An Effective Machine Learning based Heart Disease Diagnosis Analysis

Heart disease has been the prime cause of death for many decades. In 2015, research was conducted to go through studies conducted in high-income areas. The studies were limited to those performed till May of that year. They were reviewed for data pulling. A strong correlation between social reclusion/ detachment and cardiac condition was found. Possible solutions were found to be educational campaigns & other communicative exercises. Further, it was understood via a separate study that faster diagnosis can help in starting the rehabilitation process much more efficiently with more chances of recovery. MLP or Multilayer Perceptrons were one of the ways to conduct such tests for early diagnosis/detection. In 2016, the correlation between sugar industries and coronary heart disease was studied in detail. Further such studies figured out the association of cause. In this paper, we are associating and basing our predictions on symptoms. In the same year (2016), a data prediction system based on data mining was sourced. With vast data present, applying ML to predict cardiac conditions became more manageable. Even in 2017, heart disease was the leading cause of death which took more than 1500000 lives. Recently, researchers have been looking into stress echocardiography & cell therapy both focusing on prediction & recovery respectively. Here in this paper, we are using some traditional ML techniques and applying hybridization to add to the process of fast prediction.