The comparing of classification algorithm performance for identify brain patient to need an Brain Imaging Process

This paper presents the performance of classification algorithms that have been popular in general also include Neural Network, Naïve Bayes and Decision Tree is working with real data in real work situations for selection algorithm that is suitable to develop an algorithm to create an effective data classification having more precise and can be explained the model from work to understand. The application of Q-Gram algorithm to help create a new data set for testing the performance of a classification algorithm. Q-gram can help reduce errors in information of the brain patients being treated are caused by the recording error from the user or increase completeness of the information from the syntax errors language or spelling mistakes as well. The experimental results showed that the decision tree “Random Forest” effectively based on the classification algorithm is higher than others. The accuracy of 79.2% and a model of the algorithm can be used to communicate with the common people can be easily understood by that the brain patient groups that should be entered into the Brain Imaging when compared with other algorithms tested.