PyCon Pune 2017

Ever wondered “How did you find an old school buddy on Facebook, just like that?” , “How Spotify suggests you songs you haven’t listened but are matching with your listening habits?”, “How E-Commerce websites recommends you products you might like?”. Well, the answer to all these question is one-Recommendation Engines. From millions of products existing on a website, it is very difficult for a user to land on a product of his need as well as his choice. But with the inception of Artificial Intelligence, a few algorithms were introduced to perform this difficult task and with the progress in technology, these algorithms were refined as well as customized as per the user requirements by various organizations. In simplest way, Recommendation Engine learns about your behavior with different techniques and recommends you products that seems suitable for you according to the algorithm.

Outline

In this talk I’ll cover the following topics with code in Python:

1. Recommendation Engines

2. Types of recommendation Engines.

3. Various frameworks used for the same purpose (Their comparison with python).

4. Step wise guidance to make an engine- Data Acquisition, Data Cleaning, Implementing various models.

5. Problems faced during developments of recommendation systems.

6. Various libraries of Python used for the same purpose.

7. Metrics used to measure the accuracy of a recommendation engine.

8. Functions that can help to improve your engine for better results as the data gets bigger.

Speaker bio

Kajal Puri is final year student of B.Tech Computer Science. She has done her summer internship at a Startup named The Local Tribe, building a recommendation engine.

Links

• https://github.com/kajalthecoder/Recommendation-Engine

Slides

http://slides.com/kajalpuri/recommendation-engines#/