Hybrid Optimization Algorithm Based on Modified Genetic Algorithm and Back Propogation |
Author(s): |
Sushil Kumar Janardan , SHRI SHANKARACHARYA TECHNICAL CAMPUS ; Dr. Abha Choubey, SSTC BHILAI |
Keywords: |
Classification, UCI Data Base, Random Forest, Rule-Based Classifiers, Nearest Neighbors, Partial Least Squares and Principal Component Regression, Neural Network, Modified Genetic Algorithm and Back Propagation |
Abstract |
We evaluate 21 classifiers arising from 10 families (neural network, rule based classifiers, nearest neighbors, partial least squares and principal component regression, logistic and multinomial regression, multiple adaptive regression splines and other methods) executed in (with and without the caret bundle), C and Matlab, including all the significant classifiers accessible today. We use 40 data sets, which represent the whole UCI data base (excluding the large-scale problems) and other own real problems, in order to achieve significant conclusions about the classifier behavior, not dependent on the data set collection. The classifiers well on the way to be the bests are the irregular backwoods (RF) Adaptations, the distinction is not measurably significant with the second best, the Genetic Algorithm with Feature Selection portion actualized in JAVA Execution Jar File utilizing GANN-FS, which accomplishes 83.5% of the most extreme exactness. |
Other Details |
Paper ID: IJSRDV4I110065 Published in: Volume : 4, Issue : 11 Publication Date: 01/02/2017 Page(s): 70-73 |
Article Preview |
|
|