MUFTI JAZIM

Final-year B.Tech Computer Science student at the University of Kashmir with hands-on experience in Artificial Intelligence, Machine Learning, and Data Analytics. Skilled in building end-to-end data solutions, from preprocessing and feature engineering to model training and deployment with FastAPI. Proficient in predictive analytics and advanced data visualization using Power BI, with expertise in integrating machine learning workflows with business intelligence tools for data-driven decision-making. Seeking an internship in Data Science or Machine Learning to apply technical expertise and problem-solving skills in delivering impactful business solutions.

• AI/ML workshop, Workshop - NIT, Srinagar.
• Machine Learning Specialization - Coursera.
• Machine Learning A-Z: Hands-On Python in Data Science. - Udemy.
• Comprehensive course on Python, SQL, Statistics, AI/ML - Techboxx, Srinagar.
• Power BI workshop - Skill Nation.

- Obtained 94% in 10th grade.

- Obtained 94.8% in 12th grade.

- Secured 82% in the Joint Entrance Examination (JEE).

• Credit Card Fraud Detection – Developed a fraud detection model using Logistic Regression, achieving 97% accuracy on a dataset of 280,000 transactions.

• Boston Housing Price Prediction – Performed data preprocessing and feature engineering, followed by EDA and predictive modeling using Linear Regression, achieving an R² score of nearly 0.8

• Review Sentiment Prediction – Implemented NLP and vectorization techniques to classify customer reviews, achieving 81% accuracy.

• Fake Hotel Review Prediction – Built an SVM model for detecting deceptive hotel reviews with 92% precision and deployed the model as an API using FastAPI for realtime predictions.

• Mall Customer Segmentation – Applied K-Means clustering to segment 200+ customers by purchasing behavior, generating 4 actionable customer clusters, and visualized insights through a Power BI dashboard.

• Cat-Dog Image Classification (CNN) – Trained a convolutional neural network with TensorFlow to classify images, achieving 95% test accuracy on a dataset of 10,000 images.

• Bank Churn Prediction (ANN) – Designed an artificial neural network achieving 76% recall in predicting customer churn.