Skip to content

Latest commit

 

History

History
33 lines (21 loc) · 1.81 KB

README.md

File metadata and controls

33 lines (21 loc) · 1.81 KB
Raw

Table of Contents

  1. Introduction
  2. Project Summary
  3. Dependencies
  4. Running instructions

Introduction

Imagine you're at an e-commerce company that also has a social network. In addition to shopping, users can see which items their friends are buying, and are influenced by the purchases within their network. In this project a real-time platform is built to analyze purchases within a social network of users, and detect any behavior that is far from the average within that social network.

Project Summary

This project is written in Python 2.7 and contains a main.py source file and two supporting module parse_input.py and tools.py.

Processing of stream_log.json is done in the main.py file.

The functions that are used to parse batch_log.json are placed in parse_input.py module.

Additional functions that are used to find D-th degree friends of each user and functions to process purchase history are placed in tools.py module.

Dependencies

The following python libraries are used in this project:

sys, os, json, collections, numpy, networkx

Running instructions

The project can be executed with the following command:

python main batch_log.json stream_log.json flagged_purchases.json

as can be seen from the above, main.py takes three arguments from the command line. The first argument is the batch_log.json filename. This file contains the history of the social network events. The second argument is the stream_log.json filename. This file includes data that come from the real-time network stream. The code processes these two files and writes the flagged anomalous purchases in the file that is specified in the third argument flagged_purchases.json.