Sarwosri

Sarwosri

Auto-generated profile, based on publicly available information

Sarwosri

Department of Informatics, Faculty of Intelligent Electrical and Informatics Technology, Institut Teknologi Sepuluh Nopember (ITS) Sukolilo, Surabaya, Indonesia

Research Focus & Metrics

Academic Career Length

3Years (2024-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
20 November 2024 in IEEE Access
Verified
Journal Article
Open access
SASarwosri
Sarwosri
UYUmi Laili Yuhana
Umi Laili Yuhana
SRSiti Rochimah
Siti Rochimah
Functional requirement identification is the initial stage in software development and a crucial step in the success of software development. Good functional requirements sentences should be consistent with each other or conflict-free. There are already ways used to detect conflicting functional requirements. One of the methods, the Rule-based System (RBS), can pair conflicting functional requirements sentences but needs 1.6s for computing time for checking conflict or conflict-free requirements and 0.52 of accuration value. This study proposed a combination of the Clustering method and Rule-Based System to reduce computing time and increase accuracy in detecting conflict requirements. The evaluation used four public datasets containing 435 sentences and two private datasets containing 420 sentences. The proposed method reduces 97% computing time for a combination of K-Means and RBS. The proposed method can increase the 23% accuracy score for a combination of Agglomerative and RBS and Mean-Shift and RBS. The evaluation verified the proposed method to reduce computing time and increase accuracy for detecting conflict functional requirements.

1 June 2026 in IEEE Access
Verified
Journal Article
Open access
SASarwosri
Sarwosri
BNBintang Nuralamsyah
Bintang Nuralamsyah
NKNathaniel Ryo Kurniadi
Nathaniel Ryo Kurniadi
AWAbdur Rach-Man Wahed
Abdur Rach-Man Wahed
SRSiti Rochimah
Siti Rochimah
E-commerce fraud poses a significant threat to digital economies, causing financial losses and weakening consumer trust in online platforms. Existing detection systems typically rely on supervised machine learning (ML) classifiers such as Artificial Neural Networks, Random Forest, and Gradient Boosting applied to tabular transaction features. ML classifiers process transactions independently and cannot capture relationships across shared entities such as devices, IP addresses, and payment methods. Other approaches use Graph Neural Networks to capture relational dependencies between transactions. End-to-end GNN classifiers present deployment challenges including high computational cost and limited interpretability. To address this limitation, this study proposes a hybrid fraud detection pipeline to extract relational embeddings from a transaction graph and combine them with tabular features for conventional machine learning classifiers. Experiments are conducted on the Kaggle Fraudulent E-commerce Transac-tion dataset, which contains 1,472,952 training records and 23,634 test records. Three classifiers, namely Artificial Neural Networks, Random Forest, and Gradient Boosting, are evaluated under three scenarios: tabular only, end-to-end GNN, and hybrid GNN embedding. Across all three classifiers, augmenting tab-ular features with relational embeddings yields consistent improvements over the tabular-only baseline. The Random Forest hybrid pipeline obtains the most balanced result (accuracy = 0.9688, F1 = 0.6091, PR-AUC = 0.6171), although the absolute gain over the best tabular baseline is modest, indicating that GNN-based feature embedding provides a complementary rather than transformative source of predictive signal.