MER is a software that identifies and highlights manipulative communication in text from human conversations and AI-generated responses. MER benchmarks language models for manipulative expressions, fostering development of transparency and safety in AI. It also supports manipulation victims by detecting manipulative patterns in human communication.

Scanner, signatures and the largest collection of Magento malware

This research goal is to build binary classifier model which are able to separate fraud transactions from non-fraud transactions.

This GitHub repository provides a comprehensive set of tools and algorithms for detecting fraud anomalies in various data sources. Fraudulent activities can have severe consequences, impacting businesses and individuals alike. With this repository, we aim to empower researchers with effective techniques to identify and prevent fraudulent behavior.

Unsupervised learning application to fraud detection

Fraud Detection Research - Data Science Capstone Project at Penn State University, University Park Campus

The visual graph of fraud detection website

This repository contains the code components of work carried out for analyzing the Medical Provider Fraud Detection dataset with the intent to find most important features to crack down the potentially fraud providers.

Data preprocessing and classification for the detection of fraudulent transactions

Using R Language to predict whether a user will download an app after clicking a mobile app advertisement. Click on the link below to see more details!

Team project for BA810 (Supervised Machine Learning)

A collection of projects where I worked on building anomaly detection pipelines. This rep covers code for EDA, outlier detection, and stock analysis.

A book project accompanying the CopyDetect package. The book provides comprehensive coverage of response similarity analysis using R.

Test repo for the Smartboard project

Fraud Detection of a 6 million row dataset using AWS and Spark

Detecting fraud on online customer transactions

The goal of the competition was to predict fraudulent transactions on a dataset with about 40 million instances, with some characteristics similar to the datasets processed by Feedzai.

🔍 | 📈 | Life / Health Insurance Fraud Detection | 📋 | (Codeshahstra Round 1 Hackathon)

This solution performs Anomaly Detection with Statistical Modeling on Spark. The detection is based on Z-Score calculated on cpu usage data collected from servers.

Report on the performance of different machine learning algorithms in identifying persons of interest in the Enron Fraud Case