Skip to content

Personalizing unique travel experiences using data science.

Notifications You must be signed in to change notification settings

iglee/outrunJulesVerne

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

86 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Outrun Jules Verne!

This project aims to beat the best adventurer I know with the best name: Jules Verne. His book "Around the World in 80 Days" chronicles a fictional around-the-world trip on a bet that Phileas Fogg and Passepartout complete it in 80 days. Let's beat them at their game!

Jules Verne's Around the World in 80 Days

I'll help come up with an around-the-world trip path based on your selected features like sentiment and interests; you just be an adventurer and find something to bet on.

Why?

First and foremost, it's fun! Travel recommendations are somewhat of a dated art; travel agencies have largely been replaced by Google in this interwebbed and globally-minded world. However, personalizing travels is still a great interest for the most individualistic modern population with the most autonomy the history has ever seen. This is my attempt at tailoring unique travels and experiences for you.

What kind of data did I use?

  • reviews from tripadvisor.com
    • At first, I requested access to their API, but they never got back to me. So, I relied on webscraping to build my dataset from scratch. The resulting 1M+ customer reviews of attractions around the globe are stored in a private AWS S3 bucket.
  • GIS data of local attractions, i.e. the exact geographical locations of each attraction.

How did I get the data?

  • Webscraped tripadvisor.com using request and BeautifulSoup
  • GIS data: publicly available

How can I recommend a path around globe for you?

  • Natural Language Processing to extract features like sentiment, local attractions, characteristics, etc. Namely, I used spark's LDA.
  • Monte Carlo Metropolis algorithm to add some randomness.

For detailed implementation, take a look at README.md in src folder.

How well did I do?

  • Show friends and others, and ask them to evaluate. (I still have the survey scores, and it was around 4 / out of 5!)

Show & tell

The project used to be deployed as a webapp: http://outrunjulesverne.com/. However, the URL is no longer active. For how it used to look, please view this presentation. But as a quick look, here's an example of selected features:

Feature selected

And one example of an output path around the world is: Your recommended path around the world

And along the way, you'll see these beauties of the world: beauties around the world

This app may come back in the near future :]. I really enjoyed this project, and hope you did too. Welp, at least maybe inspire a grand trip.

Useful Information