A method for evaluating data-preprocessing techniques for odour classification with an array of gas sensors

Abstract

The performance of a pattern recognition system is dependent on, among other things, an appropriate data-preprocessing technique, In this paper, we describe a method to evaluate the performance of a variety of these techniques for the problem of odour classification using an array of gas sensors, also referred to as an electronic nose. Four experimental odour databases with different complexities are used to score the data-preprocessing techniques. The performance measure used is the cross-validation estimate of the classification rate of a K nearest neighbor voting rule operating on Fisher's linear discriminant projection subspace.

Short term Sensor's Drift Compensation by using Three Drift Correction Techniques

“…강원대학교 전자정보통신공학부(Division of Electronics, Information & Communication Engineering, Kangwon National University) 5-Engineering Bldg., Kangwon National University, Joongang-ro Samcheok-si Gangwon-do, 245-711, Korea Fractional, 3가지 기법이 있으며, 이 중 Differential 방법이 실험 에 사용되었다 [6]. Table 1의 T와 동일하다.…”

Research on Pattern Recognition Applied for Volume Estimation of Blood Chamber with Matrix of Optical Sensors

“…For classi cation of the test samples to particular pattern class the kNN method is used [1,6]. The classi cation tests were performed for k = 5 KNN in reduced, 3D…”

“…A pattern recognition mechanism generally consists of several subsequent stages: data acquisition and preprocessing, dimensionality reduction (also termed feature extraction ) and classi cation [1].…”

Effective Gas Identification Model based on Fuzzy Logic and Hybrid Genetic Algorithms

“…Especially, analyzing data, which changes their properties over time, is very difficult because they include a lot of inconsistent and uncertain information. In case of handling these data, a method using the useful parts of the data can be effective but requires various analysis techniques, generally principal components analysis (PCA), Fisher's linear discriminant analysis (LDA), feature subset selection (FSS), C-means clustering, radial basis function classification, K-nearest neighbor classification and so on [1][2][3][4][5]. To apply the above techniques, in addition, complex or various numerical analyses are accompanied independently [6][7][8][9][10].…”