Plasmodium falciparum causes malaria, which is an infectious and fatal disease. In early days, malaria-infected cells were diagnosed using a microscope. owing to a huge number of instances for analysis and intricacy of time, it may lead to false detection. Automated parasite detection technologies are in high demand due to increased time consumption and erroneous detection. To create effective cures and treatments, it is critical to use an accurate approach for predicting malaria parasite. Here, numerous protein sequences formulation techniques namely: discrete methods, Biochemical, physiochemical and Natural language processing techniques are applied for transformation of protein sequences in to numerical descriptors. Four classification algorithms are utilized and the anticipated results of these classifiers were then fused to establish ensemble classification model via simple majority and genetic algorithm. In addition, BCH error correction code is incorporated with support vector machine using all the feature spaces. The simulated results demonstrate the remarkable achievement of proposed compared to previous models. Thus, our proposed model may be an effective tool for discriminating the secretory and non-secretory proteins of malaria parasite.
Keywords: BCH code; Ensemble classification; Malaria parasite; NLP; PSSM; SVM.
Copyright © 2022 Elsevier Ltd. All rights reserved.