Please use this identifier to cite or link to this item:
https://repositori.uma.ac.id/handle/123456789/24108
Full metadata record
DC Field | Value | Language |
---|---|---|
dc.contributor.advisor | Lubis, Andre Hasudungan | - |
dc.contributor.author | Putra, Pagas Sukarno | - |
dc.date.accessioned | 2024-05-29T09:45:27Z | - |
dc.date.available | 2024-05-29T09:45:27Z | - |
dc.date.issued | 2024-03-28 | - |
dc.identifier.uri | https://repositori.uma.ac.id/handle/123456789/24108 | - |
dc.description | 79 Halaman | en_US |
dc.description.abstract | ChatGPT merupakan salah satu kecerdasan buatan yang saat ini tengah banyak digunakan oleh pengguna. Aplikasi chatbot ini, dirilis pada November 2022 oleh sebuah laboratorium riset kecerdasan buatan yang bernama OpenAI. Setelah peluncurannya, ChatGPT mendapatkan beragam komentar dari para pengguna internet di berbagai platform dengan isi komentar yang positif, netral, ataupun negatif, seperti pada situs Reddit. Banyaknya pengguna Reddit mengakibatkan sulitnya untuk menganalisis apakah komentar yang diutarakan bersentimen positif, netral atau negatif. Oleh karena itu, dibutuhkan suatu teknik serta metode yang sesuai untuk permasalahan tersebut. Penelitian ini menggunakan teknik analisis sentimen dengan metode Lexicon Based dan kamus kata VADER untuk mencari polaritas kata yang bersentimen positif, netral, dan negatif dan menggunakan data komentar dari situs Reddit sebagai data penelitian. Dengan tahapan penelitian yang dilakukan yaitu pengumpulan data, data reduction, text preprocessing, implementasi metode Lexicon VADER, hasil analisis sentimen, validasi penelitian dan juga visualisasi data. Hasil dari penelitian ini yaitu data komentar yang memiliki jenis sentimen positif berjumlah 14.250 komentar dengan presentase sebanyak 53,2%, sentimen netral berjumlah 6.767 komentar dengan presentase sebanyak 24.8% dan sentimen negatif berjumlah 6.009 dengan presentase sebanyak 22.0% dari total keseluruhan jumlah data komentar sebanyak 27.296 data. Hasil validasi clustering menggunakan Silhouette Index mendapatkan nilai 0.532 dengan nilai mendekati angka 1 maka kualitas klasterisasi sudah baik. Penelitian ini menunjukkan bahwa komentar pengguna Reddit terhadap ChatGPT memiliki sentimen yang positif. ChatGPT is an artificial intelligence that is currently being widely used by users. This chatbot application was released in November 2022 by an artificial intelligence research laboratory called OpenAI. After its launch, ChatGPT received various comments from internet users on various platforms with positive, neutral or negative comments, such as on the Reddit site. The large number of Reddit users makes it difficult to analyze whether the comments expressed are positive, neutral or negative. Therefore, we need a technique and method that is suitable for this problem. This research uses sentiment analysis techniques with the Lexicon Based method and the VADER word dictionary to look for the polarity of words with positive, neutral and negative sentiments and uses comment data from the Reddit site as research data. The research stages carried out were data collection, data reduction, text preprocessing, implementation of the Lexicon VADER method, sentiment analysis results, research validation and also data visualization. The results of this research are that the comment data which has a positive sentiment type is 14,250 comments with a percentage of 53.2%, neutral sentiment is 6,767 comments with a percentage of 24.8% and negative sentiment is 6,009 with a percentage of 22.0% of the total number of comment data. 27,296 records. The results of clustering validation using the Silhouette Index get a value of 0.532 with a value close to 1, meaning the quality of the clustering is good. This research shows that Reddit user comments on ChatGPT have positive sentiment. | en_US |
dc.language.iso | id | en_US |
dc.publisher | Universitas Medan Area | en_US |
dc.relation.ispartofseries | NPM;198160044 | - |
dc.subject | Analisis Sentimen | en_US |
dc.subject | VADER | en_US |
dc.subject | Lexicon Based | en_US |
dc.subject | en_US | |
dc.subject | ChatGPT | en_US |
dc.title | Analisis Sentimen tentang Chatgpt pada Kolom Komentar Situs Reddit dengan Metode Lexicon Based | en_US |
dc.title.alternative | Sentiment Analysis about Chatgpt in the Comment Column on the Reddit Site using the Lexicon Based Method | en_US |
dc.type | Thesis | en_US |
Appears in Collections: | SP - Informatic Engineering |
Files in This Item:
File | Description | Size | Format | |
---|---|---|---|---|
198160044 - Pagas Sukarno Putra - Fulltext.pdf | Cover, Abstract, Chapter I, II, III, V, Bibliography | 1.77 MB | Adobe PDF | View/Open |
198160044 - Pagas Sukarno Putra - Chapter IV.pdf Restricted Access | Chapter IV | 1.04 MB | Adobe PDF | View/Open Request a copy |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.