Claim Missing Document
Check
Articles

Found 2 Documents
Search

Klasifikasi Berita detik.com Terkait Teknologi Informasi Menggunakan TF-IDF dan Naive Bayes Nur Bainatun Nisa; Rivaldi Prima Nanda; Zahra Humaira Kudadiri; Bagus Ageng Alfahri; Mhd Furqan
Jurnal Nasional Komputasi dan Teknologi Informasi (JNKTI) Vol 8, No 3 (2025): Juni 2025
Publisher : Program Studi Teknik Komputer, Fakultas Teknik. Universitas Serambi Mekkah

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.32672/jnkti.v8i3.9171

Abstract

Abstrak – Penelitian ini membahas tentang klasifikasi berita Detik.com terkait teknologi informasi dengan menerapkan metode Term Frequency-Inverse Document Frequency (TF-IDF) sebagai ekstraksi fitur dan algoritma Naive Bayes sebagai model klasifikasi. Tujuan dari penelitian ini adalah untuk mengelompokkan berita-berita yang dimuat pada situs Detik.com ke dalam beberapa kategori utama di bidang teknologi informasi, seperti kecerdasan buatan, keamanan siber, gadget, dan aplikasi. Proses penelitian diawali dengan pengumpulan 1.050 data berita dari Detik.com menggunakan search query ‘teknologi informasi’ pada rentang Maret hingga April 2025. Data kemudian diproses melalui tahapan text preprocessing, meliputi case folding, tokenizing, stopword removal, dan stemming. Selanjutnya, fitur teks diubah menjadi representasi numerik menggunakan TF-IDF, lalu dilakukan pelatihan model klasifikasi dengan algoritma Naive Bayes. Evaluasi kinerja model dilakukan menggunakan metrik akurasi, precision, recall, dan F1-score. Hasil penelitian menunjukkan bahwa kombinasi TF-IDF dan Naive Bayes efektif dalam mengklasifikasikan berita teknologi informasi, dengan akurasi model mencapai 85%. Temuan ini menunjukkan bahwa pendekatan klasifikasi berbasis machine learning dapat membantu pengelompokan dan identifikasi topik utama secara otomatis dalam berita teknologi informasi di Detik.com.Kata Kunci: TF-IDF; Naive Bayes; Klasifikasi; Detik.com; Teknologi Informasi.Abstract – This study discusses the classification of Detik.com news related to information technology by applying the Term Frequency-Inverse Document Frequency (TF-IDF) method as a feature extraction and the Naive Bayes algorithm as a classification model. The purpose of this study is to group news published on the Detik.com site into several main categories in the field of information technology, such as artificial intelligence, cybersecurity, gadgets, and applications. The research process began with the collection of 1,050 news data from Detik.com using the search query 'information technology' in the range of March to April 2025. The data was then processed through the text preprocessing stage, including case folding, tokenizing, stopword removal, and stemming. Furthermore, text features were converted into numeric representations using TF-IDF, then training a classification model with the Naive Bayes algorithm. Model performance evaluation was carried out using accuracy, precision, recall, and F1-score metrics. The results showed that the combination of TF-IDF and Naive Bayes was effective in classifying information technology news, with a model accuracy reaching 85%. This finding suggests that a machine learning-based classification approach can help automatically cluster and identify key topics in information technology news on Detik.com.Keywords: TF-IDF; Naive Bayes; Classification; Detik.com; Information Technology.
A Discrete-Event and Monte Carlo-Based Simulation Model for Multi-Server Call Center Queueing Systems Nur Bainatun Nisa; Dafa Ikhwanu Shafa; Muhammad Yusuf Azmi; Armayanti Akhiriyah Parinduri
JITCoS : Journal of Information Technology and Computer System Vol. 1 No. 2 (2025): Journal of Information Technology and Computer System
Publisher : CV. Multimedia Teknologi Kreatif

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.65230/jitcos.v1i2.35

Abstract

This study presents the implementation and performance evaluation of a multi-server queueing system model for call center operations using discrete-event simulation combined with Monte Carlo analysis. The objective is to analyze system performance under varying numbers of service agents to identify the optimal configuration that balances service efficiency and customer satisfaction. The model assumes that customer arrivals follow a Poisson distribution, while service times are exponentially distributed to represent realistic call handling behavior. Simulation experiments were conducted over eight-hour operational periods with server counts ranging from one to eight, each replicated 500 times for statistical robustness. Performance indicators such as average waiting time, server utilization, and Service Level Agreement (SLA) compliance were analyzed to measure system efficiency. Results show that increasing the number of servers significantly reduces average waiting time and enhances service level compliance. Configurations with five or more servers achieved average waiting times close to zero and over 99% compliance with the SLA, while maintaining moderate server utilization levels between 70% and 80%. These findings demonstrate that integrating discrete-event simulation with Monte Carlo methods provides an effective and reliable framework for evaluating service system performance, optimizing resource allocation, and supporting decision-making in call center management.