A highly accurate firefly based algorithm for heart disease prediction

Rough sets and firefly algorithms is proposed to find optimal attribute reductions.Interval type-2 fuzzy logic system is used to predict heart disease.Proposed system is effective with results of fewer features and higher accuracy. This paper proposes a heart disease diagnosis system using rough sets based attribute reduction and interval type-2 fuzzy logic system (IT2FLS). The integration between rough sets based attribute reduction and IT2FLS aims to handle with high-dimensional dataset challenge and uncertainties. IT2FLS utilizes a hybrid learning process comprising fuzzy c-mean clustering algorithm and parameters tuning by chaos firefly and genetic hybrid algorithms. This learning process is computationally expensive, especially when employed with high-dimensional dataset. The rough sets based attribute reduction using chaos firefly algorithm is investigated to find optimal reduction which therefore reduces computational burden and enhances performance of IT2FLS. Experiment results demonstrate a significant dominance of the proposed system compared to other machine learning methods namely Naive Bayers, support vector machines, and artificial neural network. The proposed model is thus useful as a decision support system for heart disease diagnosis.