🚀 Why I’m Sharing My Journey
I’m currently an MSc Data Science student at the University of Greenwich, and My area of interest is, but not limited to, Natural Language Processing (NLP).
I’m sharing this overview of my skills and experience here on Medium for two reasons:
First, to connect with individuals and teams working on cutting-edge ML models. Second, to document my path toward becoming a specialist in this field, using Medium as a place to share code and learning insights.
I am actively seeking a mentorship opportunity to accelerate my growth and apply advanced skills like LSTMs/GRUs and Transformers in a dynamic organizational setting. If you’re building in the NLP or Data Science space, I would love to connect.
💡 Core Technical Expertise
My background combines seven months of practical Data Analyst experience with focused academic training in advanced modeling and statistics.
Deep Learning Models: RNNs, LSTMs, GRUs, Transformers , FastText, TensorFlow/Keras.
Core Concepts: Natural Language Processing (NLP), Machine Learning, Statistical Analysis.
Languages & Visualization: Python (Pandas, Scikit-learn), SQL, Power BI, Excel.
💻 Key Accomplishments & Projects
1. Data Analyst: ParamInfo (Nov 2023 — June 2024).
As a Data Analyst, my focus was on driving measurable improvements through clean data and clear reporting.
- Performance Improvement: Led statistical analysis using Python on website and social media metrics, directly resulting in a 25% improvement in website performance and a 50% increase in social media engagement.
- Data Pipeline Management: Managed end-to-end data pipelines, performing EDA and data manipulation with Python (Pandas) and SQL on large, diverse datasets (e.g., Google Analytics).
- Visualization: Implemented highly impactful Power BI dashboards and visualizations that increased report clarity and quality by 30%, effectively translating business needs into visual insights for board members.
2. MSc Project: Hierarchical Product Classification.
This project was a deep dive into advanced NLP models and multi-level classification.
- Model Implementation: Developed a multi-level product categorization system using both FastText (for initial levels) and a Bidirectional LSTM model.
- Data Preparation: Preprocessed and cleaned over 8300+ product data entries, including handling hundreds of null and duplicate values to ensure model fidelity.
- Results: The FastText models achieved precision and recall of 90–95% across all levels. The final LSTM model for 280 unique Level 3 categories achieved a test accuracy of over 91%.
🧠 Critical Skills & Education
- Education: MSc in Data Science (University of Greenwich) focusing on Applied Machine Learning, Big Data, Data Visualisation and Statistics.
- Critical Thinking & Resilience: Adept at evaluating complex issues and maintaining composure under pressure, a skill honed through technical problem-solving and numerous volunteering initiatives.
✉️ Let’s Connect
Thank you for reviewing my experience. I am enthusiastic about the future of deep learning and eager to contribute to a challenging team environment.
