Article Per Year (5 Year)
p-Index From 2021 - 2026
1.332
P-Index
This Author published in this journals
All Journal Infotech Journal Infotek : Jurnal Informatika dan Teknologi JURNAL PENGABDIAN KEPADA MASYARAKAT (ADI DHARMA) J-ENSITEC (Journal of Engineering and Sustainable Technology) Jurnal Pengabdian Kepada Masyarakat
Hafidz Sanjaya
Universitas Yayasan Pendidikan Imam Bonjol Majalengka
Agriculture, Biological Sciences & Forestry Humanities Civil Engineering, Building, Construction & Architecture Computer Science & IT Control & Systems Engineering Decision Sciences, Operations Research & Management Economics, Econometrics & Finance Energy Engineering Environmental Science Industrial & Manufacturing Engineering Mechanical Engineering Social Sciences

Published : 8 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 8 Documents
Search

 1 
ANALISIS SENTIMEN PUBLIK TERHADAP PENJUALAN IPHONE 16 DAN KEBIJAKAN TKDN DI INDONESIA Fajar Maula Hidayat; Hafidz Sanjaya
INFOTECH journal Vol. 11 No. 1 (2025)
Publisher : Universitas Majalengka

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.31949/infotech.v11i1.13159

Abstract

Kebijakan Tingkat Komponen Dalam Negeri (TKDN) terhadap produk Apple di Indonesia telah memicu berbagai opini di media sosial, terutama di platform X. Penelitian ini bertujuan untuk menganalisis sentimen publik terhadap kebijakan tersebut menggunakan metode machine learning. Data dikumpulkan melakukan teknik crawling, kemudian diproses dengan tahapan preprocessing untuk meningkatkan kualitas teks. Algoritma Random Forest diterapkan untuk mengklasifikasi opini menjadi kategori negatif, netral, dan positif. Hasil eksperimen menunjukkan bahwa model Random Forest mencapai akurasi 91%, presisi 91%, recall 91% dan f1-score 89%. Temuan ini memberikan wawasan bagi pelaku industri dan pembuat kebijakan dalam memahami persepsi masyarakat terkait kebijakan TKDN terhadap produk Apple, sehingga dapat menjadi pertimbangan dalam perumusan kebijakan selanjutnya.
MODEL PREDIKSI JUMLAH CALON SANTRI BARU DI PONDOK PESANTREN AT-TADZKIR MAJA MENGGUNAKAN PROPHET Hafidz Sanjaya; Dwi Purnomo; Fajar Maula Hidayat; Heri Wiranto
INFOTECH journal Vol. 11 No. 2 (2025)
Publisher : Universitas Majalengka

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.31949/infotech.v11i2.16831

Abstract

Penelitian ini bertujuan untuk memodelkan dan memprediksi jumlah calon santri baru di Pondok Pesantren At-Tadzkir Maja dengan memanfaatkan algoritma Prophet berbasis data harian periode tahun 2022–2025. Dataset yang digunakan berasal dari sistem penerimaan santri baru dan menunjukkan pola fluktuatif yang signifikan, dengan rentang nilai yang bervariasi antara hari dengan jumlah pendaftar rendah hingga hari-hari dengan lonjakan pendaftaran yang tinggi. Setelah dilakukan proses pra-pemrosesan yang mencakup normalisasi tanggal, agregasi harian, serta penyiapan struktur data yang sesuai untuk Prophet, model kemudian dilatih untuk mempelajari komponen tren, musiman mingguan, dan musiman tahunan. Hasil pemodelan menunjukkan adanya tren penurunan jumlah pendaftar dari tahun ke tahun, serta pola musiman yang kuat terutama pada awal pekan dan periode tertentu dalam satu tahun. Evaluasi menggunakan MAE sebesar 1,52 dan RMSE sebesar 2,06 menunjukkan bahwa model mampu merepresentasikan pola historis dengan tingkat kesalahan yang relatif rendah, meskipun tantangan masih muncul pada prediksi lonjakan harian yang bersifat sporadis. Secara keseluruhan, model Prophet terbukti efektif untuk memberikan gambaran prediktif yang dapat dimanfaatkan sebagai dasar perencanaan dan pengambilan keputusan strategis di lingkungan pesantren.
PEMODELAN TOPIK BERITA NASIONAL INDONESIA MENGGUNAKAN LATENT DIRICHLET ALLOCATION Fajar Maula Hidayat; Cahyadi; Hafidz Sanjaya; Dwi Purnomo; Heri Wiranto
INFOTECH journal Vol. 12 No. 1 (2026)
Publisher : Universitas Majalengka

