International Journal of Scientific Engineering and Technology (ISSN : 2277-1581) Volume No.3 Issue No.8, pp : 1054-1059 1 Aug 2014 IJSET@2014 Page 1054 Expert System Design to Predict Heart and Diabetes Diseases ShravanKumar Uppin, Anusuya M A Department of Computer science and Engineering,Sjce,Mysore, India shravan4uppin@gmail.com , anusuya_ma@yahoo.co.in Abstract— The Objective of this paper is to design an expert system that predicts the heart disease and diabetes disease with reduced number of attribute using data mining technique. Classification of knowledge objects is a knowledge mining and knowledge management process used in grouping similar knowledge objects together. There are plenty of classification algorithms available in literature but decision tree is the most often used because of its ease of implementation and simpler to understand, when compared to other classification algorithms. There are many classifiers but we have used C4.5 for more accuracy and less run time. The decision tree algorithm has been applied on the knowledge of heart and diabetes disease to foretell whether diseases present or not. The Simulation result obtained from the model enables us to establish significant patterns and relationships between the medical factors and clinical factors. Keywords-Data Mining, Artificial Intelligence, Decision Tree, Heart Disease, Diabetes ,Classification,c4.5 Algorithm. 1. Introduction Expert or knowledge-based systems are the commonest blazon of Artificial Intelligence systems in accepted analytic use. They accommodate medical knowledge, usually about an actual accurately authentic task, and are able to acumen with abstracts from alone patients to appear up with articular conclusions. Although there are abounding variations, the knowledge within a specialist method is usually represented in the type of a set of rules. Machine learning frameworks might be utilized to create the information bases utilized by master frameworks. Given a set of clinical cases, a machine learning framework can deliver an efficient depiction of those clinical gimmicks that interestingly describe the clinical conditions. This information could be communicated as basic controls, or regularly as a choice Tree. Data mining techniques plays a significant function in seeing forms and extracting knowledge from a large mass of data [1]. It is very helpful to provide better patient care and effective diagnostic capabilities [2- 4]. Various data mining techniques are employed in the diagnosis of heart disease such as: Genetic algorithm, classification via clustering, direct kernel self-organizing map, naïve Bayes, decision tree, neural network, core density, automatically defined groups, bagging algorithm, and support vector machine showing different levels of accuracies [5- 6].Among these classification algorithms decision tree algorithms is the most commonly used because it is easy to understand and cheap to implement. It provides a modeling technique that is easy for humans to comprehend and simplifies the classification Process [7-9]. Choice tree Algorithms are most generally utilized calculation as a result of its simplicity of usage and simpler to comprehend contrasted with other characterization calculations. The conclusion of the choice tree anticipated the amount of patients who have sickness disease or not. Decision tree classifiers obtain similar and sometimes better accuracy when compared with other classification methods. Decision tree algorithm can be implemented in a serial or parallel fashion based on the volume of data, memory space available on the computer resource and scalability of the algorithm. The C4.5 decision tree algorithms are applied on the dataset of heart and diabetes to predict the presence or absence of disease. 2. Literature Survey Agreeing to a recent survey by the Registrar General of India (RGI) and the Indian Council of Medical Research (ICMR), nearly 25 percent of deaths in the age group of 25- 69 years occur because of heart diseases. In 2008, five out of the top ten reasons for mortality worldwide, other than injuries, were non-transmissible diseases; this will lead up to seven out of ten by the year 2030. By then, about 76% of the deaths in the world will be due to non-communicable diseases (NCDs) [10]. Cardiovascular diseases (CVDs), also on the rise, comprise a major portion of non-communicable diseases. In 2010, of all projected worldwide deaths, 23 million are expected to be because of cardiovascular diseases. In fact, CVDs would be the single largest cause of death in the world accounting for more than a third of all deaths [11]. Cardiovascular disease includes coronary heart disease (CHD), cerebrovascular disease (stroke), Hypertensive heart disease, congenital heart disease, peripheral artery disease, rheumatic heart disease, inflammatory heart disease. The major causes of cardiovascular disease are tobacco use, physical inactivity, an unhealthy diet and harmful use of alcohol [12]. Several researchers are using statistical and data mining tools to help health care professionals in the diagnosis of heart disease [13]. Diabetes [14] is an illness that threatens the life of people of all nations. It occurs when the blood sugar in the body is increased above a definite level. It is an illness in which either pancreas in the body is not producing sufficient insulin or cells in the body are not using insulin properly. There's