Predicting Autism Using Naive Bayesian Classification Approach

The present era contains large quantity of data in the medicine field. The field of medicine is “data rich” which can be utilized in decision making. Genetics plays a major role in medicine. Large amount of data can be obtained from the human genome. Autism is a neurobehavioral disorder that impairs the ability to communicate and interact. Autism has a strong genetic basis. Mutations in the genes associated with autism can affect the brain development and its functions, starting well before birth. Mutated genes are inherited from the parents to child which becomes a parameter for autism. Herein we propose an approach to rank the set of diseased genes which is present in the autistic individual. From the training data set, the system predicts the autistic behavior of an individual by comparing the similarities between the individual's gene and the diseased gene in the training set. This is implemented using NAIVE BAYESIAN CLASSIFICATION method. For an instance, if an individual's DNA is tested and if the DNA contains the diseased genes which are present in the training set then the autism is predicted. This approach makes the diagnosis of autism easier and at an early stage.