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.31949/infotech.v12i1.16926

Abstract

Penelitian ini membahas penerapan metode Latent Dirichlet Allocation (LDA) untuk pemodelan topik berita terkini di Indonesia. Data dikumpulkan melalui RSS feed dari beberapa portal berita nasional seperti Detik, Kompas, Tribunnews, Liputan6, Tempo, CNN Indonesia, dan Antara News. Proses penelitian meliputi tahapan pengambilan data, pembersihan dan preprocessing teks, eksplorasi awal frekuensi kata, penyusunan representasi korpus, pemodelan topik LDA, visualisasi interaktif dengan pyLDAvis, serta evaluasi model menggunakan metrik coherence score. Hasil analisis menunjukkan model LDA dengan lima topik memberikan distribusi kata kunci yang relevan dengan isu-isu utama seperti bencana, politik, demonstrasi, korupsi, dan kriminal. Nilai coherence score sebesar 0,3591 mengindikasikan tingkat koherensi cukup baik, meskipun terdapat ruang optimasi melalui penyesuaian parameter. Visualisasi interaktif menunjukkan keterpisahan topik yang memadai, dengan tumpang tindih yang relatif kecil. Temuan ini memperlihatkan bahwa LDA efektif untuk mengidentifikasi topik dominan dalam berita nasional, sehingga dapat dimanfaatkan untuk analisis tren isu publik, pengelompokan konten media, serta mendukung pengambilan keputusan berbasis data.
Hardware Requirements Analysis for Administrative Staff in Higher Education Hafidz Sanjaya; Fajar Maula Hidayat
J-ENSITEC (Journal of Engineering and Sustainable Technology) Vol. 12 No. 02 (2026): June 2026
Publisher : Universitas Majalengka

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.31949/j-ensitec.v12i02.18424

Abstract

The use of computer hardware that does not match workload requirements remains a common issue in higher education administrative activities, leading to reduced efficiency and productivity. This study aims to analyze hardware requirements for administrative staff at the Faculty of Psychology, Business, and Technology, Universitas Yayasan Pendidikan Imam Bonjol Majalengka, and to propose optimal hardware specifications. A case study approach was employed by evaluating three devices used by three staff members through observation and system performance assessment. The main contribution of this study lies in providing a workload-based hardware analysis supported by simple performance benchmarking to demonstrate the impact of hardware upgrades. The results indicate that existing systems, equipped with older mid-range processors, 4–8 GB RAM, and HDD storage, are insufficient to support multitasking and data processing efficiently. Performance testing shows that data input processing time, initially ranging from 10–12 minutes, can be reduced to 4–6 minutes using the recommended specifications. This represents an efficiency improvement of approximately 50–60% and a significant reduction in system latency. Therefore, this study offers a more measurable and evidence-based approach to hardware planning compared to conventional descriptive methods.
PELATIHAN ENGLISH SPEAKING BERBASIS ARTIFICIAL INTELLIGENCE UNTUK MENINGKATKAN SELF-CONFIDENCE MAHASISWA Cahyadi; Dwi Purnomo; Heri Wiranto; Fajar Maula Hidayat; Hafidz Sanjaya
Jurnal Pengabdian Kepada Masyarakat Vol 4 No 2 (2026): JURNAL PENGABDIAN KEPADA MASYARAKAT (PENGMAS)
Publisher : Pusat Penelitian dan Pengabdian pada Masyarakat (P3M)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.59820/pengmas.v4i2.534

Abstract

The rapid development of Artificial Intelligence (AI) has significantly influenced various educational practices, including English language learning. Many students still experience difficulties in speaking English due to low self-confidence, fear of making mistakes, and limited opportunities for speaking practice. This community service activity aimed to enhance students’ self-confidence through AI-based English speaking training. The program involved 25 students who participated in interactive speaking practices using AI applications such as ChatGPT and AI voice assistants. The methods employed included training sessions, demonstrations, guided speaking practice, discussions, and evaluations through questionnaires and observations. The results indicated that participants experienced improvements in self-confidence, motivation, and speaking participation after attending the training. Students perceived AI-based learning as a flexible, interactive, and supportive medium for independent speaking practice. Furthermore, the implementation of AI in English language learning created a more comfortable learning environment, reducing speaking anxiety. Therefore, AI-based speaking training was found to be effective in enhancing students’ self-confidence and communication skills in English language learning.
Perbandingan Algoritma Machine Learning untuk Klasifikasi Diabetes Menggunakan Feature Selection dan Hyperparameter Tuning Fajar Maula Hidayat; Hafidz Sanjaya
Infotek: Jurnal Informatika dan Teknologi Vol. 9 No. 2 (2026): Infotek : Jurnal Informatika dan Teknologi
Publisher : Fakultas Teknik Universitas Hamzanwadi

