This project uses Python 2. I have used the Anaconda distribution of python for development. It includes libraries numpy, scipy, pandas, sklearn, matplotlib and seaborn used in this project. Apart from these libraries, I have used the igraph library for the graph operations. All of the mentioned libraries can be installed using 'pip install library_name' or 'conda install library_name' if you have the Anaconda distrinution. The SNAP (Stanford Network Analysis Project) Facebook dataset is used in this project. Refer to the documentation in the 'report' folder for implementation details.
Sourish1997/anomaly-detection
A statistical and machine learning based approach to anomaly detection in social networks.
Jupyter Notebook