Analisis Sentimen Rangka Esaf Pada Sepeda Motor Honda Matic Menggunakan Metode Naïve Bayes dan Support Vector Machine

KAMI, Yohanes De Britto Dhana (2024) Analisis Sentimen Rangka Esaf Pada Sepeda Motor Honda Matic Menggunakan Metode Naïve Bayes dan Support Vector Machine. Undergraduate thesis, Universitas Katolik Widya Mandira Kupang.

	Text ABSTRAK.pdf Download (421kB)
	Text BAB I.pdf Download (492kB)
	Text BAB II.pdf Restricted to Repository staff only Download (718kB)
	Text BAB III.pdf Restricted to Repository staff only Download (614kB)
	Text BAB IV.pdf Restricted to Repository staff only Download (800kB)
	Text BAB V.pdf Download (458kB)
	Text DAFTAR PUSTAKA.pdf Download (890kB)

Abstract

The rapid development of automotive technology has become a primary focus for automotive companies. Innovation, as well as the development of design and safety features, have increasingly become central to meeting market demands. eSAF is the latest innovation in the frame of automatic motorcycles, launched by HONDA at the end of 2019. By combining elements such as safety, comfort, and efficiency, HONDA successfully created an automatic motorcycle frame called eSAF. Starting in mid-August 2023, the eSAF frame began to trigger various reactions and responses related to issues of quality that do not meet safety standards. Various reactions and responses concerning the eSAF frame issue have spread widely on social media, one of which is on X (formerly Twitter). Analyzing public sentiment regarding the eSAF frame, using the Naïve Bayes and Support Vector Machine methods, is the goal of this study. The performance results of both methods will be compared to determine which method has the best performance in analyzing sentiment regarding the eSAF frame. Out of a total of 6,041 data points, 80% or 4,832 data points were used as training data, and 20% or 1,209 data points were used as test data, which will be analyzed by the Naïve Bayes and Support Vector Machine methods. The research results show that the Naïve Bayes method analyzed 1,016 data points as negative sentiment, 192 data points as neutral sentiment, and 1 data point as positive sentiment, while the Support Vector Machine method analyzed 800 data points as negative sentiment, 390 data points as neutral sentiment, and 19 data points as positive sentiment. The research findings indicate that the Support Vector Machine method achieved an accuracy rate of 0.6360, slightly outperforming the Naïve Bayes method, which had an accuracy rate of 0.6195.

Actions (login required)

View Item