiments indicate that the algorithm outperforms state-of-the-art methods in terms of classification accuracy in the problem of person identification and gesture recognition.

Highlights

  • Classification algorithms, an important tool in Computational Intelligence Methods and Statistical learning [1,2,3], are widely used in many areas, for example biometrics, economic trend analysis, human-computer interfaces, medical diagnostics, etc

  • The common way to construct a person identification or gesture recognition system based on hand gestures is to collect, with the respect to overfitting, as a large database, as required to reach satisfactory values of the classification coefficients accompanying the receiver operating characteristic (ROC) curve

  • The comparison of the new algorithm and other classification methods optimized with respect to the mean error rate there is presented as follows

Read more

Summary

Introduction

Classification algorithms, an important tool in Computational Intelligence Methods and Statistical learning [1,2,3], are widely used in many areas, for example biometrics, economic trend analysis, human-computer interfaces, medical diagnostics, etc These methods include random forest [3], k-nearest neighbour (kNN) [4], probabilistic neural network (PNN) [5], multi-layer perceptron (MLP) [6], support vector machine (SVM) [7], Gaussian processes [8], adaptive neuro-fuzzy inference system (ANFIS) [9], decision trees [3], radial basis function-based neural network (RBF NN) [10], generalized regression neural network (GRNN) [5] as well as siamese neural network (SNN) [11]. The real user may not be so patient, or may be an elder person, or be a disabled person which limits the possibility to record a large number of repetitions of a single gesture, and the biometric acceptance factor [12] might be lowered

Methods
Results
Discussion
Conclusion

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call

Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.