Course: CS421-Machine Learning

Project: Product rating fraud/anomaly detection

Overview

The task is to build and optimize a model that can correctly identify fraudulent users based on their ratings for various items.

Data are published as batches on a weekly basis, and labels for previous weeks are made available after submissions and metrics have been computed.

It assumed that real and fradulent user reviews are drawn from different probability distributions.

Training datasets are located within the /batches directory, and output predictions are located within the /predictions directory

Getting started

• Install all dependencies with pip install -r requirements.txt

• Run the notebook!