Optimal Data Mining Classification Algorithm for Bio Medicinal Facts

The most significant application of bio medicinal data analysis is used to determine or classify the unknown samples with the help of known samples to make a decision. Several general purpose data mining classification techniques have been proposed to identify the bio medicinal patterns. Because of their high dimensionality, multiple classes, unlearned data and hidden values, bio medicinal datasets pretense a distinctive challenge of machine learning and data mining algorithms for classification. This paper endow with a complete evaluation of a set of diverse machine learning schemes on a number of bio medicinal datasets. The intend of this paper shows that data mining can be applied to the medicinal databases, which will predict or classify the data with a reasonable accuracy also find the optimal classification algorithm for bio medicinal data sets. It focused to study a few classification techniques such as Support Vector Machine (SVM), Nearest Neighbor Classifier (k-NN), Decision Tree Induction, Navie Bayes, C4.5, Genetic Algorithm, Rule based algorithm and Back Propagation through bio medicinal data bases. These Classification Techniques applies to UCI, KDD, PUBMED and publicly available in common medicinal datasets, and compared how these classification techniques performed in class prediction of test datasets to find the best classifier for bio medicinal identification. The performance of these classifiers can evaluate in terms of minimum threshold level to identify the bio medicinal pattern, subset merit, noise, imbalance ratio, missing values, execution time, accuracy of an algorithm, correlation-based feature selection information gain, training time, memory usage, and memory utilization has been analyzed. The effect of these complexity measures on classification accuracy is evaluated using above mentioned data mining machine learning algorithms. Keywords— Bio medicinal, Machine learning, UCI, KDD, Lazy learner, SVM, Back propagation.