Article Per Year (5 Year)
p-Index From 2020 - 2025
0.444
P-Index
This Author published in this journals
All Journal JURNAL MEDIA INFORMATIKA BUDIDARMA
Samsir Samsir
Universitas Al Washliyah, Rantauprapat
Computer Science & IT Control & Systems Engineering Electrical & Electronics Engineering

Published : 2 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 2 Documents
Search

 1 
Analisis Sentimen Pembelajaran Daring Pada Twitter di Masa Pandemi COVID-19 Menggunakan Metode Naïve Bayes Samsir Samsir; Ambiyar Ambiyar; Unung Verawardina; Firman Edi; Ronal Watrianthos
JURNAL MEDIA INFORMATIKA BUDIDARMA Vol 5, No 1 (2021): Januari 2021
Publisher : STMIK Budi Darma

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30865/mib.v5i1.2580

Abstract

The WHO announced that more than 52 million people tested positive for Covid-19, and 1.2 million died in the second week of November 2020. Meanwhile, Indonesia recorded 463 thousand individuals with 15,148 deaths that were confirmed positive. Strategy against pandemics by incorporating socialization. However, learning that was initially bold as a technique became controversial due to the briefness of the adaptation process. a wide continuum of social reactions has resulted in the sudden transition from face-to-face learning to bold learning on a large scale. This research focuses on public opinion on online learning during the Indonesian COVID-19 pandemic in early November 2020. The analysis was carried out on Twitter by mining document-based text that was interpreted using the Naïve Bayes algorithm. The results show that online learning has a positive sentiment of 30 percent, a negative sentiment of 69 percent, and a neutral 1 percent over the period. Due to community dissatisfaction about online learning, a significant amount of negative sentiment is created. Some tweets indicate disappointment with the words' stress 'and' lazy 'in the conversation being high-frequency words.
BERTopic Modeling of Natural Language Processing Abstracts: Thematic Structure and Trajectory Samsir Samsir; Reagan Surbakti Saragih; Selamat Subagio; Rahmad Aditiya; Ronal Watrianthos
JURNAL MEDIA INFORMATIKA BUDIDARMA Vol 7, No 3 (2023): Juli 2023
Publisher : Universitas Budi Darma

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30865/mib.v7i3.6426

Abstract

The rapid growth in the academic literature presents challenges in identifying relevant studies. This research aimed to apply unsupervised clustering techniques to 13,027 Scopus abstracts to uncover structure and themes in natural language processing (NLP) publications. Abstracts were pre-processed with tokenization, lemmatization, and vectorization. The BERTopic algorithm was used for clustering, using the MiniLM-L6-v2 embedding model and a minimum topic size of 50. Quantitative analysis revealed eight main topics, with sizes ranging from 205 to 4089 abstracts per topic. The language models topic was most prominent with 4089 abstracts. The topics were evaluated using coherence scores between 0.42 and 0.58, indicating meaningful themes. Keywords and sample documents provided interpretable topic representations. The results showcase the ability to produce coherent topics and capture connections between NLP studies. Clustering supports focused browsing and identification of relevant literature. Unlike human-curated classifications, the unsupervised data-driven approach prevents bias. Given the need to understand research trends, clustering abstracts enables efficient knowledge discovery from scientific corpora. This methodology can be applied to various datasets and fields to uncover overlooked patterns. The ability to adjust parameters allows for customized analysis. In general, unsupervised clustering provides a versatile framework for navigating, summarizing, and analyzing academic literature as volumes expand exponentially.
Co-Authors Ambiyar, Ambiyar Firman Edi Rahmad Aditiya Ronal Watrianthos Saragih, Reagan Surbakti Subagio, Selamat Unung Verawardina