• Jitendra Soni

The deranged dataset is one in which a class is represented with very few example, this reduces efficiency of classifier, to be a deranged dataset in any case one class should be represented with very few examples, no of solutions like data case, bagging, cost assessment of model Genetic Programming (GP) based some methods been proposed in the papers. Researchers have introduced many methods to improve the efficiency of classifier for deranged data. In this paper, a function is planned to handle data disparity problem, by modifying learning algorithm but the original dataset remain intact. Based on Darwin’s theory of natural selection, GP as a learning algorithm used to evolve classifiers by applying various GP operators. Distance Parameter is introduced to address data disparity problem. Distance parameter classification is achieved so that performance of classifier is measured in each class of datasets. Fitness value calculated for data disparity problem is taken as input along with parameter where values of parameters are predefined, to evolve in number of generations those classifier having less number of nodes with good fitness values are preferred. This paper represents various problems with their solutions introduced by researchers to improve the performance of classifier, various advantages and disadvantage of different techniques are studied to make reasonable comparison between them.

Dataset, Classifier, Fitness function, Data Sampling. Code Blot

"Novel approach to resolve data disparity problem in Genetic Programming", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.5, Issue 1, page no.23-26, January-2018, Available :http://www.jetir.org/papers/JETIR1801321.pdf

"Novel approach to resolve data disparity problem in Genetic Programming", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.5, Issue 1, page no. pp23-26, January-2018, Available at : http://www.jetir.org/papers/JETIR1801321.pdf