A Copula Statistic for Measuring Nonlinear Dependence with Application to Feature Selection in Machine Learning
Journal Title: International Journal of Advanced Computer Science & Applications - Year 2017, Vol 8, Issue 7
Abstract
Feature selection in machine learning aims to find out the best subset of variables from the input that reduces the computation requirement and improves the predictor performance. In this paper, a new index based on empirical copulas, termed the Copula Statistic (CoS) to assess the strength of statistical dependence and for testing statistical independence is introduced. It is shown that this test exhibits higher statistical power than other indices. Finally, the CoS is applied to feature selection in machine learning problems, which allow a demonstration of the good performance of the CoS.
Authors and Affiliations
Mohsen Ben Hassine, Lamine Mili, Kiran Karra
Keywords
Related Articles
- EP ID EP260212
- DOI 10.14569/IJACSA.2017.080720
- Views 126
- Downloads 0
MULTITHREADING IMAGE PROCESSING IN SINGLE-CORE AND MULTI-CORE CPU USING JAVA
Multithreading has been shown to be a powerful approach for boosting a system performance. One of the good examples of applications that benefits from multithreading is image processing. Image processing requires many re...
A Novel Secure Fingerprint-based Authentication System for Student’s Examination System
In the fingerprint image processing, various methods have been suggested as using band pass filter, Fouries transform filter and Fuzzy systems. In this paper, we present a useful and an applicable fingerprint security sy...
New Modified RLE Algorithms to Compress Grayscale Images with Lossy and Lossless Compression
New modified RLE algorithms to compress grayscale images with lossy and lossless compression, depending on the probability of repetition of pixels in the image and the pixel values to reduce the size of the encoded data...
Social Computing: The Impact on Cultural Behavior
Social computing continues to become more and more popular and has impacted cultural behavior. While cultural behavior affects the way an individual do social computing, Hofstede’s theory is still prevalent. The results...
Optimization of OADM DWDM Ring Optical Network using Various Modulation Formats
In this paper, the performance of the ring optical network is analyzed at bit rate 2.5 Gbps and 5 Gbps for various modulation formats such as NRZ rectangular, NRZ raised cosine, RZ soliton, RZ super Gaussian, RZ raised c...