MLT finds potentially useful patterns in the data. Three MLT deployed for the diabetes mellitus prediction is presented subsequently with a brief on proposed method, experimental set up, test results and performance comparison. The proposed classifiers are tested with the original dataset. The results are recorded first. Subsequently the dataset will be subject to cluster and the this will be the first fold of the proposed technique. In the expansion step the assigned cluster will be a separate

more »

... instance in the dataset. This will be the second fold of the proposed technique. Classification will be deployed as the third fold of the proposed technique. This proposed three fold integrated clustering-classification technique for diabetes mellitus prediction significantly improves the performance of the diabetes mellitus prediction. After the proposed strategy is carried out, results are recorded and compared.