Claim Missing Document
Check
Articles

Found 34 Documents
Search

Perbandingan Algoritma Klasifikasi untuk Prediksi Sentimen Emoji dalam Teks Moh Ainol Yaqin; Mochammad Faid
JOKI: Jurnal Komputasi dan Informatika Vol 1 No 01 (2024): Juni 2024
Publisher : CV. Laskar Karya

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.65678/joki.v1i01.16

Abstract

Penelitian ini bertujuan untuk mengevaluasi dan membandingkan efektivitas berbagai algoritma pembelajaran mesin dalam mengklasifikasikan sentimen emoji. Data yang digunakan dalam penelitian ini terdiri dari emoji yang masing-masing diberikan label sentimen positif, negatif, atau netral. Kami menerapkan teknik Count Vectorization untuk mengubah data teks menjadi format numerik yang sesuai untuk pemrosesan algoritma pembelajaran mesin, Empat algoritma yang dievaluasi dalam penelitian ini adalah Multilayer Perceptron (MLP), Decision Tree Classifier, Naive Bayes Classifier, dan K-Nearest Neighbors (KNN). Setiap model dilatih menggunakan data latihan dan diuji menggunakan data uji yang telah dipisahkan sebelumnya dengan rasio pembagian 80:20,Hasil penelitian menunjukkan bahwa terdapat perbedaan signifikan dalam akurasi antara algoritma yang diuji, dengan masing-masing algoritma menunjukkan kekuatan dan kelemahannya. Efektivitas setiap model dievaluasi berdasarkan akurasi prediksi dan diinterpretasikan melalui matriks kebingungan yang dihasilkan,Visualisasi data menggunakan seaborn dan matplotlib memberikan wawasan lebih lanjut tentang performa masing-masing model, memungkinkan analisis mendalam tentang kesalahan klasifikasi yang umum terjadi. Studi ini memberikan panduan yang berguna untuk pemilihan algoritma dalam aplikasi klasifikasi sentimen berbasis emoji, yang memiliki implikasi signifikan dalam analisis media sosial dan interaksi manusia-komputer
Analisis Dampak Route Summarization Terhadap Performa OSPF Multi-Area Menggunakan GNS3 MikroTik Muhammad Ilfan Asykuri; Zainal Arifin; Mochammad Faid
TIN: Terapan Informatika Nusantara Vol 7 No 3 (2026): August 2026
Publisher : Forum Kerjasama Pendidikan Tinggi (FKPT)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47065/tin.v7i3.10209

Abstract

Open Shortest Path First (OSPF) multi-area networks at medium to large scales face the problem of Link State Database (LSDB) bloat caused by uncontrolled Link State Advertisement (LSA) propagation, particularly when each subnet is individually advertised by the Area Border Router (ABR) across all network areas. This condition increases memory consumption, the computational load of the Shortest Path First (SPF) algorithm, and routing convergence delays. This study aims to quantitatively analyze the impact of applying route summarization on the ABR regarding the number of LSAs per type, LSDB size, routing table entries, convergence time, and CPU and memory utilization of MikroTik RouterOS CHR devices in an OSPF multi-area environment. The method employed is quantitative experimental with two controlled scenarios: without route summarization as a baseline and with route summarization on the ABR as the treatment condition. The topology was built on the GNS3 simulation platform using 12 MikroTik CHR routers distributed across three areas with 24 active subnets and 1,000 dummy prefixes. Results show that route summarization reduced Type-3 LSAs by 99.71%, decreased LSDB size by 99.32%, reduced routing table entries by 99.22%, lowered peak ABR CPU usage from 11.33% to 8.33%, and achieved perfect fault isolation on the Area 0 control plane with an average convergence time of 0.2 ms compared to 1,023 ms in the baseline condition. This study contributes practical guidelines for network administrators to optimize router memory utilization and accelerate convergence in multi-area network architectures.
Plant Disease Identification Using Deep Learning: A Systematic Literature Review Mochammad Faid
Khatulistiwa SMART: Science, Methodology, Artificial intelligence, Research, and Technology Vol 1 No 1 (2024): December 2024
Publisher : CV Narasi Khatulistiwa Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.67490/kjarti.v1i1.232

Abstract

This research aims to analyze and summarize recent approaches in plant disease identification and classification using deep learning techniques. Through a systematic literature review, we evaluate the various methodologies, neural network architectures, and datasets used in recent studies in this field. Our findings show that the use of deep learning, especially by utilizing complex neural network architectures, has led to significant improvements in plant disease identification accuracy. One of the key findings is the highest accuracy achieved by the Inception Net CNN architecture-based Deep Learning method in detecting diseases in tomato plants, reaching 99.89%. These results confirm that deep learning approaches have great potential to optimize plant disease management and improve agricultural productivity globally.
Python-Based Telegram Frequently Asked Questions (FAQ) Chatbot Using Text Retrieval Faid, Mochammad; Sudriyanto, Sudriyanto; Sukron, Moh; Syaiful, Syaiful
COREAI: Jurnal Kecerdasan Buatan, Komputasi dan Teknologi Informasi Vol 7, No 1 (2026): Sustainable Information Technology Innovation Supports a Digital-Based Smart Eco
Publisher : Universitas Nurul Jadid

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33650/coreai.v7i1.15554

Abstract

This study presents the implementation of a Telegram FAQ chatbot developed in Python to address the difficulty of providing rapid and consistent responses to recurring public-service questions. The system stores question-and-answer pairs in memory, extracts FAQ entries from free-form text, and retrieves the closest answer using text similarity through Python's difflib library. The prototype also provides inline menus, an interactive quiz, FAQ reset and update commands, and an optional large language model refinement stage constrained by the retrieved context. The research followed an implementation-oriented prototyping approach comprising requirements identification, code analysis, system design, implementation review, and scenario-based functional inspection. Static analysis identified 14 principal functions, 273 built-in FAQ pairs, and 100 quiz questions. The implementation analysis indicates that lightweight retrieval can support a practical and resource-efficient FAQ automation prototype while maintaining answers within the available knowledge base. Nevertheless, production deployment requires stronger secret management, persistent storage, access control, logging, and empirical evaluation using real user questions. The study contributes a replicable architecture for controlled FAQ automation on Telegram and establishes priorities for subsequent reliability and usability testing.