Perbandingan Algoritma Naïve Bayes Classifier Dan K-Nearest Neighbor Pada Sentimen Review Aplikasi Mobile JKN

BPJS Health must provide health services for the people of Indonesia. With the availability of the Mobile JKN application, it is useful to facilitate services for participants of the National Health Insurance-Indonesian Health Card (JKN-KIS). Mobile JKN is an innovation in electronic government health insurance services, making it easier for the public to access services and information quickly in the palm of their hand. With this innovation, many pros and cons flowed from the community, various comments appeared in the Play Store review column, sentiment analysis could be used to assess and rate applications. Therefore, these sentiments can be analyzed into information that can be used as material for evaluation and consideration by BPJS Kesehatan regarding Mobile JKN. This study aims to look at the results of the accuracy comparison between the Naïve Bayes Classifier (NBC) and K-Nearest Neighbor (KNN) algorithms on the sentiment review of the Mobile JKN application on the Play Store. This study used the Naïve Bayes Classifier (NBC) and K-Nearest Neighbor (KNN) methods with data scrapping techniques to collect Play Store data for the past year, namely 2,847 data and divided into 3 classes, namely positive, neutral and negative. Distribution of data using 10 K-Fold Cross Validation so that a comparison of the accuracy level of the Naïve Bayes Classifier (NBC) is 61.15%, while the accuracy level of K-Nearest Neighbor (KNN) is 87.59%.