Mahfudh, Adzhal Arwani
Unknown Affiliation

Published : 4 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 4 Documents
Search

Klasifikasi Berita Hoax Dengan Menggunakan Metode Naive Bayes Mustofa, Hery; Mahfudh, Adzhal Arwani
Walisongo Journal of Information Technology Vol 1, No 1 (2019): Walisongo Journal of Information Technology
Publisher : Universitas Islam Negeri Walisongo Semarang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.21580/wjit.2019.1.1.3915

Abstract

Hoaxes contain false news or non-sourced news. Today, hoaxes are very widely spread through internet media. The development of information technology that has so quickly triggered the spread of hoax information through the internet has become uncontrolled. So we need an intelligent system that can classify hoax news content that is spread through internet media. The hoax classification process can be done through the preprocessing stage then weighting the word and classification using naive bayes. Measurements were made using the 10-ford cross validation method. The results obtained from these measurements, it is known that the value of fold 6 has the highest accuracy, which is equal to 85.28% which is classified as relevant documents as much as 307 and irrelevant as much as 53 or an error rate of 14.72%. While the average value based on hoax news and true news value precision 0.896 and recall 0.853
Tingkat Ketergantungan (Usability) E-learning di Fakultas Saitek UIN Walisongo Semarang Mahfudh, Adzhal Arwani; Rizki, Favian Agung; Alfaza, Ahilla Salma
Walisongo Journal of Information Technology Vol 2, No 2 (2020): Walisongo Journal of Information Technology
Publisher : Universitas Islam Negeri Walisongo Semarang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.21580/wjit.2020.2.2.7172

Abstract

Utilizing technology, especially the internet to support the teaching and learning process, is what happened in the 21st century. E-learning is one of the learning innovations in the network that allows users to access material or collect assignments online. The benefits of e-learning are so numerous and can help lectures, users can access anywhere without being limited by space and time. The method used is descriptive research, which in its implementation consists of data collection, analysis and interpretation of the meaning of the data obtained. . This study aims to determine the level of usability or usability that exists on the e-leraning website of UIN WALISONGO Saintek Faculty, whether the system has been made that meets the usability criteria or not, in terms of Learnability, Efficiency, Memorability, Errors, and Satisfaction. After conducting research, the authors obtain data that e-learning meets usability criteria seen from the data in the field that is the result of the survey. So, Walisongo e-learning can be used as a support for learning effectively and efficiently.
Klasifikasi Pemahaman Santri Dalam Pembelajaran Kitab Kuning Menggunakan Algoritma Naive Bayes Berbasis Forward Selection Mahfudh, Adzhal Arwani; Mustofa, Hery
Walisongo Journal of Information Technology Vol 1, No 2 (2019): Walisongo Journal of Information Technology
Publisher : Universitas Islam Negeri Walisongo Semarang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.21580/wjit.2019.1.2.4529

Abstract

Kitab kuning merupakan kitab tradisional yang mengandung diraasah islamiyah yang diajarakan pada pondok pesantren, mulai dari struktur bahasa arab (ilmu nahwu dan shorof), ‘ulumul qur’an, hadits, aqidah, tasawuf/akhlaq, tafsir, fiqh sampai ilmu sosial dan kemasyarakatan (mu’amalah).  Disebut juga dengan kitab gundul karena tidak memiliki harakat (fathah, kasroh, dhammah, sukun) untuk bisa membaca dan memahami secara menyeluruh dibutuhkan waktu yang relatif lama. Penelitian ini bertujuan untuk mendapatkan model klasifikasi dari data pembelajaran kitab kuning di pondok pesantren. Metode yang digunakan dalam penelitian ini adalah forward selection sebagai praproses dalam mengurangi dimensi data, menghilangkan data yang tidak relevan dan naive bayes yang berguna untuk mengklasifikasi data. Hasil dari klasifikasi data pembelajaran kitab kuning menggunakan atribut yang telah diklasifikasi berdasarkan fitur-fiturnya dan dilakukan iterasi pada cross validation sehingga menghasilkan akurasi yang tepat. Berdasarkan hasil pengujian dengan dua metode, pengujian dengan algoritma Naive bayes saja menghasilkan akurasi 96,02%, untuk algoritma Naive bayes berbasis forward selection menghasilkan akurasi 97,38% . Terdapat peningkatan akurasi dengan penambahan fitur seleksi.
Comparative Study of SVM and Decision Tree Algorithms on the Effect of SMOTE Technique on LinkAja Application Faruq, Muhammad Kholfan; Umam, Khothibul; Mustofa, Mokhamad Iklil; Mahfudh, Adzhal Arwani
Journal of Applied Informatics and Computing Vol. 9 No. 6 (2025): December 2025
Publisher : Politeknik Negeri Batam

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30871/jaic.v9i6.9806

Abstract

The widespread adoption of digital wallets like LinkAja in Indonesia has led to a surge in user-generated reviews, which are valuable for assessing service quality. This study compares the classification performance of Support Vector Machine (SVM) and Decision Tree algorithms on user reviews from the LinkAja application. 7.000 reviews were gathered through web scraping and processed with standard text cleaning, tokenization, stopword removal, and stemming, resulting in 6,261 usable entries. These were divided into training and testing sets in a 70:30 ratio. The performance of each algorithm was evaluated both before and after the application of Synthetic Minority Oversampling Technique (SMOTE) to address class imbalance. Prior to SMOTE, SVM recorded an accuracy of 77.97%, precision of 0.74, recall of 0.33, and F1 score of 0.45, while Decision Tree reached 72.01% accuracy, 0.50 precision, 0.62 recall, and 0.55 F1 score. After SMOTE, SVM accuracy slightly improved to 78.29%, with notable increases in recall (0.74) and F1 score (0.60); Decision Tree also saw an accuracy rise to 74.56% but experienced a slight decline in F1 score to 0.52. These findings demonstrate that SVM, particularly when used with SMOTE, offers better overall performance and class balance in classifying reviews with imbalanced sentiment distribution, making it more suitable than Decision Tree for this application.