Building reliable AI pipelines for network security and IoT intrusion detection.
Machine Learning Researcher · Data Analyst · Python Instructor · IT Administrator. I combine academic research in ML and computer networks with real-world experience in IT administration and programming education.
About Me
Machine Learning Researcher · Data Analyst · Python Instructor · IT Administrator
Mohammadali Mousavireineh
I combine academic research in machine learning and computer networks with real-world experience in IT administration and programming education. My current research focuses on optimized feature engineering, leakage-free machine learning pipelines, and robust intrusion detection for IoT and network security environments.
Featured Research
Manuscript under review · Multimedia Tools and Applications (Springer)
Optimized Feature Engineering for IoT Intrusion Detection: Synergy of Feature Selection, Feature Extraction, and Classifier Ensemble
This study investigates how carefully designed feature selection and feature extraction pipelines can improve machine learning-based intrusion detection in IoT and heterogeneous network environments. A unified, leakage-free experimental framework is evaluated across UNSW-NB15, AWID, and CSE-CIC-IDS2018. The work compares filter-based feature selection methods, feature extraction techniques, twelve classifier families, and stacking-based meta-learning to identify robust and efficient IDS configurations.
- Leakage-free, cross-dataset IDS benchmark
- Comparison of FS-only and FS→FE pipelines
- Ablation-driven analysis of each pipeline component
- Focus on efficient and scalable IDS design for IoT environments
Pipeline Methodology
A controlled, leakage-free experimental framework
Preprocessing
Cleaning, encoding, stratified partitioning, and training-only fitting to avoid information leakage.
Feature Selection
Variance Threshold, ANOVA F-test, and Chi-Squared filters to reduce redundant attributes.
Feature Extraction
PCA, LDA, ICA, and truncated SVD to produce compact discriminative representations.
Modeling & Ensemble
Classical, ensemble, boosting, and stacking classifiers evaluated under a controlled protocol.
Experience
Combining research, IT infrastructure, and education
Researcher – IoT Intrusion Detection Project
Designed and evaluated feature engineering pipelines for intrusion detection using Python, Scikit-learn, benchmark datasets, and ensemble learning strategies.
IT Administrator
Maintaining servers, client systems, internal IT infrastructure, security monitoring, troubleshooting, technical support, and data management.
Programming and Data Analysis Instructor
Teaching Python, data analysis, computer vision fundamentals, frontend/backend web development, and project-based programming skills.
Technical Skills
Tools and technologies I work with
Programming
Machine Learning & AI
Data Analysis
Domains
Education
Academic background and qualifications
M.Sc. in Computer Engineering – Computer Networks
GPA: 19.43/20 · Top student · Thesis: Optimized Feature Engineering for IoT Intrusion Detection
B.Sc. in Software Engineering
GPA: 17.50/20
Certificates in Data Analytics, Machine Learning, Machine Vision, UI/UX Design, and Cybersecurity. Languages: Persian (Native), English (B2), German (A2).
Courses
From fundamentals to production — practical AI & ML learning paths
Articles
Latest thoughts on AI, ML, and technology
💡 Tips
Quick educational tips on AI, ML, and programming
💬 Forum
Join discussions, ask questions, share knowledge
🎓 Top Students
Our highest achievers
Ready to Dive In?
Join thousands of learners exploring AI, ML, and intelligent systems. Start with a free course.