In recent years, many researchers are focusing on computational intelligence-based biomedical system to solve the problematic medical issues and to attain higher accuracy in predicting various diseases. Associative classification technique is one of the essential supervised learning methods that integrate association rule mining and classification to achieve higher accuracy. Heart disease is a major cause of death in many developing countries. So, this paper provides an efficient weighted decision support system for identifying heart disease by using an evolutionary memetic algorithm based on random walk algorithm. The experiments are carried out on heart disease data set, which has been collected from UCI repository. Various well-known validation techniques have been employed to evaluate the proposed system efficiency in the field of biomedical, which helps the physicians to diagnose the heart disease in the patient at early stage for starting further treatment. © 2020, Springer Nature Singapore Pte Ltd.