|Institution:||KTH Royal Institute of Technology|
|Keywords:||Natural Sciences; Computer and Information Science; Computer Science; Naturvetenskap; Data- och informationsvetenskap; Datavetenskap (datalogi)|
|Full text PDF:||http://urn.kb.se/resolve?urn=urn:nbn:se:kth:diva-166593|
Heart disease is the leading cause of death in the world. Being able to conduct an early screening diagnosis of heart disease at home, could potentially be a tool to reduce the amount of people who lose their lives to the disease in the future. This report aims at investigating if an early screening diagnostic aid using no attributes requiring advanced medical equipment to be measured can be created, that acquires the same level of accuracy as previous data sets and studies. A litera- ture study of medical background, patient data sets and attributes, as well as data mining was conducted. A unique home data set consisting of attributes that can be obtained from home was created and data mining experiments were run in WEKA, using classification algorithms Naive-Bayes and Decision Trees. The results are compared to the Cleveland data set in regards to accuracy. The study shows that the home data set does not deliver the same accuracy level as the Cleveland data set. The idea that similar accuracy can be obtained for the dierent sets has not been disproven and more exhaustive research is encouraged.