Veri Madenciliği İle Kalp Hastalığı Teşhisi

Abstract

Developing and changing environmental conditions, the globalization of the borders and the globalization of the world, different marketing and R&D (research and development) methods reveal the importance of "information" rather than "data". The widespread and easing of the Internet makes it difficult for R&D teams to access “information”. Research on the internet using search engines often results in a different way than desired. The ability of a large retailer to identify customer trends from the invoice information and produce marketing tactics accordingly will prevent them from getting ahead of their competitors. If attention is given to the examples given, it will be seen that the process of turning “data” into “information” is emphasized. Data mining is the business of accessing and mining information among large-scale data. Or, in a sense, it is the search for the relations that can enable us to make predictions about the future from large data stacks using a computer program. Data mining is the extraction of implicit, unclear, previously unknown but potentially useful information from the available data. At various stages of the data mining process; Algorithms such as statistical methods, memory-based methods, genetic algorithms, neural networks and decision trees can be used. Heart diseases (cardiovascular diseases) are one of the most common diseases in the world today. It is estimated that cardiovascular diseases will continue to be the number one cause of death for a long time on a global scale. Cardiovascular disease deaths in developed western countries show a decreasing tendency in developing countries. The positive factor in mortality rates in the world is that they are largely preventable in terms of cardiovascular diseases. Therefore, based on the data of patients diagnosed with heart disease, the study was carried out to predict pre-cardiac disease by using text mining and algorithms. This study was conducted to show how much importance and place data mining has on the study of big data sets. From the heart data set containing hundreds of information, by using WEKA program, by applying various algorithms, the study was made to diagnose people with heart disease. There are various applications and methods for the definitive diagnosis of heart disease and detection of disease severity. In this study, the use of data mining, which could provide a cheaper and more effective approach, was studied. In this study, the results obtained by classification methods and correct classification rates were compared. In order to obtain the necessary calculations and models, classification algorithms such as ZeroR, OneR, Naive Bayes, J48 Decision Tree, Random Forest, Multiplayer Perceptrons, k-nearest neighbor (k-NN), Logistic Regression, support vector machine (SVM), have been applied in Weka packet program. As a result of the application of the best results in the determination of heart disease algorithm has been tried to be determined. There are many different studies that determine heart disease by data mining algorithms. But there is no study that implements 9 different algorithms to the data set and this paper will be the first one.