Anomaly Detection Techniques as a Quality Evaluation of graphs

Abstract

The goal of this project is the implementation of PyGQE, a software package that given a graph measures its quality by measuring the possible anomaly detections. The aim of this application is to help data scientists evaluate how important a dataset in graph form is and its level of quality. The program is implemented in python, it takes a list of edges in CSV format and a feature map (optional) and returns a list of anomalous nodes and uncommon features patterns.