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.29408/jit.v9i2.35199

Abstract

Diabetes mellitus is a chronic disease with an increasing prevalence worldwide, requiring early detection to support faster and more accurate disease management. This study aims to compare the performance of several machine learning algorithms for diabetes classification using feature selection and hyperparameter tuning. The dataset used was the Pima Indians Diabetes Dataset obtained from the Kaggle platform. The research consisted of data preprocessing, feature selection using SelectKBest, training and testing data splitting, hyperparameter tuning using GridSearchCV, and model evaluation using accuracy, precision, recall, F1-score, ROC-AUC, and cross validation. The evaluated algorithms included Logistic Regression, Support Vector Machine (SVM), Random Forest, K-Nearest Neighbor (KNN), and Naive Bayes. The results showed that the KNN algorithm achieved the best performance with an accuracy of 74.02%, precision of 63.46%, recall of 61.11%, F1-score of 62.26%, and ROC-AUC of 79.60%. The findings indicate that integrating data preprocessing, feature selection, hyperparameter tuning, and cross validation provides a more comprehensive evaluation process for machine learning models in diabetes classification.
PELATIHAN SISTEM PEMESANAN KAMAR HOTEL BERBASIS DIGITAL DI TWINS HOTEL SYARIAH BANDUNG Dwi Purnomo; Cahyadi; Heri Wiranto; Hafidz Sanjaya
JURNAL PENGABDIAN KEPADA MASYARAKAT (ADI DHARMA) Vol 3 No 2 (2025): JURNAL PENGABDIAN KEPADA MASYARAKAT (ADI DHARMA)
Publisher : ABISATYA DINAMIKA ISWARA PUBLISHING

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.58268/adidharma.v3i2.178

Abstract

Twins Hotel Syariah Bandung is currently developing a digital-based service system to improve the efficiency and quality of services provided to guests. In the room reservation process, the hotel continues to face challenges such as manual data recording, errors in data management, and limitations in reporting. To address these issues, a team of lecturers from YPIB Majalengka of University initiated a training program on a digital hotel room reservation system designed to be simple yet capable of overcoming the classic limitations of manual processes. Furthermore, the need for ease of information delivery, service availability, and accessibility is crucial for the future development of the hotel's business. The activity was carried out in several stages, including system needs identification, system analysis and design, system development, preparation of training materials, implementation of the system in the hotel environment, and intensive training for relevant staff. The methods used include in-person and online training, hands-on practice with the system, and evaluation sessions to identify challenges during and after the training. The results showed an increase in staff skills in using the hotel room reservation application, a reduction in administrative errors, and improved time efficiency in serving guests. Additionally, the system is capable of generating automatic reports that assist management in decision-making. Overall, this community service activity had a positive impact and can serve as a model for the application of information technology-based community engagement that is practical and sustainable.
IMPLEMENTASI MODEL PREDIKSI JUMLAH CALON SANTRI PADA SISTEM PENERIMAAN SANTRI BARU Hafidz Sanjaya; Fajar Maula Hidayat; Dwi Purnomo; Heri Wiranto
JURNAL PENGABDIAN KEPADA MASYARAKAT (ADI DHARMA) Vol 4 No 1 (2025): JURNAL PENGABDIAN KEPADA MASYARAKAT (ADI DHARMA)
Publisher : ABISATYA DINAMIKA ISWARA PUBLISHING

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.58268/adidharma.v4i1.231

Abstract

This community service program was conducted to address the need of Pondok Pesantren At-Tadzkir Maja for a more effective planning mechanism regarding the number of new student applicants, which previously lacked a data-driven prediction system. The purpose of this program is to implement a prediction model for forecasting the number of prospective students by utilizing the Prophet algorithm developed in prior research, and to provide an accessible predictive dashboard for operational use by the institution. The implementation involved needs assessment, model integration into the online admission system, dashboard development, user training, and evaluation of the administrators’ understanding. The results indicate that the model runs automatically and produces informative daily forecasts, while the dashboard enables administrators to interpret trend and seasonal patterns to support capacity planning and admission strategies. The program concludes that the implementation successfully improves the pesantren’s ability to make data-driven decisions, and it is recommended that future work integrates additional external variables and ensures regular model updates to maintain predictive accuracy.
Co-Authors Cahyadi Dwi Purnomo Fajar Maula Hidayat Heri Wiranto