Perbandingan Performa Labeling Lexicon Inset dan Lexicon Vader pada Analisa Sentimen Rohingya di Aplikasi X dengan Support Vector Machine

Fathoni, Muhammad Fernanda Naufal (2024) Perbandingan Performa Labeling Lexicon Inset dan Lexicon Vader pada Analisa Sentimen Rohingya di Aplikasi X dengan Support Vector Machine. Undergraduate thesis, UPN Vteteran Jawa Timur.

	Text (cover) 20081010257.-cover.pdf Download (507kB)
	Text (bab 1) 20081010257.-bab1.pdf Download (143kB)
	Text (bab 2) 20081010257.-bab2.pdf Restricted to Repository staff only until 4 June 2028. Download (352kB)
	Text (bab 3) 20081010257.-bab3.pdf Restricted to Repository staff only until 4 June 2028. Download (495kB)
	Text (bab 4) 20081010257.-bab4.pdf Restricted to Repository staff only until 4 June 2028. Download (656kB)
	Text (bab 5) 20081010257.-bab5.pdf Download (8kB)
	Text (daftar pustaka) 20081010257.-daftarpustaka.pdf Download (141kB)

Abstract

The Rohingya influx in Indonesia has become a topic of conversation on social media. One way to see how the public responds to it is to analyze sentiment. The amount of data makes the problem of time efficiency in the labeling process, therefore the lexicon dictionary is needed for the labeling process. A lexicon dictionary is a dictionary that contains word data with word values or weights which are then calculated to determine the sentiment of a sentence. The existence of a lexicon makes the training data labeling process faster so that it simplifies the data processing process. In the current era, data is growing and circulating very rapidly so it takes a fast and efficient time. Although it is fast and makes it easier to solve problems, it is still necessary to question the accuracy produced when using the lexicon labeling. A comparison of the labeling process between the InSet lexicon and the VADER lexicon was conducted to determine the accuracy of the labeling. It was done by combining lexicon with machine learning method of support vector machine and TF-IDF weighting. The results obtained will be evaluated using confusion matrix and the accuracy of the results will be calculated. Data obtained from social media X as many as 9117 lines and when labeled with InSet lexicon produces 5241 negative sentiments, 1369 positive sentiments, and 521 neutral sentiments. Then the labeling results with the VADER lexicon produced 2749 positive sentiments, 2523 negative sentiments, and 1881 neutral sentiments. After being labeled, weighting is carried out and then processed with SVM and the confusion matrix is calculated with the results of InSet lexicon accuracy having an average of 85.8% while the VADER SVM lexicon has an average of 82.65%.

Item Type:

Thesis (Undergraduate)

Contributors: