Software Developer focused on Data Mining and Machine Learning Applications
Welcome to my GitHub Pages portfolio! Here you will find detailed information about my qualifications, professional experience, education, technical skills, and publications.
Table of Contents
Summary of Qualifications
- Bilingual, with over 4 years of intensive research experience in Data Mining and Machine Learning.
- Published high-quality research papers in Data Mining and Machine Learning.
- Specialized in developing innovative Deep Learning models for OpenStreetMap data quality assessment.
- Excellent troubleshooting and network management skills.
- Strong collaboration skills, thriving both as a team member and working independently.
- Excellent communication, critical thinking, and decision-making skills.
- Fluent in English and French.
Professional Experience
Postdoctoral Researcher @ Space Techniques Center, Algerian Space Agency, Algeria (Jan 2022 - Sep 2024)
- Spearheaded the design and construction of Deep Learning models, taking ownership of initiatives to assess and enhance OpenStreetMap (OSM) data quality (3-person team).
- Pioneered the design and training of a bespoke segmentation model on Alsat-2 satellite images for land cover classification, delivering high-precision results (5-person team).
- Authored a comprehensive literature review on cutting-edge models for OSM data quality assessment, providing actionable insights.
- Presented a research paper at the flagship conference GIScience 2023, demonstrating the impact of the work on a global stage.
Computer Science Engineer @ EHS LES PINS, Algeria (May 2014 - Jun 2016)
- Developed Java applications for remote machine monitoring and communication between radiologists and technicians, achieving a 70% reduction in radio film usage and significantly enhancing radiology service quality.
- Maintained the hospital information system, ensuring data integrity and improving the system’s performance and reliability.
- Managed and extended the hospital’s local area network, ensuring robust and reliable connectivity.
Computer Science Instructor @ Institute of Management and Enterprise Development of Oran, Algeria (Jan 2014 - Mar 2014)
- Delivered comprehensive courses in Computer Science, Office Automation, and Java programming to a group of 15 students, ensuring a high level of understanding.
- Supervised and evaluated student assignments and exams, providing constructive feedback to promote learning and improvement.
Part-time Teaching Assistant @ University of Science and Technology of Oran Mohamed Boudiaf, Algeria (Jan 2012 - June 2012)
- Instructed courses in Visual Basic Application for Excel and Matlab programming to a group of 20 students, fostering a strong foundation in programming skills.
- Guided students through complex programming concepts, enhancing their problem-solving abilities and technical knowledge.
Education
D.Eng. in Software Engineering (软件工程博士) | School of Computer Science (计算机科) | Wuhan University (武汉大学), China (Dec 2021)
- Conducted research on quasi-convex optimization for gradient-based clustering algorithms, resulting in improving the performance of multi-database clustering.
- Received a fully funded scholarship to pursue a Doctor of Engineering degree in China.
- Published 2 high-quality journal papers and a China Computer Federation conference paper.
- Degree evaluated by WES as equivalent to a doctoral degree obtained in Canada.
- Defended a thesis on Local Pattern Analysis strategy using graph-based classification algorithms and disjoint forests-based data structures resulting in enhancing the quality of multi-database mining and the decision-making processes.
- Received Best Paper Presentation award for exceptional participation at the COSI conference.
- Published 3 high-quality journal and conference papers.
Technical Skills
Operating Systems
Programming Languages
Data Science & Machine Learning
Web Development
Database Management Systems
Version Control
Geospatial Analysis
Big Data
Data Visualization
Dimensionality Reduction
Journal Articles and Conferences
A Gradient-Based Clustering for Multi-Database Mining
Publication
Multinational corporations have multiple databases distributed throughout their branches, which store millions of transactions per day. For business applications, identifying disjoint clusters of similar and relevant databases contributes to learning the common buying patterns among customers and also increases the profits by targeting potential clients in the future. This process is called clustering, which is an important unsupervised technique for big data mining. In this article, we present an effective approach to search for the optimal clustering of multiple transaction databases in a weighted undirected similarity graph. To assess the clustering quality, we use dual gradient descent to minimize a constrained quasi-convex loss function whose parameters will determine the edges needed to form the optimal database clusters in the graph. Therefore, finding the global minimum is guaranteed in a finite and short time compared with the existing non-convex objectives where all possible candidate clusterings are generated to find the ideal clustering. Moreover, our algorithm does not require specifying the number of clusters a priori and uses a disjoint-set forest data structure to maintain and keep track of the clusters as they are updated. Through a series of experiments on public data samples and precomputed similarity matrices, we show that our algorithm is more accurate and faster in practice than the existing clustering algorithms for multi-database mining.
Training and Certifications