Article Per Year (5 Year)
p-Index From 2021 - 2026
0.947
P-Index
This Author published in this journals
All Journal Jurnal Optimasi Sistem Industri Journal of Information Technology and Computer Science JATI (Jurnal Mahasiswa Teknik Informatika) SPEKTA (Jurnal Pengabdian Kepada Masyarakat : Teknologi dan Aplikasi) Jurnal Pengabdian Masyarakat (ABDIRA)
Sulthan Rafif
Universitas Telkom
Aerospace Engineering Computer Science & IT Control & Systems Engineering Decision Sciences, Operations Research & Management Earth & Planetary Sciences Economics, Econometrics & Finance Electrical & Electronics Engineering Energy Engineering Environmental Science Industrial & Manufacturing Engineering Languange, Linguistic, Communication & Media Materials Science & Nanotechnology Mechanical Engineering Nursing Social Sciences Other

Published : 5 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 5 Documents
Search

 1 
Multilabel Classification for Keyword Determination of Scientific Articles Rafif, Sulthan; Setya Perdana, Rizal; Pandu Adikara, Putra
Journal of Information Technology and Computer Science Vol. 8 No. 2: August 2023
Publisher : Faculty of Computer Science (FILKOM) Brawijaya University

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.25126/jitecs.202382560

Abstract

In writing scientific articles, there are provisions regarding the structure or parts of writing that must be fulfilled. One part of the scientific article that must be included is keywords. The process of determining keywords manually can cause discrepancies with the specific themes discussed in the article. Thus, causing readers to be unable to reach the scientific article. The process of determining the keywords of scientific articles is determined automatically by the classification method. The classification process is carried out by determining the set of keywords owned by each scientific article data based on the abstract and title. Therefore, the classification process applied is multi-feature and multi-label. Classification is done by applying the Contextualized Word Embedding Method. The implementation of Contextualized Word Embedding Method is done by applying BERT Model. By applying the BERT Model, it is expected to provide good performance in determining the keywords of scientific articles. The evaluation results by applying the BERT Model to the case of multi-label classification on abstract data for keyword determination resulted in a loss value of Training Data is 0.514, loss value of Validation Data is 0.511, and an accuracy value of 0.71, a precision value of 0.71, a recall value of 0.71, an error value of 0.29 and f-1 score of 0.83. Based on the results of the evaluation, it shows that the BERT Classification Model can carry out a classification process to determine a set of keywords from each abstract data in scientific articles.
Developing and Managing MSME Websites to Improve Kampoeng Ilmu's Operational Performance Alifah, Amalia Nur; Rachmaniar, Desita Nur; Mustaqim, Tanzilal; Rafif, Sulthan
SPEKTA (Jurnal Pengabdian Kepada Masyarakat : Teknologi dan Aplikasi) Vol. 6 No. 2 (2025)
Publisher : Universitas Ahmad Dahlan

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.12928/spekta.v6i2.13871

Abstract

Background: MSMEs in Kampoeng Ilmu Surabaya continue to face obstacles in product marketing and digital adoption, which limit their operational growth. This program aims to address these issues by developing an integrated website and improving digital literacy among MSME partners to strengthen their online visibility and business sustainability. Contribution: The program contributes to the community by providing a digital platform (UMKMCerdas website) that enables MSMEs to independently manage product data, storefront profiles, and promotional information. It also enhances partners’ capacity to operate digital tools, supporting long-term empowerment and competitiveness. Method: The implementation consisted of five structured stages: needs analysis through interviews, website design with UI/UX and database planning, website development using PHP (Laravel), MySQL, Tailwind, and Filament, training and mentoring for MSME partners, and evaluation through continuous monitoring. A participatory approach was used to ensure active involvement and skill transfer to 82 MSME actors. Results: The integrated website successfully provides features such as bookstore profiles, product catalogs, Google Maps integration, WhatsApp contacts, and an admin dashboard. Post-training responses showed significant enhancement in partners’ confidence and ability to use digital tools for promoting their businesses and managing information. Conclusion: The program effectively strengthens the digital capabilities of Kampoeng Ilmu MSMEs, enabling them to manage business content independently and expand their market reach.
A Hybrid DDMRP-OUTL Inventory Policy with Defect Prediction for Resilient Supply Chains Erly Ekayanti Rosyida; Ilyas Mas'udin; Nisa Isrofi; Sulthan Rafif
Jurnal Optimasi Sistem Industri Vol. 25 No. 1 (2026): Published in June 2026
Publisher : The Industrial Engineering Department of Engineering Faculty at Universitas Andalas

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.25077/josi.v25.n1.p63-93.2026

