An Efficient Decision-Making Disease Diagnosing System Using Healthcare Datamining Techniques

Abstract

Healthcare uses Data Mining techniques for knowledge discovery and identifying successful prescription patterns for diseases and prediction using computer aided diagnosis or expert learning. Integrating Data Mining with forecasting can provide dependable and high quality forecasts. Prediction of diseases using data mining techniques is a motivating task for increasing diagnostic accuracy. Hence the objective of this research is in using data mining as they help decrease cost and time. Knowledge discovery from medical data is a complicated task, mainly due to irrelevant and unwanted data. Using more than one data mining technique for predicting diseases can also result in better accuracy. Hence the main objective of this research work is to predict Healthcare diseases from patient’s records and suggesting a non-invasive data mining model. Moreover, Features provide state-of-the-art performance for recognition of abnormalities. While the accuracy of action recognition has been continuously improved over the recent years, the extraction of lesser number of features and subsequent identifications based on these extractions have been preventing methods from scaling up to real-life issues. This problem is addressed in this research work by the development of highly efficient features using feature information in disease recognitions. Moreover the speed of feature extraction and feature selection can help disease classification perform better at the cost of a negligible reduction in recognition accuracy. The main goal of this work is efficient disease recognition while exploring the speed-memory trade-off in feature extraction and selection.