Enhanced Fuzzy based Deep Learning Prediction Model for Biological Data Applications

Biological data often deals with class imbalance problem. For the most part, information examination related issues in bioinformatics can be separated into three classes as indicated by the sort of biological data: sequences, structures, and networks. Classification and clustering strategies of data mining plays a critical part to dissect biological data such as genomic/DNA microarray data classification and analysis. Learning from imbalanced datasets is a common problem found in many bioinformatics applications, such as gene prediction, splice site prediction, promoter prediction, protein classification and many more. Improving accuracy, precision, recall and F-1 score is the primary objective of this research work. Also, analyzing the performance of the classifiers (existing and proposed) under injection of noisy data is aimed. The evaluation metrics prove that proposed model  performs better than that of existing classifiers.