• Neha sharma
• Deepak kumar

Drug discovery using text mining is the antidote for laying the stone for accurate classification of diabetic disease. The objective of this study is to uncover diabetic based on medicines prescribed to patients using the applications of natural language processing and text mining. To this end, proposed work is partitioned into phases. First phase, pre-processing yield purified data by eliminating any abnormalities or noise using IVR(insignificant value removal) mechanism. Natural language processing then plays critical part on pre-processed data by extracting meaningful words and eliminating stop words. Next phase i.e text mining forms clusters. Earlier work kept the hold rate at 0.2 but in proposed work hold out rate is changed to 0.3 and classification accuracy is greatly improved. Result in terms of sensitivity , specificity and F-score shows improvement by 2%. Method Used Entire simulation is divided into primarily two phases. 1. Preprocessing: At this stage resizing operation is performed since input layer of the network requires predefined size. 20% hold out rate is maintained for testing data. 2. Text parsing and Classification: Multi class text parsing is used to extract critical and non critical segments from within the training dataset. After extracting the features, classification is performed on the test data. Parameters Parameters used for optimization includes classification accuracy. Simulation and Result Simulation is conducted in MATLAB using image processing and neural network toolbox. The proposed mechanism shows improvement in terms of classification accuracy by the margin of 3% which is significant, thereby enhancing the recognition rate.

Diabetes, Deep learning, Accuracy

"Drug discovery with high accuracy using hybrid text mining through IVR and natural language processing", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.6, Issue 6, page no.109-120, June 2019, Available :http://www.jetir.org/papers/JETIR1907M17.pdf

"Drug discovery with high accuracy using hybrid text mining through IVR and natural language processing", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.6, Issue 6, page no. pp109-120, June 2019, Available at : http://www.jetir.org/papers/JETIR1907M17.pdf