Browsing by Author "Esmeli, Ramazan"

Now showing 1 - 5 of 5

AI-Driven Food Safety Risk Prediction: a Transformer-Based Approach With RASFF Database
(Emerald Group Publishing Ltd, 2025) Sari, Omer Faruk; Bader-El-Den, Mohamed; Leadley, Craig; Esmeli, Ramazan; Mohasseb, Alaa; Ince, Volkan
PurposeThis study aims to enhance food safety risk classification by systematically evaluating the effectiveness of machine learning and transformer-based AI models using the RASFF dataset. While AI-powered surveillance has gained attention, most research focuses on isolated applications of machine learning without systematically comparing them to advanced transformer architectures. This research addresses this gap by evaluating the predictive accuracy and interpretability of the model to ensure that AI-driven risk assessments are both effective and transparent for regulation.Design/methodology/approachThe study employs a structured evaluation framework in which traditional machine learning models, including logistic regression, support vector machines and random forest, are compared with advanced transformer-based models such as BERT, RoBERTa and BioBERT. Additionally, explainable AI (XAI) techniques, particularly SHAP analysis, enhance the interpretability of the models by identifying the key food safety risk factors that influence classification decisions.FindingsTransformer-based models significantly outperform traditional machine learning methods, with RoBERTa achieving the highest classification accuracy. The SHAP analysis highlights key hazards salmonella, aflatoxins, listeria and sulphites as primary factors in serious risk classification, while procedural attributes like certification status and temperature control are less impactful. Despite improvements in accuracy, computational efficiency and scalability remain challenges for real-world deployment.Originality/valueWe introduce a novel end-to-end AI framework that integrates state-of-the-art transformers with Explainable AI for the RASFF database. By integrating explainable AI, it bridges the gap between AI research and regulatory implementation and provides actionable insights for policymakers and industry stakeholders to improve risk management and early hazard detection.
An Analysis of Consumer Purchase Behavior Following Cart Addition in E-Commerce Utilizing Explainable Artificial Intelligence
(Mdpi, 2025) Esmeli, Ramazan; Gokce, Aytac
To optimize personalized offers and reduce cart abandonment, it is essential to understand customer behavior in e-commerce after products are added to the cart. Although purchase prediction models are well researched, session-level changes, including price variations, product category shifts, and geographical context, are less examined concerning their impact on machine learning models for predicting purchase behavior after cart additions. This study incorporates these factors into machine learning models to examine their impacts on predictions using explainable AI techniques. The comprehensive experimental results obtained from two datasets and eight models demonstrate that machine learning algorithms can achieve an F1 score of 89% in predicting purchase behavior following cart additions. This study highlights the significant impact of session-specific factors, like price fluctuations, category transitions, and geographical context, coupled with consumers' previous browsing patterns, on model predictions.
Predicting Spoilage Intensity Level in Sausage Products Using Explainable Machine Learning and GAN-Based Data Augmentation
(Springer, 2025) Ince, Volkan; Bader-El-Den, Mohamed; Esmeli, Ramazan; Maurya, Lalit; Sari, Omer Faruk
Spoilage in processed meat products, such as poultry and pork sausages, presents significant challenges for food safety, quality control, and waste reduction. This study presents a machine learning-based framework to classify spoilage intensity levels using sensory, physicochemical, and microbiological features. To overcome limitations caused by small datasets, we applied synthetic data augmentation using a tabular variational autoencoder (TVAE) to generate high-fidelity samples that enhance model generalization. Additionally, traditional oversampling techniques such as SMOTE and ADASYN were employed for comparative purposes and to further address class imbalance issues. Seven machine learning classifiers were evaluated logistic regression, support vector machine, K-nearest neighbors, random forest, gradient boosting, voting classifier, and multilayer perceptron. The best classification performance was achieved when models were trained on GAN-based synthetic data and tested on real samples. For poultry sausage spoilage prediction, the gradient boosting classifier reached the highest accuracy of 97%. For pork sausages, random forest achieved the highest accuracy of 95%. These results confirm the effectiveness of data augmentation in improving predictive robustness. To ensure model transparency, we integrated explainable AI techniques SHAP and LIME into the pipeline. These analyses revealed that sampling time, CO2\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$_2$$\end{document} concentration, pH, and microbial species such as Lactobacillus curvatus and Leuconostoc carnosum were among the most influential features in spoilage prediction. The combination of synthetic data generation and interpretable machine learning enables a reliable, scalable, and explainable approach to spoilage classification. This methodology has strong potential for enhancing quality control systems in the meat industry while reducing waste and improving safety along the food supply chain.
Session Context Data Integration To Address the Cold Start Problem in E-Commerce Recommender Systems
(Elsevier, 2024) Esmeli, Ramazan; Abdullahi, Hassana; Bader-El-Den, Mohamed; Can, Ali Selcuk
Recommender systems play an important role in identifying and filtering relevant products based on the behaviours of users. Nevertheless, recommender systems suffer from the 'cold-start' problem, which occurs when no prior information about a new session or a user is available. Many approaches to solving the cold-start problem have been presented in the literature. However, there is still room for improving the performance of recommender systems in the cold-start stage. In this article, we present a novel method to alleviate the cold-start problem in session-based recommender systems. The purpose of this work is to develop a session similarity-based cold-start session alleviation approach for recommendation systems. The developed method uses previous sessions' contextual and temporal features to find sessions similar to the newly started one. Our results on three different datasets show that, based on the provided Mean Average Precision and Normalised Discounted Cumulative Gain scores, the Session Similarity-based Framework consistently outperforms baseline models in terms of recommendation relevance and ranking quality across three used datasets. Our approach can be used to address the challenges associated with cold start sessions where no previously interacted items are present.
Understanding Customer Loyalty-Aware Recommender Systems in E-Commerce: an Analytical Perspective
(Springer, 2025) Esmeli, Ramazan; Can, Ali Selcuk; Awad, Aya; Bader-El-Den, Mohamed
The selection of relevant variables is critical for providing personalized product and service recommendations on e-commerce businesses. However, the integration of e-loyalty-related features into recommender systems remains underexplored. This study aims to investigate the impact of incorporating e-loyalty indicators, such as purchase frequency and platform engagement, on the performance of recommender systems in the context of e-commerce businesses. Using three well-established recommender system models and four real-world datasets, we conducted computational experiments to assess performance improvements when e-loyalty features are incorporated. The results show that integrating e-loyalty-related features significantly enhances the performance of recommendation systems, with sequential deep neural networks outperforming other algorithms. Our study contributes to the literature by highlighting the value of leveraging customer loyalty data to enhance recommendation accuracy. Theoretical implications include underscoring the importance of using longitudinal user engagement data in recommender systems to move beyond static personalization toward adaptive, behavior-aware technologies. From a practical perspective, our findings suggest that incorporating e-loyalty features can improve recommendation accuracy, offering valuable insights for e-commerce businesses seeking to personalize their services. This research offers original contributions by focusing on the role of loyalty-driven features in improving recommender systems, an area that remains largely underexplored.