Loading presentation...

Present Remotely

Send the link below via email or IM

Copy

Present to your audience

Start remote presentation

  • Invited audience members will follow you as you navigate and present
  • People invited to a presentation do not need a Prezi account
  • This link expires 10 minutes after you close the presentation
  • A maximum of 30 users can follow your presentation
  • Learn more about this feature in our knowledge base article

Do you really want to delete this prezi?

Neither you, nor the coeditors you shared it with will be able to recover it again.

DeleteCancel

DataSlaves BigData Travel Hackings

No description
by

Chandra Jacobs

on 20 October 2013

Comments (0)

Please log in to add your comment.

Report abuse

Transcript of DataSlaves BigData Travel Hackings

DataSlaves BigData Travel Hackings
Technology
Analysis
Query (Amadeus) - Date of Travel: Oct 29
Result (TravelPort) - October Travel (searched in September)
Booking (ARC) - Date of Travel: Oct 29 - Dec 31

Organize by:
Departure Date/Time then
Origin-Destination
Airline (or preferred airline)
Price
Visualization
FlightStats visualization

The problem
I want to book a flight. When should I book it?
Parameters:
Price
# of stops
Origin
Destination
Class of service
Thank you!
Conventional Wisdom
Shop on Tuesday mornings for sales.
Fact or myth?
7 weeks in advance. 11 for int'l.
On flights that are less full.
Never buy within 14 days.
No day of the week is best.

Ideas
Use Travelport data
Determine average price for a route (trip origin / destination) based on days in advance
Use ARC data
Determine average price paid for a route
Determine days to book in advance to determine consumer purchase profiles
Use Amadeus Data
How many days in advance do customers start their search?
Compare above data to visualize number of days in advance and price as a percent above/below average
Break down based on number of stops
Further granularity by drilling down by route
Issues (Technology)
Limited time
Spent significant time/effort learning tools we were not familiar with
Data Model (index) was not initially setup with appropriate data types for statistics
Would this benefit the Travel Industry?
Does the trend of when consumers book their tickets correspond to the best price?
How do major holidays impact this?
Does willingness to book flights with connections change based on days before flight?
Facets allow very fast analysis by terms (popular airports), dates, and perform statistics (average price, etc)
Richmond is popular!
Python
Parsed raw data to JSON
Made minor transformations
Days booked in advance
Primary Cabin
Origin / Destination
Import JSON data to ElasticSearch
Used ElasticSearch
Query data
Use facets for statistical/term analysis
Tableau
Visualize the results
ElasticSearch
Takeaways
The travel industry has lots of data!
Shiny new technologies aren't always the most appropriate!
Conduct more research on tools output prior to using
Python scripting
Tableau can do cool things
ElasticSearch has many capabilities once familiar with DSL
Prezi allows collaborative presentation creating
Travelport data only came from search in a specific period
Ignores seasonal or event differences
Human intervention from Revenue Management
Query Data does not match up with result data
Data (Issues)
Full transcript