This repository contains multiple machine learning projects I have worked on. Each project applies data analysis, feature engineering, and ML models to solve real-world problems.
- Goal: Recommend movies/shows similar to a given title or based on user description.
- Approach:
- Data cleaning and preprocessing (
description,genres,cast). - TF-IDF vectorization of text fields.
- Feature fusion with weights.
- Cosine similarity for recommendations.
- Data cleaning and preprocessing (
- Features:
- Recommend by movie title (if present in dataset).
- Recommend by free-text query (e.g., “romantic comedy in New York”).
- Libraries:
pandas,numpy,scikit-learn,scipy,matplotlib,seaborn.
- Goal: Analyze Uber ride data to find trends and insights.
- Tasks:
- Data cleaning and preprocessing.
- Exploratory Data Analysis (EDA): trips by day, hour, and location.
- Visualizations: heatmaps, bar plots, line plots.
- Feature extraction for demand forecasting.
- Libraries:
pandas,numpy,matplotlib,seaborn.
- Goal: Predict the likelihood of a heart attack based on patient data.
- Dataset: Heart disease dataset with attributes such as age, cholesterol, blood pressure, etc.
- Approach:
- Data preprocessing and handling missing values.
- Feature scaling and selection.
- Classification models: Logistic Regression, Decision Tree, Random Forest, SVM.
- Model evaluation with accuracy, precision, recall, confusion matrix.
- Libraries:
pandas,numpy,scikit-learn,matplotlib,seaborn.