Abstract

The high variability of consumer demand makes the development of inventory strategies crucial, especially regarding operational inventory resilience. Combining defect prediction with inventory strategies is crucial amidst uncertainty related to quality. Conventional Demand-Driven Material Requirements Planning (DDMRP) strategies are sensitive to shifts in consumer demand based on buffers for replenishment. However, this strategy has the disadvantage of not considering losses due to defective production output quality. This study develops a hybrid inventory model by combining DDMRP with Order-Up-To-Level (OUTL) replenishment management, and defect rate prediction. Production output is assessed from estimated defect rates converted into yield factors. OUTL is used for conditional quantity setting by determining the amount of excess replenishment. Defect rate prediction uses a manufacturing defect dataset along with production volume, supplier quality, maintenance hours, time percentage, and worker productivity. The initial predictive element in inventory simulation using a random forest regressor configuration achieved an R² value of 0.7208. Numerical experiments to evaluate the inventory model used 24 scenarios over a 200-day daily review period. Scenarios were conducted by integrating various demand patterns, production process conditions, and production capacity limitations. The DDMRP-OUTL hybrid strategy model can reduce the Bullwhip Effect Ratio (Ratio of Echelon Logistics - REL) compared to conventional DDMRP for various scenarios, and the most significant reduction is close to 24% under intermittent demand. Furthermore, it demonstrates a higher average inventory increase as a trade-off between replenishment stability and inventory load. Stockout events are not consistently reduced across all scenarios, although the integration of defect rate prediction and the DDMRP-OUTL hybrid model leads to replenishment stability, and inventory load and service reliability must be balanced when implementing this policy.
REKOMENDASI KATA KUNCI ARTIKEL ILMIAH BERBASIS CONTEXTUALIZED WORD EMBEDDING DAN AGGLOMERATIVE CLUSTERING Sulthan Rafif; Faisal Zaidan Pradana
JATI (Jurnal Mahasiswa Teknik Informatika) Vol. 10 No. 2 (2026): JATI Vol. 10 No. 2
Publisher : Institut Teknologi Nasional Malang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.36040/jati.v10i2.17805

Abstract

Kata kunci adalah alat untuk memudahkan pembaca menemukan artikel ilmiah. Kata kunci yang ditentukan dengan tepat akan menarik perhatian pembaca untuk mengkaji dan mengutip artikel ilmiah. Kata kunci yang tidak ditentukan secara tepat dapat mengurangi kesempatan artikel ilmiah dijangkau oleh pembaca. Berdasarkan permasalahan tersebut perlu menerapkan penentuan kata kunci secara otomatis untuk mengurangi kesalahan dan mempercepat pekerjaan. Pada penelitian sebelumnya diterapkan pendekatan Contextualized Word Embedding, Clustering, dan Part-of-Speech Tagging dalam ekstraksi kata kunci. Kombinasi pendekatan Contextualized Word Embedding dan Part-of-Speech diterapkan pada penelitian sebelumnya dengan Model PatternRank. Model PatternRank dalam ekstraksi kata kunci hanya mempertimbangkan perspektif dari satu artikel ilmiah yang menyebabkan informasi yang diperoleh terbatas sehingga perlu menerapkan pendekatan clustering. Pendekatan clustering dalam ekstraksi kata kunci pada penelitian sebelumnya menerapkan TF-IDF dalam proses embedding. TF-IDF tidak memperhatikan urutan kata, sehingga menyebabkan nilai embedding yang dihasilkan dari TF-IDF tidak sesuai dengan konteks kalimat. Sehingga berdasarkan kelemahan dari pendekatan Contextualized Word Embedding dan Clustering, pada penelitian ini menerapkan kombinasi dari kedua pendekatan tersebut dalam ekstraksi kata kunci. Berdasarkan nilai indikator precision, recall, dan f1-score hasil dari evaluasi menunjukkan kombinasi dari Metode Clustering dan Contextualized Word Embedding mengungguli Metode PatternRank.
Penguatan Literasi Data Siswa SMA Wachid Hasyim 2 Taman melalui Pelatihan Excel Desita Nur Rachmaniar; Rifdatun Ni’mah; Zerlina Agustiya Putri; Sulthan Rafif
Jurnal Pengabdian Masyarakat (ABDIRA) Vol 6, No 3 (2026): Abdira, Juli
Publisher : Universitas Pahlawan Tuanku Tambusai

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.31004/abdira.v6i3.2127

Abstract

Data literacy is an essential competency that enables students to understand, process, and interpret information in the digital era. However, the use of computers in schools is often limited to basic activities, resulting in insufficient skills in data processing and visualization. This community service program aimed to strengthen the data literacy of eleventh-grade students at SMA Wachid Hasyim 2 Taman through training in data processing and visualization using Microsoft Excel. The program employed a hands-on practice approach based on the learning-by-doing method, consisting of lectures, guided exercises, case studies, and quizzes for evaluation. A total of 38 students participated in the training, which covered data cleaning, basic data processing techniques, and chart creation and interpretation. The evaluation results showed an average achievement score of 67%, with the highest performance in basic data concepts and data quality (68%). These findings indicate that practice-based training is effective in enhancing students’ data literacy and supporting the development of essential twenty-first-century skills.
Co-Authors Alifah, Amalia Nur Desita Nur Rachmaniar Erly Ekayanti Rosyida Faisal Zaidan Pradana Ilyas Mas'udin Nisa Isrofi Pandu Adikara, Putra Rifdatun Ni’mah Setya Perdana, Rizal Tanzilal Mustaqim Zerlina Agustiya Putri