<![CDATA[This study aims to compare the performance of two algorithms in detecting diabetes mellitus, which is a metabolic disorder caused by insufficient insulin production by the pancreas. In this research, we used two algorithms, namely Naive Bayes and K-Nearest Neighbor (KNN), to carry out analysis on the diabetes mellitus dataset used. The Naive Bayes algorithm is a statistical algorithm used to classify and predict the probability of certain classes. Meanwhile, the K-Nearest Neighbor (KNN) algorithm is used to classify new objects based on their similarity to nearby objects. This study utilized 9 variables, including number of pregnancies, glucose levels, blood pressure, skin thickness, insulin, Body Mass Index (BMI), family history of diabetes, age, and diagnosis results. The dataset used consists of 2000 data obtained from KAGGLE. The classification process is carried out by importing data into Microsoft Excel, designing the process, and then analyzing the data using Google Colab by applying the K-Nearest Neighbor and Naive Bayes algorithms. The research results show that the K-Nearest Neighbor algorithm provides a higher level of accuracy compared to the Naive Bayes algorithm.]]>
Copyrights © 2023
