ANALYSIS OF PUBLIC SENTIMENT TOWARDS TOKOPEDIA BASED ON OPINION FROM TWITTER USING NAÏVE BAYES CLASSIFIER METHOD AND SPAM DETECTION

PRAYITNO, YOEL AMAZIA (2024) ANALYSIS OF PUBLIC SENTIMENT TOWARDS TOKOPEDIA BASED ON OPINION FROM TWITTER USING NAÏVE BAYES CLASSIFIER METHOD AND SPAM DETECTION. Skripsi thesis, UNIVERSITAS KATOLIK SOEGIJAPRANATA.

Preview	Text 20.K1.0022-YOEL AMAZIA PRAYITNO_COVER_1.pdf Download (196kB) | Preview
	Text 20.K1.0022-YOEL AMAZIA PRAYITNO_BAB I_1.pdf Restricted to Registered users only Download (171kB)
	Text 20.K1.0022-YOEL AMAZIA PRAYITNO_BAB II_1.pdf Restricted to Registered users only Download (182kB)
	Text 20.K1.0022-YOEL AMAZIA PRAYITNO_BAB III_1.pdf Restricted to Registered users only Download (246kB)
	Text 20.K1.0022-YOEL AMAZIA PRAYITNO_BAB IV_1.pdf Restricted to Registered users only Download (441kB)
	Text 20.K1.0022-YOEL AMAZIA PRAYITNO_BAB V_1.pdf Restricted to Registered users only Download (165kB)
Preview	Text 20.K1.0022-YOEL AMAZIA PRAYITNO_DAPUS_1.pdf Download (260kB) | Preview
	Text 20.K1.0022-YOEL AMAZIA PRAYITNO_LAMPIRAN_1.pdf Restricted to Registered users only Download (276kB)

Abstract

Sentiment analysis is a text analysis technique that aims to identify and evaluate the sentiments, opinions, or emotions contained in the text being analyzed. Specifically, sentiment analysis can be used to analyze the sentiment associated with a topic, product, brand, or service in a text. Of the several journals that have been reviewed and identified, a particular journal revealed relevant limitations that could be incorporated into the research. Specifically, the absence of a spam detection system in the existing research hindered the completeness of the research and potentially affected the final results. Therefore, the inclusion of a spam detection system is crucial to improve the accuracy and reliability of the findings. In this research, the author will use the naïve bayes classifier and SVM methods for comparison. This research seeks to identify existing sentiment and understand the factors underlying the sentiment. The results to be achieved are the effectiveness of the naive bayes algorithm compared to the svm algorithm with the accuracy value as the assessment parameter and also the effectiveness of spam detection. Naive Bayes outperforms SVM after 5 experiments with different split data with an average accuracy value of 95.27% Naive Bayes and 93.56% SVM. And the result of using spam detection is that positive tweets decreased by 1.24%, neutral and negative tweets increased to 0.73% and 0.51% after using spam detection

Item Type:	Thesis (Skripsi)
Subjects:	000 Computer Science, Information and General Works > 004 Data processing & computer science
Divisions:	Faculty of Computer Science > Department of Informatics Engineering
Depositing User:	Mr Yosua Norman Rumondor
Date Deposited:	23 Apr 2024 03:00
Last Modified:	23 Apr 2024 03:00
URI:	http://repository.unika.ac.id/id/eprint/35218

Actions (login required)

View Item