PROJECTS
Machine Learning and Data professional with extensive experience in software development, data analytics, and AI-driven solutions. Skilled in designing, developing, and evaluating machine learning techniques across Artificial Neural Networks, Evolutionary Algorithms, Recommender Systems, and Computer Vision. Passionate about transforming complex data into actionable insights and building intelligent systems that support data-driven decision-making.
E. Bojnordi, A. Mitrovic, M. Galster, S. Malinen, J. Holland, “Adaptive Comment Recommendations to Support Social Learning in Active Video Watching”, User Modeling and User-Adapted Interaction, under review.
E. Bojnordi, A. Mitrovic, M. Galster, S. Malinen, J. Holland, “Fostering Interactive Engagement in Active Video Watching via Adaptive Comment Recommendations”, Artificial Intelligence in Education (AIED 2025), Springer, Cham, 6: 83–90, 2025.
E. Bojnordi, A. Mitrovic, M. Galster, S. Malinen, J. Holland, N. Mohammadhassan, “Enhancing Social Learning in Active Video Watching”, Proceedings of the 32nd International Conference on Computers in Education (ICCE 2024), Asia-Pacific Society for Computers in Education, 1: 171–176, 2024.
E. Bojnordi, A. Mitrovic, M. Galster, S. Malinen, J. Holland, “Personalized Comment Reviewing in Active Video Watching: Investigation of Learners’ Cognitive Load”, Proceedings of the 32nd International Conference on Computers in Education (ICCE 2024), Asia-Pacific Society for Computers in Education, 2: 780–782, 2024.
E. Bojnordi, A. Mitrovic, M. Galster, S. Malinen, J. Holland, “Adding Interactive Mode to Active Video Watching”, Proceedings of the 31st International Conference on Computers in Education (ICCE 2023), Asia-Pacific Society for Computers in Education, 2: 1042–1044, 2023.
E. Bojnordi, S.J. Mousavirad, M. Pedram, G. Schaefer, D. Oliva, “Improving the Generalisation Ability of Neural Networks Using a Lévy Flight Distribution Algorithm for Classification Problems”, New Generation Computing, 41(2): 225-242, 2023.
M. Hemmati, S.J. Mousavirad, E. Bojnordi, M. Shaeri, “A New Hybrid Method for Text Feature Selection through Combination of Relative Discrimination Criterion and Ant Colony Optimization”, 7th International Conference on Harmony Search, Soft Computing and Applications (ICHSA 2022), Seoul, South Korea, 2022.
E. Bojnordi, S.J. Mousavirad, G. Schaefer, I. Korovin, “MCS-HMS: A Multi-Cluster Selection Strategy for the Human Mental Search Algorithm”, IEEE Symposium Series on Computational Intelligence (SSCI), Orlando, Florida, USA, December 5-8, 2021.
E. Bojnordi, P. Moradi, “A Novel Collaborative Filtering Model Based on Combination of Correlation Method with Matrix Completion Technique”, IJST, Transaction of Electrical Engineering, 37 (E1), 93-100, 2013.
E. Bojnordi, P. Moradi, “A Novel Collaborative Filtering Model Based on Combination of Correlation Method with Matrix Completion Technique”, The 16th CSI International Symposium on Artificial Intelligence and Signal Processing, Shiraz, Iran, 2012.
S.J. Mousavirad, K. Nasri, E. Bojnordi, F. Akhlaghian Tab, “A New License Plate Detection Method by Using Controlled Threshold Changes”, 4th Conference on Information and Knowledge Technology, Babol, Iran, 2012.
Manila, Philippines • November 2024
New Zealand • Dec 2022 – Nov 2025
Iran • Sep 2009 – Oct 2012
Iran • Sep 2001 – Oct 2005
AVW-Space is an online Active Video Watching platform that promotes active and social learning through educational videos and peer interactions. In this project, I developed an adaptive comment recommender system using an ontology-based learner model to estimate a learner's domain knowledge and recommend relevant peer comments for review. The system aims to enhance knowledge components that are less represented in the learner model by providing personalised recommendations.
An end-to-end deep learning-based web application for detecting pneumonia from chest X-ray images using a ResNet-18 model and a FastAPI-based web interface. Users can upload a chest X-ray image through a simple UI, preview the image in real time, and receive an instant prediction indicating whether the image is NORMAL or PNEUMONIA, along with model performance metrics including Accuracy, Precision, Recall, F1-score, and Balanced Accuracy.
A Flask-based AI-powered security monitoring system that integrates camera-based image capture with a YOLO object detection model for real-time visual analysis. The system processes captured images, detects objects, and stores security events with timestamps in both log files and JSON format for monitoring and analysis.
a Flask-based web application that forecasts cryptocurrency prices using the Prophet time-series forecasting model. The application retrieves historical market data from the CoinGecko API, predicts future prices over a user-defined time horizon, and estimates potential investment profits based on configurable investment amount and profit thresholds. The system presents forecasted prices, confidence intervals, and recommended investment opportunities through an interactive web interface.
An executive reporting and analytics platform developed using C# and SQL Server to monitor auditor performance and operational efficiency across the Iranian National Tax Administration. The system transformed reporting workflows by reducing report generation time from days to minutes, enabling faster, data-driven decision-making. The platform centralized reporting and provided role-based dashboards for users at every organizational level, from individual tax auditors to Directors General. It delivered real-time insights into auditor productivity and regional office performance, helping management quickly identify underperforming offices, detect workflow bottlenecks, and allocate resources more effectively.
A C# image processing application for automatic grayscale image contrast enhancement using a genetic algorithm (GA)-based optimization approach. The system addressed limitations of traditional histogram-based enhancement methods, which can introduce unnatural brightness variations and visual artifacts. It automatically optimized transformation parameters to improve image contrast, preserve brightness, and enhance visual details without requiring manual parameter tuning.