Application of Artificial Neural Network, K-Nearest Neighbor and Naive Bayes Algorithms for Classification of Obesity Risk Cardiovascular Disease
Keywords:Artificial Neural Network, Classification, K-Nearest Neighbor, Naive Bayes, Obesity
The rate of obesity sufferers continues to increase every year. This happens due to improper lifestyle and diet, as well as various physical conditions. This research aims to analyze the level of obesity using data mining techniques with classification algorithms. This research was conducted on people from countries on the American continent between the ages of 14 and 61 years. Data is collected and information is processed using a web platform that includes surveys where anonymous users answer each question to obtain 17 attributes and 2111 records. This research uses 3 algorithms, namely the Artificial Neural Network algorithm, K-Nearest Neighbor and Naive Bayes. People who are obese are also at higher risk of experiencing health problems, such as asthma, stroke, heart disease, diabetes and cancer. The results after comparing the three algorithms, it is better to use the k-nearest neighbor algorithm compared to Artificial Neural Network and Naive Bayes because the accuracy is 95.74%. Therefore, the K-Nearest Neighbor algorithm is very suitable to use when classifying data.