FAHLEVI, ZUL HAM (2024) PENERAPAN ALGORITMA NAIVE BAYES PADA ANALISIS SENTIMEN TWITTER TERHADAP HARGA BERAS. S1 thesis, Universitas Mercu Buana Jakarta.
|
Text (HAL COVER)
01 COVER.pdf Download (483kB) | Preview |
|
|
Text (ABSTRAK)
02 ABSTRAK.pdf Download (141kB) | Preview |
|
![[img]](https://repository.mercubuana.ac.id/style/images/fileicons/text.png) |
Text (BAB I)
03 BAB 1.pdf Restricted to Registered users only Download (158kB) |
|
|
Text (BAB II)
04 BAB 2.pdf Restricted to Registered users only Download (295kB) |
|
|
Text (BAB III)
05 BAB 3.pdf Restricted to Registered users only Download (445kB) |
|
|
Text (BAB IV)
06 BAB 4.pdf Restricted to Registered users only Download (657kB) |
|
|
Text (BAB V)
07 BAB 5.pdf Restricted to Registered users only Download (174kB) |
|
|
Text (DAFTAR PUSTAKA)
08 DAFTAR PUSTAKA.pdf Restricted to Registered users only Download (163kB) |
|
|
Text (LAMPIRAN)
09 LAMPIRAN.pdf Restricted to Registered users only Download (544kB) |
Abstract
Rice is the primary staple food for 98% of Indonesia's population and plays a crucial role in the national food structure. Fluctuations in rice prices can significantly impact the economy, especially by increasing the number of people living in poverty. Twitter has become a popular platform for expressing public opinions on various issues, including rice prices. Sentiment analysis can help classify public opinions into positive or negative using various algorithms. This research employs the Naïve Bayes algorithm for sentiment analysis of Twitter data regarding rice prices in Indonesia. The application of the Naïve Bayes algorithm to Twitter sentiment analysis on rice prices involves several stages, including data collection, data cleaning, automatic data labeling using the InSet Lexicon, word weighting using TF-IDF, splitting data into training and testing sets, and classification with the Naïve Bayes algorithm across four trials. The comparison of classification results shows that the trial with a 90:10 data split yields the best results with an Accuracy of 81.54%, Precision of 86.57%, Recall of 71.65%, and F1-score of 78.42%. From the analysis of the 10% testing data, 38.7% of the sentiments were categorized as positive, while 61.3% were negative. These results provide a clear picture of public opinion on rice prices on Twitter. Keywords: Sentiment Analysis, Naïve Bayes, Algorithm, Twitter Beras merupakan bahan pangan pokok utama bagi 98% penduduk Indonesia dan memiliki peran penting dalam struktur pangan nasional. Fluktuasi harga beras dapat berdampak signifikan pada ekonomi, terutama pada peningkatan jumlah penduduk miskin. Twitter menjadi platform yang populer untuk menyuarakan opini publik mengenai berbagai isu, termasuk harga beras. Analisis sentimen dapat membantu mengklasifikasikan opini publik menjadi positif atau negatif menggunakan berbagai algoritma. Penelitian ini menggunakan algoritma Naïve Bayes untuk analisis sentimen Twitter terhadap harga beras di Indonesia. Penerapan algoritma Naïve Bayes pada analisis sentimen Twitter terhadap harga beras melibatkan beberapa tahapan seperti seperti pengumpulan data, pembersihan data, pelabelan data otomatis menggunakan InSet Lexicon, pembobotan kata menggunakan TFIDF, pembagian data training dan data testing, serta klasifikasi algoritma Naïve Bayes dengan 4 kali percobaan. Hasil perbandingan klasifikasi menunjukkan bahwa percobaan dengan pembagian data 90:10 memberikan hasil terbaik dengan Accuracy: 81,54%, Precision: 86,57%, Recall: 71,65%, dan F1-score: 78,42%. Dari analisis data testing sebesar 10%, sebanyak 38,7% sentimen dikategorikan sebagai positif dan 61,3% sebagai negatif. Hasil ini memberikan gambaran yang jelas tentang opini publik mengenai harga beras di Twitter. Kata Kunci: Analis Sentimen, Naïve Bayes, Algoritma, Twitter
Actions (login required)
 |
View Item |
