ANALISA SENTIMEN TWITTER TERHADAP PENGGUNAAN ALAT DETEKSI COVID-19 GENOSE MENGGUNAKAN METODE KNN

AKBAR, MUHAMMAD RIZQI (2021) ANALISA SENTIMEN TWITTER TERHADAP PENGGUNAAN ALAT DETEKSI COVID-19 GENOSE MENGGUNAKAN METODE KNN. Undergraduate thesis, Universitas Stikubank.

	PDF (HLM JUDUL) Download (965kB)
	PDF (ABSTRAK) Download (419kB)
	PDF (BAB I) Download (83kB)
	PDF (BAB II) Restricted to Repository staff only Download (95kB)
	PDF (BAB III) Restricted to Repository staff only Download (229kB)
	PDF (BAB IV) Restricted to Repository staff only Download (643kB)
	PDF (BAB V) Restricted to Repository staff only Download (85kB)
	PDF (DAFTAR PUSTAKA) Download (31kB)
	PDF (lAMPIRAN) Restricted to Repository staff only Download (713kB)

Abstract

GeNose C19 merupakan salah satu alat untuk mendeteksi virus covid-19 yang dikembangkan oleh Universitas Gajah mada (UGM). GeNose bekerja dengan cara mendeteksi Volatile Organic Compund (VOC) yang terbentuk karena adanya infeksi virus covid-19 yang keluar bersamaan dengan nafas yang dimasukkan ke dalam kantong khusus. Saat GeNose mulai digunakan sebagai syarat untuk melakukakan perjalanan dengan transportasi umum terutama kereta api dan pesawat terbang, masyarakat memberikan komentar terhadap penggunaan alat tersebut melalui media sosial terutama twitter. Komentar tersebut memiliki peranan penting karena dapat digunakan untuk melakukan analisis sentimen dalam memprediksi penilaian masyarakat tentang alat tersebut apakah bernilai positif atau negatif. Implementasi analisis sentiment twitter terhadap penggunaan alat deteksi covid -19 GeNose menggunakan metode K-Nearest Neighbor dapat membantu mengklasifikasikan penilaian masyarakat tentang alat tersebut menjadi sentimen negatif atau positif. Untuk menunjang proses analisis, sebelum data di analisis, ada beberapa tahap yang dilakukan yaitu preprocessing yang didalamnya terdapat proses cleaning, case folding, stemming, stopword removal dan tokenisasi. Proses setelahnya yaitu pembobotan TF-IDF dan yang terakhir klasifikasi metode K-Nearest Neighbor.Setelah proses klasifikasi dan dilakukan pengujian, sistem dapat mengklasifikasikan komentar akurasi terbaik sebesar 87% dengan penggunaan nilai k 1,3, dan 7. GeNose C19 is one of the tools to detect the covid-19 virus developed by Gajah Mada University (UGM). GeNose works by detecting Volatile Organic Compounds (VOCs) that are formed due to the Covid-19 virus infection that comes out together with breath that is put in a special bag. When GeNose began to be used as a condition for traveling by public transportation, especially trains and airplanes, people commented on the use of these tools through social media, especially Twitter. These comments have an important role because they can be used to conduct sentiment analysis in predicting the public's assessment of whether the tool is positive or negative. The implementation of Twitter sentiment analysis on the use of the GeNose COVID-19 detection tool using the K-Nearest Neighbor method can help to classify people's assessments of the tool into negative or positive sentiments. To support the analysis process, before it is processed into several stages, namely preprocessing which includes the cleaning process, case folding, stemming, stopword removal and tokenization. The next process is TF-IDF weighting and the last is the classification of the K-Nearest Neighbor method. After the classification process and testing, the system can classify the comments with the best accuracy of 87% using k values of 1,3, and 7.

Item Type:	Thesis (Undergraduate)
Additional Information:	SKR.I.05.01.2012 NIM 17.01.53.0004
Uncontrolled Keywords:	analisis sentimen, K-Nearest Neighbor, TF-IDF, Data mining sentiment analysis, K-Nearest Neighbor, TF-IDF, Data mining
Subjects:	Q Science > QA Mathematics > QA76 Computer software
Faculty / Institution:	Fakultas Teknologi Informasi > Program Studi Teknik Informatika
Depositing User:	Teteh Hayati
Date Deposited:	09 Nov 2021 05:59
Last Modified:	09 Nov 2021 07:07
URI:	https://eprints.unisbank.ac.id/id/eprint/8010

Actions (login required)

View Item