Skripsi
ANALISIS SENTIMEN PELAYANAN BPJS MENGGUNAKAN ALGORITMA NAIVE BAYES DAN PARTICLE SWARM OPTIMIZATION
BPJS is a legal entity aimed at providing healthcare coverage to the public. However, the increasing number of people using BPJS services has generated both support and opposition among the community. This research aims to analyze public opinions regarding the Social Security Administration Agency (BPJS) on Twitter. The method employed in this study is the Naïve Bayes Classifier with Particle Swarm Optimization (PSO) as the feature selection method. In this research, the PSO feature selection parameters are set with 40 particles, a cognitive coefficient (c1) of 1.2, a social coefficient (c2) of 0.6, and an inertia (w) of 0.1. The evaluation is conducted by comparing the Naïve Bayes method without feature selection and the Naïve Bayes method with PSO feature selection. The results indicate that the Naïve Bayes method without feature selection successfully classifies text with an accuracy of 79.1%. However, with the implementation of PSO feature selection, the classification accuracy increases to 92.2%. This demonstrates that utilizing PSO feature selection with properly set parameters improves the accuracy by 13.1%. This research providesa better understanding of public sentiment towards BPJS services and highlights the effectiveness of Naïve Bayes classification when combined with PSO feature selection in classifying public opinions related to BPJS services. Keywords: BPJS, Feature Selection, Naïve Bayes Classifier, Particle Swarm Optimization (PSO), Sentiment analysis.
Inventory Code | Barcode | Call Number | Location | Status |
---|---|---|---|---|
2307005299 | T125198 | T1251982023 | Central Library (REFERENS) | Available but not for loan - Not for Loan |
No other version available