[MICCAI2019 & TMI2020] Overfitting under Class Imbalance: Anaylsis and Improvements for Medical Image Segmentation.

Dropout in Deep Learning

Classification of signatures in image format as genuine or fake. Created two models - one from scratch using deep learning layers and other using pre trained model VGG16. Before training used image pre processing techniques as well.

This project demonstrates the use of multi-class SVM on the Adult Census Income dataset from the UCI Machine Learning Repository. T

This project explores the working of various Boosting algorithms and analyzes the results across different algorithms. Algorithms Used are: Random Forest, Ada Boost, Gradient Boost and XG Boost

Intro to Machine Learning Course By Kaggle

This repository explores how data augmentation helps mitigate overfitting in CNNs with limited training data.

Reducing overfitting in perdiction in decision trees

Bayesian PRS methods model uncertainty in effect size estimates and shrink small effect sizes to mitigate spurious associations and biases from sample overlap. By using full posterior distributions rather than point estimates, they effectively account for estimation errors and reduce the impact of artificially inflated associations.

Data Analysis and Visualization in the US Health Insurance industry - UEH

A Performance Study of Naive Bayes Classifier in Advertisement Analysis

Regularization is a crucial technique in machine learning that helps to prevent overfitting. Overfitting occurs when a model becomes too complex and learns the training data so well that it fails to generalize to new, unseen data.

This is the dataset used in the second chapter of Aurélien Géron's recent book 'Hands-On Machine learning with Scikit-Learn and TensorFlow'. It serves as an excellent introduction to implementing machine learning algorithms because it requires rudimentary data cleaning, has an easily understandable list of variables and sits at an optimal size b…

The model uses CNNs to guess the flower in the image. At each epoch, the model's neurons undergo a random dropout and the data is augmented. Overfitting is eliminated. The dataset can be downloaded storage.googleapis.com/download/example_images/flower_photos.tgz.

The primary objective of this project is to design and train a deep neural network that can generalize well to new, unseen data, effectively distinguishing between rocks and metal cylinders based on the sonar chirp returns.

Health Profile Analysis:Revealing Disorder Paterns,Medication Guidance and Risk Classification-ML Project