Project Summary: Travel Scout

  • Project title
    • Travel Scout
  • Accessible at www.inventacorp.com/TravelScout
  • Project members (names and emails)
  • Motivating Questions
    • Historically, the recreational travel market was intermediated by professional travel agents who could suggest destinations to vacationers based on important factors such as the time of year, the traveler's budget, languages spoken, and the distance the traveler wanted to fly.

      In the contemporary travel market, travel agents have become a luxury service reserved for the technologically incompetent and those demanding additional services like aggregated invoices (i.e. corporate travelers).  However, the important advising services previously rendered for casual vacationers is still in demand because most travel websites (e.g. Expedia, Orbitz) only offer point-to-point deal searching.

      This context suggests our motivating question: "If I am a casual traveler looking to get out of town, where can I go, and what's the best value?"

      This question is not answered by the profuse number of point-to-point airfare search engines because it requires the user to exhaustively search a large number of possible destinations and to conduct offline research into those locations.  Only recently, a small number of online services have cropped up that allow for more of a "browsing" experience than the "searching" one available since the '90s.  However, none of these takes full advantage of users' visual comprehension abilities.
  • Data Sources
    • The source of airfare data is currently SkyScanner (http://www.skyscanner.net)  We are lucky that this website allows "searching" at varying levels of granularity (both the "from" and "to" locations can be countries, cities, or airports, and the "to" location is entirely optional).
    • The second set of data is our weather data, which informs the user's temperature filter.  We collected historical average temperatures for every city and every month of the year.  All of this data was scraped from www.WeatherBase.com using Python.
    • The third set of data is the geocoding of locations (e.g. "Boston Logan") to latitude/longitude coordinates so that we can plot the user's itineraries on a map.  We use a combination of The Global Airport Database and the Google Maps API geocoder.
    • In the future, we may incorporate "languages spoken" data, which is currently in table form here and here.
  • Design Details
    • We have "details on demand" provided by Google Map's zooming capabilities and the aggregation features of the tree list.
    • Temperatures are encoded on a blue to red scale.
    • Flight distances (correlated with durations) are encoded as the length of a line between airports.
    • Itineraries are encoded with different colors.  In the future, we hope to match the color displayed on the Google map with a coloring of the itinerary box.
    • Aggregated vs. exact travel destinations are displayed with different markers.
    • Country flags represent countries in the Travel Options box.
    • Double-ended sliders allow different types of filtering based on temperature and budget.
  • Screenshot
[top]