Muhammad Yusril Helmi Setyawan
Universitas Logistik dan Bisnis Internasional

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

Found 4 Documents
Search

Model Switching Hybrid Untuk Menangani User dan Item Cold-Start Muhammad Ilman Aqilaa; Muhammad Yusril Helmi Setyawan; Cahyo Prianto
Jurnal Teknik Informatika dan Sistem Informasi Vol 12 No 1 (2026): JuTISI
Publisher : Maranatha University Press

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.28932/jutisi.v12i1.12779

Abstract

Recommender systems face significant challenges under cold-start conditions, where information about users or items is still limited. This study proposes a hybrid switching approach that adaptively combines Content-Based Filtering (CBF), User-Based Collaborative Filtering (CF), and Item-Based CF based on the number of user and item interactions. The evaluation was conducted through cold-start scenario testing for a single user, accuracy measurement using RMSE and MAE with 5-Fold Cross-Validation, and adaptivity testing under varying levels of cold-start conditions (5%, 20%, and 50%). Experimental results show that the hybrid model effectively handles all cold-start scenarios by falling back to CBF or CF User-Based when data is insufficient, and opting for CF Item-Based when sufficient information is available. The model achieved the best performance with an average RMSE of 0.8165 and MAE of 0.6592, along with low standard deviations, indicating stable performance across folds. Furthermore, the hybrid system demonstrated dynamic adaptability to data completeness levels, with a gradual shift in fallback algorithm usage as cold-start severity increased. Therefore, the hybrid switching approach not only excels in accuracy but also offers flexibility and robustness, making it an effective solution for improving the quality of recommender systems in scenarios with incomplete data.
Text-Based Recommender Systems for User Cold-Start Problems in E-Commerce Product Recommendation: A Systematic Literature Review Muhammad Yusril Helmi Setyawan; Haris Saefuloh
SISTEMASI Vol 15, No 9 (2026): Sistemasi: Jurnal Sistem Informasi
Publisher : Universitas Islam Indragiri

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.32520/stmsi.v15i9.6817

Abstract

The user cold-start problem remains a major challenge in personalized e-commerce product recommendation because new users have little or no interaction history, limiting the effectiveness of conventional collaborative filtering and rating-based models. Textual information, such as user reviews, item reviews, product descriptions, metadata, and semantic product attributes, can provide valuable initial signals for modeling user preferences and product characteristics. This study aims to identify textual data sources, AI/NLP methods, and model families suitable for addressing the user cold-start problem in e-commerce product recommendation. A systematic literature review was conducted following the PRISMA 2020 framework, covering 55 selected articles from the Scopus database that were screened and extracted with the assistance of Watase. The studies were categorized into 31 core studies and 24 supporting studies and assessed using five quality criteria. Unlike previous cold-start reviews that broadly map mitigation strategies, this review narrows its scope to text-based evidence and links each model family to the data conditions required for its application. The results show that review text is the dominant textual data source, appearing in 33 of the 55 studies, followed by product descriptions and metadata. The studies were mapped into seven model families, and 43 studies (78.2%) explicitly reported evaluation metrics. Prominent model directions include hybrid deep sentiment-aware recommenders, review-based transformer or attention models, SBERT/BERT-based recommendation, graph-based recommendation, and ontology-based semantic recommendation. Based on the reported evidence, text-semantic hybrid recommenders represent a promising direction, although differences in datasets, evaluation protocols, and metrics across studies prevent direct performance comparisons.
Explainable Service Boundary Decision Framework for Monolith Migration Using DDD and Quantitative Metrics Irgi Achmad Fauzi; Muhammad Yusril Helmi Setyawan; Mohamad Nurkamal Fauzan
Jurnal Sistem Komputer dan Informatika (JSON) Vol. 8 No. 1 (2026): September (2026)
Publisher : Universitas Budi Darma

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30865/json.v8i1.10013

Abstract

Identifying a service boundary is a critical decision in monolith-to-microservices migration. Intuition alone is hard to reproduce, while technical metrics without domain context can overlook business semantics. This study develops an explainable boundary-decision framework linking Domain-Driven Design (DDD), implementation evidence, and quantitative metrics for the FemaleFit e-commerce monolith. Domain mapping traced business capabilities to endpoints, controllers, models, and tables, producing 13 initial contexts and seven atomic candidates. Candidates were scored with M01–M13 under three weighting schemes (W1–W3), a practical-tie rule, and boundary sensitivity analysis. C01 Identity & Access and C03 Craft Catalog formed a practical tie, differing by 0.002803 under W1. C03 was selected on the judgement that it had clearer domain alignment, an explicit routing seam, a verified core flow, and a more contained blast radius. It was extracted as a Catalog Service behind a Strangler Gateway, with 34 unique HTTP method–path pairs reconciled against the migration contract. Functional evaluation covered one representative case for each of the 34 frozen Catalog contracts, including negative/error cases. All 540 retained read requests completed without failure across three independent runs per condition; latency findings were descriptive and mixed. The metrics indicate lower structural coupling and process-level deployment independence, but not database independence. Findings are bounded by one case, one service, a sequential local workload, and a shared database.
A non-invasive IoT model for acquiring learning behaviour data: A design science approach Marwanto Rahmatuloh; Virdiandry Putratama; Saepudin Nirwan; Muhammad Yusril Helmi Setyawan
Jurnal Inovasi dan Teknologi Pembelajaran Vol. 13 No. 2 (2026)
Publisher : Universitas Negeri Malang in collaboration with APSTPI and IPTPI

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.17977/um031v13i22026p158

Abstract

Digital learning environments still rely heavily on software interaction logs and self-report instruments to understand student learning behaviour, approaches that are indirect, subjective, and weakly representative of real behaviour. This study designs and evaluates a non-invasive Internet of Things (IoT) model for acquiring learning behaviour data, analysed using Artificial Intelligence, within a Design Science Research framework. The model integrates RFID/NFC, PIR, LDR, and DHT22 sensors to record presence, light physical activity, and environmental conditions at room level, without capturing images, audio, or any identifiable field. A multi-session proof-of-concept experiment (2,582 complete records across three independent sessions in a computer laboratory) was analysed using K-Means clustering, Decision Tree classification, and correlation analysis. Pooled analysis identified two well-separated activity clusters (silhouette = 0.42) and 96.8% hold-out accuracy; cross-day validation, however, showed that these results reflect between-session differences rather than generalisable behaviour, with leave-one-session-out accuracy falling to 44–60% and pooled environment–activity correlations failing to replicate within sessions. The findings demonstrate the technical feasibility of the acquisition model while providing concrete evidence that session-agnostic validation of IoT-based behavioural data is misleading, and caution against feeding such indicators into automated learning-management decisions before cross-session validation.