Please use this identifier to cite or link to this item:
https://repositori.uma.ac.id/handle/123456789/20208
Full metadata record
DC Field | Value | Language |
---|---|---|
dc.contributor.advisor | Khairina, Nurul | - |
dc.contributor.advisor | Noviandri, Dian | - |
dc.contributor.author | Pasaribu, Fordinand Halomoan | - |
dc.date.accessioned | 2023-07-06T03:04:35Z | - |
dc.date.available | 2023-07-06T03:04:35Z | - |
dc.date.issued | 2023-01-19 | - |
dc.identifier.uri | https://repositori.uma.ac.id/handle/123456789/20208 | - |
dc.description | 93 Halaman | en_US |
dc.description.abstract | Organisasi Kesehatan Dunia (WHO) mengumumkan adanya klaster pneumonia di Wuhan, China pada akhir tahun 2019 yang kemudian diberi nama COVID-19 (Corona Virus Disease 2019). Bulan Maret 2020 WHO menyatakan COVID-19 menjadi pandemi dunia dikarenakan penyebarannya dengan begitu cepat serta menginfeksi orang di seluruh dunia. Kemudian pada bulan November 2020 vaksin COVID-19 dengan tingkat efikasi di atas 90% akhirnya ditemukan dan siap untuk digunakan. Munculnya vaksin COVID-19 telah mengakibatkan pro serta kontra pada masyarakat. Ada yang mendukung vaksin, ada yang mewaspadai vaksin, bahkan meskipun pemerintah memberikan vaksin secara gratis, tetapi masih ada juga yang menolaknya, dan informasi tentang vaksin menyebar luas pada media sosial terutama twitter. Pengadaan vaksin corona menyebabkan timbulnya opini-opini yang beragam di masyarakat. Vaksin COVID-19 menjadi trending topik di media sosial twitter. Opini di twitter kemudian akan menjadi data untuk dilakukan analisis sentimen yang merupakan suatu proses bertujuan untuk mengetahui apakah polaritas suatu data berupa teks akan mengarah ke positif, negatif, atau netral. Metode yang digunakan pada penelitian ini yaitu pengumpulan data, text preprocessing, TF-IDF, algoritma multilayer perceptron, serta pengujian dengan confusion matrix. Dari jumlah data 228208 data opini positif, negatif, dan netral pada pengguna tweet mengenai vaksin COVID-19 dengan perbandingan training 90% dan testing 10%, model akan lebih banyak belajar menggunakan data training dengan jumlah yang besar. Hasi performa dari penelitian ini mendapatkan performa tertinggi pada akurasi 81.2%, presisi 83.8%, dan recall 71.2%. Hasil wordcloud pada opini positif terdapat 3 topik yaitu mengenai ketersediaan, berbayar, dan dosis. Opini negatif pengguna tweet terhadap vaksin COVID-19 mendapatkan 2 pokok permasalahan seperti efek samping vaksin dan kematian. Opini netral terdapat 3 topik seperti dosis, ketersediaan, umur, dan tanggal kadaluwarsa. The World Health Organization (WHO) announced a pneumonia cluster in Wuhan, China at the end of 2019 which was later named COVID-19 (Corona Virus Disease 2019). In March 2020 WHO declared COVID-19 to be a world pandemic because it spread so quickly and infected people all over the world. Then in November 2020, a COVID-19 vaccine with an efficacy level above 90% was finally found and ready to be used. The emergence of the COVID-19 vaccine has raised pros and cons in society. Some support vaccines and some are wary of vaccines, even though the government provides vaccines for free, there are still those who reject them, and information about vaccines is spreading widely on social media, especially Twitter. Procurement of the corona vaccine has led to the emergence of various opinions in society. The COVID-19 vaccine has become a trending topic on Twitter and social media. Opinions on Twitter will then become data for sentiment analysis, which is a process that aims to find out whether the polarity of textual data is positive, negative, or neutral. The methods used in this study are data collection, text preprocessing, TF-IDF, multilayer perceptron algorithms, and testing with a fusion matrix. From the total data of 228208 positive, negative, and neutral opinion data on tweet users about the COVID-19 vaccine with a training ratio of 90% and testing 10%, the model will learn more using large amounts of training data. The performance results from this study obtained the highest performance at 81.2% accuracy, 83.8% precision, and 71.2% recall. Wordcloud results in positive opinions there are 3 topics, namely regarding availability, payment, and dosage. Negative opinions of tweet users about the COVID-19 vaccine get 2 main issues such as vaccine side effects and death. Neutral opinion on 3 topics such as dosage, availability, age, and expiration date. | en_US |
dc.language.iso | id | en_US |
dc.publisher | Universitas Medan Area | en_US |
dc.relation.ispartofseries | NPM;188160061 | - |
dc.subject | analisis sentimen | en_US |
dc.subject | en_US | |
dc.subject | vaksin covid-19 | en_US |
dc.subject | algoritma multilayer perceptron | en_US |
dc.subject | TF-IDF | en_US |
dc.subject | sentiment analysis | en_US |
dc.subject | en_US | |
dc.subject | covid-19 vaccine | en_US |
dc.subject | multilayer perceptron algorithm | en_US |
dc.subject | TF-IDF | en_US |
dc.title | Penerapan Algoritma Multilayer Perceptron pada Sentimen Pengguna Tweet terhadap Vaksin Covid-19 | en_US |
dc.title.alternative | Application of the Perceptron Multilayer Algorithm to Tweet User Sentiments towards the Covid-19 Vaccine | en_US |
dc.type | Skripsi Sarjana | en_US |
Appears in Collections: | SP - Informatic Engineering |
Files in This Item:
File | Description | Size | Format | |
---|---|---|---|---|
188160061 - Fordinand Halomoan Pasaribu - Fulltext.pdf | Cover, Abstract, Chapter I, II, III, V, Bibliography | 1.57 MB | Adobe PDF | View/Open |
188160061 - Fordinand Halomoan Pasaribu - Chapter IV.pdf Restricted Access | Chapter IV | 1.51 MB | Adobe PDF | View/Open Request a copy |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.