Cardiovascular diseases are some of the most common cases in the United States. Whether it be High Blood Pressure, Coronary Artery Disease or Cardiac Arrest these diseases are unpredictable at an early stage. This makes it difficult to prevent and can affect 1 in every 4 people in the United States according to Centers for Disease Control and Prevention.
Associative Classification is a way to organize subjects by specificity to a required subject. Along with this a genetic algorithm is used to predict, in this case, if an individual inherited qualities from an initial population.
In the image above a sequence of steps is being used to compare genetic data. You begin with an initial population and target something specific. Then you crossover with your following data. If a mutation is present that is an indication of your target gene meaning it is more likely for you to have a specific disease.
Heart disease is a result of a combination of multiple circumstances. Blood pressure, age, height, weight muscle activity among other things determine your likeliness of being a candidate for a certain disease. By comparing association rules through a decision tree and other techniques, researchers examined a standard model used to predict heart disease. They divided the data into partitions and applied an association rule to classify whether or not an individual can develop a cardiac disease in that area of the heart given their circumstances. This research has mainly been applied in India where the mortality rate based on cardiac illness is 30% in the countryside making it a necessity to incorporate a machine learning technique with such accuracy before that rate rises.
To learn more about how associative learning is processed compared to another technique watch the video below: