/data_dump
============
Folder contains the data required for the existing and the new algorithm to function. This data is generated by executing the data generation code inside ./data-creation

/data-creation
===============
1) data_creation.py : creates possible label set combinations and randomly assigns cost to the label set. The Cost range is controlled based on the number of attributes in each label set.
2) user_data_creation.py: Generates number of users as specified in the input. Randomly assigns characteristics to users from the domain of available characteristics.

budget-optimization-existing.py
================================
1) Implementation of OSN Perspective approximation algorithm in "C. Xia, S. Guha and S. Muthukrishnan. Targeting Algorithms for Online Social Advertising Markets. In 2016 IEEE/ACM International Conference on ASONAM"

budget-optimization-new.py
1) Implementation of Alternate Greedy Approximation algorithm for the OSN Perspective problem in "C. Xia, S. Guha and S. Muthukrishnan. Targeting Algorithms for Online Social Advertising Markets. In 2016 IEEE/ACM International Conference on ASONAM"