ANALISIS SENTIMEN OPINI MASYARAKAT TERHADAP WISATA DI BATURRADEN MENGGUNAKAN KLASIFIKASI NAÏVE BAYES CLASSIFIER (NBC)

RAMADHAN, ANDIKA PUTRA (2023) ANALISIS SENTIMEN OPINI MASYARAKAT TERHADAP WISATA DI BATURRADEN MENGGUNAKAN KLASIFIKASI NAÏVE BAYES CLASSIFIER (NBC). S1 thesis, Universitas Muhammadiyah Purwokerto.

Preview	Text ANDIKA PUTRA RAMADHAN_COVER.pdf Download (45MB) | Preview
Preview	Text ANDIKA PUTRA RAMADHAN_BAB I.pdf Download (1MB) | Preview
Preview	Text ANDIKA PUTRA RAMADHAN_BAB II.pdf Download (1MB) | Preview
	Text ANDIKA PUTRA RAMADHAN_BAB III.pdf Restricted to Registered users only Download (1MB)
	Text ANDIKA PUTRA RAMADHAN_BAB IV.pdf Restricted to Registered users only Download (1MB)
	Text ANDIKA PUTRA RAMADHAN_BAB V.pdf Restricted to Registered users only Download (1MB)
Preview	Text ANDIKA PUTRA RAMADHAN_DAFTAR PUSTAKA.pdf Download (1MB) | Preview
	Text ANDIKA PUTRA RAMADHAN_LAMPIRAN.pdf Restricted to Registered users only Download (7MB)

Abstract

Social media is currently growing rapidly, one of which is on Google Maps. Here everyone can access easily because Google Maps can help us guide the way to where we are going. The topic I discuss here is regarding the review given by Google Maps to vent words that lead to a place on Google Maps and the topic discussed coincides with the Baturraden tourist spot. This has led to the emergence of pros and cons phenomena among the public regarding giving reviews on Google Maps. In this study, data was obtained by scraping from reviews on Google Maps using one of the tools available in Chrome, namely the Instant Data Scraper/Data Scraper. Data from scraping results obtained as many as 11,822 data. Furthermore, the data is preprocessed with the stages of cleaning, case folding, tokenizing, normalization, filtering, and stemming. After data preprocessing, the remaining 7527 is used to process data labeling. The data labeling process was carried out using the lexicon senticnet 7 method and obtained 4,572 positive numbers, 2,072 negative numbers, and 883 neutral numbers. Then the naïve Bayes classifier (NBC) classification algorithm was applied with a comparison of training and test data 8:2. The results of the tests conducted in this review obtained an accuracy of 74.00%, a precision of 73.15%, a recall of 91.88%, and a fi-score of 81.45%.

Actions (login required)

View Item