Predicting Thyroid Disease using Linear Discriminant Analysis (LDA) Data Mining Technique

Thyroid disease is very common disease in human. Nowadays most of the women suffering from thyroid disease than male. There are two types in thyroid disease like hypothyroid and hyperthyroid disease. These diseases giving many side effects such as weight gain, weight loss, stress and so on to our human body .If this disease is detected in earlier stage, then physician can give proper treatment to the patients .Data Mining is playing important role in predicting many diseases. Classification is one the most significant Technique in Data Mining. It is a supervised learning. It is used to classify predefined data sets. Health care data are having exponential growth in volume and complexity. Data mining Technique is mainly used in healthcare organizations for decision making, diagnosing disease and giving better treatment to the patients. In this paper hypothyroid disease is to be predicted using data mining. The dataset used for the study on hypothyroid is taken from UCI repository. Classification of this thyroid disease is a considerable task. An experimental study is carried out using Linear Discriminant Analysis (LDA) to achieve better accuracy. There are many data mining classification Algorithms such as CART, REP Tree, and J48 and so on. The LDA Algorithm gives accuracy is 99.62% with cross validation k=6.