Abstract:

In bio medical science, health disorders and their characteristics have a huge relationship with the gene expressions. The key elements used for diagnosis and prediction of health disorder. The data mining technique has a huge impact and application in human genetics and gene sequence data analysis. The huge size of data in the electronic format is considered as the big data. The storing, transferring and mining of genetic information within a big data are posed to be the current challenges in the process of huge data analysis. Classification of gene database is one of the most fundamental but yet a challenging problem that exists in the field of bio medical engineering and bioinformatics. There are a number of competent gene classification models which exist in current practice. These classification models in general have been used for natural language processing text classification, image recognitions, data prediction, reinforcement training etc. Some materials and methods required in this research are Association Rules, Clustering genes, Correlation Clusters in Gene Sequence, Cluster Editing in Gene sequences, Correlation Based Clustering, Logistic Regression. This technology was followed by Support Vector Machine Classification which reduced the execution time but still inquired high computational cost due to increased number of iterations. The salient features of the proposed technique Correlation Based Clustering Algorithm include fastest execution time, it has reduced cost considerably. It has improved the accuracy of result and reduced the number of rules.

Keywords: