Analisis Sentimen Terhadap Program Kampus Merdeka Menggunakan Algoritma Naive Bayes Classifier Di Twitter
The independent campus program or independent learning campus independence (MBKM) is a new policy launched in January 2019 by the ministry of education, culture, research, and technology of the republic of Indonesia. This policy is the government's strategy in improving the quality of students...
Gespeichert in:
Veröffentlicht in: | Jurnal Tekno Kompak (Bandarlampung. Online) 2023-02, Vol.17 (1), p.25 |
---|---|
Hauptverfasser: | , |
Format: | Artikel |
Sprache: | eng |
Online-Zugang: | Volltext |
Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
Zusammenfassung: | The independent campus program or independent learning campus independence (MBKM) is a new policy launched in January 2019 by the ministry of education, culture, research, and technology of the republic of Indonesia. This policy is the government's strategy in improving the quality of students to respond to the needs of the times along with significant changes in technology, the world of work, social and culture. As a new breakthrough in the world of higher education, this program is very much discussed on social media. Currently, many people use social media Twitter to provide feedback or opinions on a government policy or trend that is developing. This independent campus program has reaped the pros and cons of the community, especially on Twitter social media since its inception. This study aims to analyze the sentiment of public opinion on the independent campus program on twitter to determine the level of accuracy in the method and the proportion of sentiment as an evaluation of the algorithm, performance and program of the independent campus itself. Data collection using the vicinitas.io website was carried out in real time with #kampus merdeka and #mbkm from tweets and retweets of twitter users during November 2021 to March 2022. Analysis of 501 tweet data was carried out by classifying text in negative and positive forms using the naive bayes classifier algorithm. . The implementation of the classification in the Naive Bayes algorithm is carried out in several stages, namely text preprocessing, TF-IDF calculations, classification calculations, and K-fold cross validation. K-fold is used for applications that are used to get maximum accuracy results. The program is made in the python programming language on the google colab tools provided by google. The visualization of the results displayed in this study is a word cloud with the most dominant word results appearing on positive sentiments, namely campus, merdeka, mbkm, and programs, while on negative sentiments, namely campus, money, pocket, and conversion. Based on the results of the research, the classification that can be done by the system gets 272 positive sentiment classification results and 229 negative sentiment opinions with an average accuracy of 60%, precision 64%, recall 58% and f1-score 58%. |
---|---|
ISSN: | 1412-9663 2656-3525 |
DOI: | 10.33365/jtk.v17i1.2061 |