Sentiment Analysis with Spark Streaming

spam/ham classifier

A simple classifier of music genders, Naive bayes, SGD and SVM used.

Supervised Machine Learning methods (Random Forest and SGD Classifier) to classify short conversations extracted from Reddit

Training binary classifier and multi-class classifier to classify the MNIST datase

The primary aim is to apply image processing techniques to diverse datasets using different machine learning algorithms.

Using MNIST dataset and classifying images

Tag Prediction Model for the Doubt Asking Platform. Suggests tags based on the user input question and question description.

Image Recognition using Python on MNIST dataset with the help of CNN, Multiclass Logistic Regression and SGD

Sentimental analysis on IMDB Movies reviews using Machine Learning classifier (Logistic Regression, Naive Bayes, SVM) and Deep Learning Models (Deep Neural Network, RNN, CNN)

Email Spam Classification with Spark streaming and Predictive Data Modelling

This project was completed as part of the Applied Machine Learning course at Drexel University, Philadelphia. The project aimed to apply machine learning algorithms to analyze a dataset of consumer complaints and categorize them into different groups based on the issues related to goods or services.

Applying predictive models on "Employee Turnover" dataset.

A Python based project used to Predict whether the patient has heart Diseases or not. i have use various classification algorithm for prediction of heart disease

A machine learning system that makes crop recommendations based on soil and weather data.

Improving Performance for Distributed SGD using Ray

Classification of Body postures using different ML algorithms and comparing their performances.

"DressMeUp" project utilizes fashion images and color combinations to achieve image classification for clothing combinations. Algorithms include SGD (SVM), Passive Aggressive Classifier, ResNet50 CNN, and EfficientNetV2-S CNN with K-Means for color analysis. Achieved accuracy exceeds 90%. Built with Python, Scikit-Learn, TensorFlow, and Streamlit.

We built a model for Film Junky Union to detect negative reviews.