Heart Disease Prediction using an optimized Extreme Learning Machine with Bacterial Colony optimization

Heart disease (HD) is often regarded most complex and life-threatening human infection. Recently, the extreme learning machine (ELM) method is used to handle a variety of health disease detection and prediction problems. The ELM algorithm, on the other hand, determines the input weights and biases at random. As a result, the random selection of connection biases and weight provides an unexpected result, resulting in an error and lower accuracy. In this paper, bacterial colony optimization (BCO) uses to optimize the weights and bias to address the above-mentioned drawbacks of ELM. Three separate metrics are used to calculate the performance of the BCO+ELM which is applied to heart disease prediction. The experimental effects show that the BCO+ELM produces better results when compared to other methods.