Resampling exercise to predict accuracy, precision, and sensitivity in credit-loan risk

Compared the metrics and performance of different classification algorithms on Heart Failure dataset from UCI ML Repository

Evaluate a detection model performance

Human Resources Analytics

Amex Analyze This is a data science competition held by American Express across all the Indian Institute of Technology Institutes across India. I had participated in this competition in 2018, it was based on predictive modelling where we need to train a model to solve a bank problem - Analyze This 2018

Trained MATLAB models for 82% precision/80% recall, optimized with blob analysis for 25% performance boost. User-friendly alarm system with 500+ engaged users.

Developed a Convolutional Neural Network based on VGG16 architecture to diagnose COVID-19 and classify chest X-rays of patients suffering from COVID-19, Ground Glass Opacity and Viral Pneumonia. This repository contains the link to the dataset, python code for visualizing the obtained data and developing the model using Keras API.

Insurance Cross Sell Opportunity Forecast through machine learning algorithm

Mail SPAM Detector

BGU, Information Retrieval final project. Search-engine, Wikipedia corpus.

ML-FinFraud-Detector is a machine learning project for detecting financial transaction fraud. Utilizing XGBoost, precision-recall, and ROC curves, it provides accurate fraud detection. Explore feature importance, evaluate model performance, and enhance financial security with this comprehensive fraud detection solution.

The goal of this project is to develop a machine learning model that can help banks to identify customers who are likely to churn and take appropriate measures to retain them

A Swift implementation of mAP computation for Yolo-style detections

Information Retrieval models implemented in Python

(projeto ainda não finalizado) - Este repositório contém um projeto de uma seguradora deseja começar a vender seguro de veículos para clientes que já possuem plano de saúde.

Built a simple search system using Lucene. Indexed 100 text documents using the bbc-news sports dataset. Showed the impact of indexing the data well on precision and recall. Have included the queries used to arrive at the precision and recall.

📔 This repository delves into Logistic Regression for loan approval prediction at LoanTap. It covers data preprocessing, model development, evaluation metrics, and strategic business recommendations. Explore model optimization techniques such as confusion matrix, precision, recall, Roc curve and F1 score to effectively mitigate default risks.

Resample precision-recall curves correctly!

Supervised Machine Learning and Credit Risk