ارسل ملاحظاتك

ارسل ملاحظاتك لنا







الرئيسية > Sentiment Analysis of Twitter...>الوصف

Sentiment Analysis of Twitter Users on the Topic of Nusantara Vaccines Using the Naïve Bayes Algorithm

المصدر: مجلة دراسات المعلومات والتكنولوجيا
الناشر: جمعية المكتبات المتخصصة فرع الخليج العربي ودار جامعة حمد بن خليفة للنشر
المؤلف الرئيسي: Syahputra, Ismail (Author)
مؤلفين آخرين: Samihardjo, Rosalin (Co-Author)
المجلد/العدد: مج5, ع1
محكمة: نعم
الدولة: قطر
التاريخ الميلادي: 2022
الشهر: مارس
الصفحات: 1 - 12
ISSN: 2616-4930
رقم MD: 1266597
نوع المحتوى: بحوث ومقالات
اللغة: الإنجليزية
قواعد المعلومات: EduSearch, HumanIndex
مواضيع:
الأوبئة العالمية | فيروس كورونا (كوفيد-19) | اللقاحات | لقاح نوسانتارا | تحليل المشاعر | خوارزمية نايف بايز | وسائل التواصل الاجتماعي | تويتر
كلمات المؤلف المفتاحية:
Sentiment Analysis | Nusantara Vaccine | Naïve Bayes | Twitter | Social Media
رابط المحتوى:
PDF (صورة)     
صورة الغلاف QR قانون

عدد مرات التحميل

4

حفظ في:
المستخلص: Vaccine development and production is an effort to combat the Covid-19 outbreak. A vaccine that is being developed in Indonesia, and which has drawn public attention, especially on social media, is introduced as the Nusantara vaccine. From its first appearance to the temporary suspension of research by the BPOM (National Agency of Drug and Food Control), the Nusantara vaccine has raised pros and cons and has become a public conversation, especially on social media such as Twitter. The number of conversations made by social media users, especially on Twitter about the Nusantara vaccine, shows that the topic has attracted a great deal of user interests. This study aims to determine the polarization of Twitter users in Indonesia towards the Nusantara vaccine, which can be used as a reference for policy-makers. It also aims to determine the performance of the naive Bayes algorithm in classifying Indonesian texts. The research method used in analyzing sentiment was text mining. Sentiment analysis was performed using the naive Bayes algorithm. This study created a classification with two models, namely a two-class model (positive, negative) and a non-class model (positive, negative, neutral). From the processed data, it was evident that 55.51% of users expressed positive sentiment, 27.03% had negative sentiment, and the remaining 17.46% had neutral sentiment. The results of the naive Bayes classification showed that the best accuracy rate was 68.75% and 50% for the two-class and three-class classifications, respectively.
ISSN: 2616-4930

عناصر مشابهة


© 2024 المنظومة. جميع الحقوق محفوظة.