Fahima Jannat - Data Science Portfolio
Dear Sir! I’m Fahima Jannat, a passionate Data Scientist with a strong foundation in statistics, machine learning, and data storytelling. I love turning raw data into actionable insights and building models that solve real-world problems.
🧠 Skills & Technologies
- Languages: Python, SQL, R
- Libraries: Pandas, NumPy, Scikit-learn, TensorFlow, Keras, XGBoost, Matplotlib, Seaborn
- Databases: MySQL, PostgreSQL, MongoDB
- Tools: Jupyter Notebook, Tableau, Power BI, Git, Docker, AWS (S3, EC2)
- Concepts: Regression, Classification, NLP, Time Series Forecasting, A/B Testing, Clustering
📁 Projects
1. 📊 Customer Churn Prediction
Goal: Identify customers likely to churn for a telecom company.
Tech Stack: Python, Scikit-learn, Logistic Regression, Random Forest
- Cleaned and explored 6K+ customer records
- Trained multiple ML models with GridSearchCV
- Achieved 89% accuracy with optimized random forest
- Deployed using Streamlit for demo purposes
🔗 View Project
2. 🧾 Retail Sales Forecasting
Goal: Forecast future sales using time series data from a chain of stores.
Tech Stack: Python, ARIMA, Facebook Prophet
- Implemented time series models with proper decomposition
- Compared Prophet vs. ARIMA performance
- Improved forecast accuracy by 15% using seasonal tuning
🔗 View Project
3. 💬 Sentiment Analysis on Product Reviews
Goal: Classify customer reviews as positive, negative, or neutral using NLP.
Tech Stack: Python, NLTK, TF-IDF, XGBoost
- Preprocessed 50K+ product reviews
- Built text classification models using pipeline
- Achieved 92% F1-score on test data
🔗 View Project
4. 📈 Airbnb Data Dashboard (Tableau)
Goal: Create an interactive dashboard to explore Airbnb trends in NYC.
Tech Stack: Tableau
🏅 Certifications
- IBM Data Science Professional Certificate (Coursera)
- Google Advanced Data Analytics Certificate
- Tableau Data Visualization Specialist
📬 Let’s Connect!
“Data is a precious thing and will last longer than the systems themselves.” – Tim Berners-Lee