Skip to content

The easiest way to optimize Facebook Ads using Upper Confidence Bound Algorithm. 💻

Notifications You must be signed in to change notification settings

vaibhavmagon/FB-Ads-Opt-UCB

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

21 Commits
 
 
 
 
 
 

Repository files navigation

Facebook Ads Optimisation Script Using Real Life Ads 💻

This code analyzes 43 Facebook ad sets and helps find the best ad set using the Upper Confidence Bound machine learning algorithm.

More on UCB here: http://banditalgs.com/2016/09/18/the-upper-confidence-bound-algorithm/

To Run

  1. Export Ad set from Facebook ads directly for Impressions : Signups (Can use it for other data points as well).
  2. Replace the data in ads-data.csv with current ads data.
  3. Run: sudo python fb-ucb-ads.py

Data

Folder contains the ads-data.csv file with Ads Impressions : Signups data.

Ads Histogram for current Data

Clearly Ad: 2 is the winner and should have higher budget allocation.

screen shot 2019-01-31 at 11 46 28 am

Prerequisits

  • Python 2.7
  • Machine learning libraries
  • Anaconda/Spyder

Contributers

  • Vaibhav Magon