-
Campus Placement Prediction
This project leverages data mining to analyze student profiles and predict placement outcomes using machine learning models like Support Vector Machines,KNN, Logistic Regression and Decision Trees. Feature engineering plays a critical role in enhancing model accuracy. By analyzing academic and demographic data, the project aims to forecast employability, providing valuable insights for educational institutions and students alike. -
Churn Prediction
Focused on customer retention, this project models customer churn based on a rich dataset including demographics and account behavior, applying ensemble methods like Gradient Boosting and logistic regression to user data, segmented by demographics and interaction metrics. Applied classification algorithms and evaluating model performance using confusion matrix and ROC curves for precision in prediction. It quantifies churn risk and identifies key factors influencing customer retention for strategic business interventions. -
FloraClassify
A botanical classification system that employs Support Vector Machine (SVM) algorithm to distinguish species based on morphological data, with a focus on hyperparameter tuning for model optimization and cross-validation to ensure generalizability. -
Studying the Correlation between Tweet Sentiment and Stock Prices
A computational finance project that combines sentiment analysis using NLP techniques with quantitative stock data. It employs time series analysis and machine learning to evaluate the predictive power of public sentiment on market movements. -
Violence Detection Using Deep Learning
An advanced computer vision project that uses Convolutional Neural Networks (CNNs) to detect violent actions in video data. The project entails rigorous training with a labeled dataset to fine-tune the neural network for high accuracy and real-time detection capabilities. It integrates image processing techniques with deep learning frameworks to identify violent events, contributing to the field of intelligent surveillance systems.
Each project contains all the neccessary scripts, model files, and documentation detailing the methodology, algorithms used, and the results obtained from the machine learning models.