BS Computer Science Graduate · Faisalabad, Pakistan 🇵🇰

I'm an aspiring machine learning researcher focused on applying data-driven methods to real-world problems, particularly in education and social impact. My core interests lie in predictive modeling, exploratory data analysis, and AI for social good, and I'm currently working toward a master's in research where I can deepen this work.

Languages & Libraries Python NumPy pandas matplotlib seaborn scikit-learn

Core Competencies Data Cleaning EDA Data Visualization Predictive Modeling Classification Regression Clustering Model Evaluation

Tools Jupyter Notebook Visual Studio Code Google Colab GitHub

Binary classification model predicting student pass/fail outcomes

Designed and built a complete machine learning pipeline in Python to predict whether a student will pass or fail, using a synthetically generated dataset of 1,000 student records modeled on realistic behavioral patterns: study hours, attendance, previous scores, sleep hours, and extracurricular participation.

• Conducted full exploratory data analysis, including distribution plots and a feature correlation heatmap
• Compared a Logistic Regression baseline against a Random Forest classifier
• Evaluated performance using accuracy, confusion matrices, and classification reports
• Ran feature importance analysis to identify the strongest predictors of student success
• Achieved 80%+ accuracy on held-out test data
• Built a live prediction function to test outcomes for new student inputs

Tech: Python · pandas · NumPy · matplotlib · seaborn · scikit-learn 📁 View Repository

Segmented 22,000+ customers from real transaction-level retail data

Engineered RFM (Recency, Frequency, Monetary) features from 64,000+ transactions, validating data quality at each stage, then applied K-Means clustering to uncover distinct customer segments. Log-transformed skewed features before scaling to prevent extreme values from dominating the clusters.

Tech: Python · pandas · scikit-learn · matplotlib · seaborn

Exploratory analysis of product, regional, and time-based sales trends

Analyzed e-commerce sales data to identify top-performing products, monthly trends, and category-level breakdowns, and delivered a written summary report of key findings.

Tech: Python · pandas · matplotlib · seaborn

Predicting hiring likelihood for Pakistani university students

Cleaned and analyzed a dataset of roughly 48,000 student records, then built and compared Logistic Regression and Decision Tree classifiers to predict hiring outcomes despite a 96/4 class imbalance. Also developed regression models to estimate expected salary and applied K-Means clustering on academic performance data.

Tech: Python · pandas · scikit-learn · matplotlib · seaborn

Binary sentiment classification on 50,000 movie reviews

Cleaned raw review text, extracted 5,000 TF-IDF features, and compared Logistic Regression against Multinomial Naive Bayes for sentiment classification.

• Achieved 88.85% accuracy with Logistic Regression

Tech: Python · scikit-learn · TF-IDF

🔗 Credly Profile · Microsoft Learn Profile

BS Computer Science — Government College University, Faisalabad (2020 – 2025) GPA: 3.22 / 4.00

Machine Learning · Pattern Recognition · Predictive Modeling · Transfer Learning · AI for Social Good