Advancement and applicability of classifiers for variant exponential model to optimize the accuracy for deep learning

The exploratory increment in database technology has facilitated researchers and scientist’s throughout the globe to determine best possible knowledge for discovery of hidden patterns and rules among large databases. Unfortunately, several technologies were intervened to measure the hidden patterns but tend to be incompetent, but soft computing techniques solely evaluated the different application domains and its success has potentially driven in prediction of future prognosis. In proposed study we have generalized our approach to discover a combinational model to measure the accuracy among the applicability of the classifiers. A soft computing solutions that we have utilized three different classifiers such as Random Forest, Naïve Bayes and K Nearest Neighbor with pancreatic cancer datasets utilizing varied training test data and ten fold cross validation techniques. Further, varied performance indicators were utilized to measure accuracy among the classifiers which include Area under Curve, F measure, Specificity and others. The Experimental results prove that the proposed approach can benefit end users to discriminate diversified method which can explicitly has potentially higher performance.