Prediction of Disease through Data Mining

  • Alok Nagargoje, Anand Bohara, Vijay Kakade, Mrunal Kale and Prof.P.S.Hanwate


According to the World Bank collection of development indicators in 2018, shows that out of the total population of India, the Rural population was reported as 65.97%. Due to less availability of public heath-care, lack of awareness, inadequate facilities and less knowledge regarding various diseases are among the main causes for improper health issues. With increase in various types of technologies supporting in the field of health-care, it can be possible to help to reduce the number of people dying by various diseases in rural. It can be achieved through various types of data of symptoms regarding different diseases that are generated by various hospitals, NGO’s and various government organizations. We propose to build our project with the main focus on creating an application that classifies and predicts diseases on basis of prior data-sets regarding various symptoms on diseases and the user’s knowledge about their health and symptoms. It consists of three different algorithms - Navies Bayes, Random Forest and Decision Tree to classify and predict the diseases. Due to lack of adequate health facilities, often people are not aware of their disease until it’s too late. The private health-care is too costly and located at far distances, while most often the public health-care doctors have knowledge regarding common viral’s and influenza. The starting symptoms can be quite similar and this leads to wrong prediction of disease and treatment, and this causes late treatment of disease after non-effectiveness of treatment from prior. At times when particular disease is volatile, late treatment can lead to fatal death. With the help of data mining, data-sets of symptoms and algorithms we can achieve the predicting of disease before its too late and thus improve the future health conditions of people in rural as well as urban areas.