Main Article Content
Diabetes mellitus is one among the complex diseases for which specific causes have not yet been identified. Nevertheless, many medical science researchers believe that complex diseases are caused by environmental, genetic and abnormal cholesterol and triglyceride levels. Detection of such diseases becomes an issue because it is not free from false presumptions and is accompanied by unpredictable effects. To solve this problem an existing system introduced multiple classifier approach base type-2 diabetes mellitus detection. In this system, we introduced a voting scheme which is dynamic called multiple factors weighted combination for classifiers’ decision combination. However, it does integrate the genetic information and cannot discover complex disease more accurately. To solve this problem the proposed system is introduced a sequential pattern mining approach which is called Frequent Pattern growth approach. The main objective of the sequential pattern algorithm is to check and mine data sets based on the sequential order. Based on the gene sequence structure the sequence pattern algorithm discovers the set of frequent sub sequences in the dataset. The minimum support count value is identified to produce interesting patterns which satisfy the conditions. Hence this algorithm is used to detect the complex disease more accurately. The experimental results show that the proposed system achieves high performance compared with the existing system in terms of accuracy, precision, recall and fmeasure.