• Azhagammal
• Dr. k. RubaSoundar

Due to the technology development, the data’s are increased in their volume, velocity, and variety, referred to as large data or Big Data. From this large data, we can gain knowledge by using techniques like classification, association rule learning, outlier detection, clustering, and regression. By definition, classification canvass the specific set of data, procreate a set of grouping rules which can be used to codify the future data. The aim of classification is to meticulously forecast the target class for each case in the data. The classification stint starts with a dataset in which the class assignments are known. Available methods for classifications are support vector machines, quadratic classifiers, k-nearest neighbor, boosting, decision trees, neural network, learning vector quantization. From the above list, we use Neural Network for classification in this paper, Neural network is one of the most effective methods for classification which are relatively rude electronic networks of neuron based on the neural structure of the brain. Neural network technique is either supervised or unsupervised. The supervised method contains algorithms like Back Propagation, Feed Forward, and so many. A deep neural network comprises more than one hidden layers which actually improves the accuracy in classification. The Deep belief network(DBN) is an unsupervised learning. It includes more than one trained RBMs. In this paper, we implement DBN in a large dataset (Readmission rate of Diabetic patients in medical data) and observe the accuracy based on different test cases.

Deep Belief Network, DBN, Classification, large Dataset

"Classification with Deep Belief Network for large Dataset", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.5, Issue 10, page no.89-95, October-2018, Available :http://www.jetir.org/papers/JETIR1810713.pdf

"Classification with Deep Belief Network for large Dataset", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.5, Issue 10, page no. pp89-95, October-2018, Available at : http://www.jetir.org/papers/JETIR1810713.pdf