Description:

We have always been taught that the earlier you book a flight, the cheaper it is. What if I said it isn't? You see it's not a straight line and it has a minimum at some point(someday before the flight). We are going to see how historical Airfare data can help us derive the best day to book a flight so that you 'actually' get the cheapest fares.

The talk would talk about the entire process, from getting the data, to training a basic Neural network on the data. With advancements in deep learning in these few years, it is very easy to train a simple statistical model to predict the prices.

Also, my thesis at IIT Kharagpur was titled 'Forecasting of Airfare prices using Neural networks' and the talk is based on that along with a few improvements I made on top of that.

Prerequisites:

A brief understanding of neural networks or any machine learning model in general could help you make the most out of your talk.

Content URLs:

To be updated soon.

Speaker Info:

I am an IIT Kharagpur graduate(2017) who spent over 4 years coding in Python. Worked with all styles of python from website development using Django and Flask to scientific computing using numpy and scikit-learn to web-scraping using Selenium. It's been a wonderful journey all along and I'm now looking forward to bring as many people on board as I can to experience what I've experienced.

I am also the founder of Papercop, an examination preparation portal for the students of IIT Kharagpur which has about 70k+ hits. I am a very passionate speedcuber( Can solve the rubiks cube in about 10s odd). Won plenty of medals in speedcubing competitions across the country. I now work as an analyst with American Express.

Speaker at Pycon India '17 and invited to Pycon Italy'18

Speaker Links:

Links to previous talks: Pycon India'17

Twitter

Linkedin