Pattern recognition for insect behavior by wrapper approach to subset selection of wavelet-multifractal attributes

The goal of this paper was to apply data mining subset selection techniques and wavelet-multifractal to describe insect behavior. It was proposed wavelet modulus maxima to extract multifractal parameters of sound attributes for pattern recognition of an insect behavior. Wrapper data mining approach was used to select relevant attributes. It has been found that, in general, wavelet-multifractal-based schemes perform better for sound, particularly in terms of minimizing noise distortion influence. The results from wavelet-multifractal-based method can also be improved by applying different mother wavelets; however, these schemes often have greater set-up requirements.