Abdul Rehman

Results-driven data analyst with expertise in Python, SQL, and advanced Excel. Skilled in predictive modeling, data visualization, and leveraging insights to drive business growth. Proven track record in enhancing data accuracy, improving forecasting models, and extracting actionable insights from complex datasets. Adept at transforming data into strategic recommendations and collaborating with teams to boost efficiency and decision-making.

• Microsoft Excel - ADVANCED Certificate | Reed.co.uk
• Google Data Analytics Certificate | Google
• SQL for Data Science | University of California, Davis
• Python for Data Science, AI, and Development | IBM
• Python Programming | University of Helsinki
• Google Advanced Data Analytics | Google
• Supervised Machine Learning: Regression and Classification | Stanford University
• Python programming CS50P | Harvard University
• Advanced Learning Algorithms | Stanford University
• Introduction to Computer Science CS50X | Harvard University

• Customer Churn Analysis - EDA and Logistics Regression.

Conducted a comprehensive churn analysis for a telecommunications company using Python (Pandas, NumPy, Matplotlib), employing univariate and bivariate analysis to identify key customer characteristics, and developed a high-accuracy predictive neural network model (Tensorflow, Keras) to provide strategic recommendations based on findings.

• Bitcoin Price Movement - Sentiment Analysis & Xgboost.

Extracted Bitcoin data using Python (Mwclient, Yfinance) and conducted predictive analysis with XBoost, achieving 70% accuracy in forecasting price movements through feature engineering.

• E-Commerce Marketplace Analysis - Using Web Scraping to Identify Profitable Product Opportunities.

Scraped and analyzed product trends on Daraz(Pakistan's largest e-commerce platform) using Python (BeautifulSoup, Requests) to provide actionable insights for product selection, aiding in strategic decision-making.

• Traffic Sign Recognition System - using Convoluted Neural Networks.

Developed a traffic sign recognition system using CNNs in Python (Keras, TensorFlow), achieving 96% accuracy and surpassing standard neural network performance.

• Customer Segmentation Using K-Means.

Performed K-means clustering on customer data using Python (Pandas, Scikit-learn, Matplotlib) to identify distinct segments, analyze behavior patterns, and develop targeted marketing strategies, enhancing customer retention and acquisition efforts.

• Used Car Price Prediction - Polynomial Regression.

Developed a polynomial regression model to predict used car prices with over 92% accuracy using Python (Pandas, NumPy, Scikit-learn) for data pre-processing and model training, implemented feature engineering techniques, and provided actionable insights for pricing strategies and market analysis.

• Steel Plate Defect Detection - Random Forest Classifier.

Used a Random Forest Classifier with Python (Pandas, Scikit-learn) for data preprocessing, model training, and evaluation to detect defective steel plates with 88% accuracy, enhancing model accuracy through feature selection and engineering, and delivering insights to improve quality control processes in manufacturing.