Permalink
Find file
Fetching contributors…
Cannot retrieve contributors at this time
executable file 152 lines (108 sloc) 5.91 KB

PennApps 2011 Hackstravaganza

5 live demos in 5 minutes, powered by the Hunch API. 5x the danger, 10x the network bandwidth...

TL;DR Check out the Hunch API reference.

Demos

Fast Forward

  • Discover New Music for Old-Schoolers

  • API calls used

Popcorn

  • Adds Hunch recommendations to Google's Movie Showtimes pages

  • API calls used

  • Protip: result_ids can be from:

    • Amazon (amz_)
    • Facebook (fb_)
    • Foursquare (4sq_)
    • Hunch (hn_)
    • IMDB (imdb_)
    • Musicbrainz (mb_)
    • Spotify (sp_)
    • TripAdvisor (tp_)
    • Twitter (tw_)
    • Yelp (yelp_)

Restaurantology

  • I'm homesick for that restaurant I used to love back in High School but I live in Juneau, Alaska.

  • API calls used

Brdg.me

  • Find awesome people nearby

  • API calls used

AutomaticDJ

BONUS: Hunch.fm

BONUS: Spotify Playlist Generator

BONUS: Hunch Movie Map

  • Find what movies connect you and your friends

  • (Log in with your Hunch.com username and password)

  • API calls used

    • batch (allows you to execute multiple API request in one call. very snazzy)

          http://api.hunch.com/api/v1/batch/?auth_basic=1&calls=
      
      
        	  [
        		{"name":"get-recommendations",
        		 "params":
        			{"friend_id":"tw_gleitz",
        			 "limit":"10",
        			 "topic_ids":"list_movie"}},
        		{"name":"get-recommendations",
        		 "params":
        			{"_result_ids":"responses[0].recommendations.result_id",
        			 "minimal":true,
        			 "limit":10,
        			 "popularity":".1",
        			 "friend_id":"tw_pennapps"},
        		 "requires":["responses[0].recommendations.result_id"]}
        	  ]
      

What Should I Build??

  • Group recommendations

    • I'm going to dinner with Wesley Snipes, Kenny G., and Staff Sergeant Max Fightmaster. Where should we go? Check out the group recommendations call.
  • A browser extension that takes the current page you are on (Amazon, Facebook, Yelp) and shows your affinity to that page.

  • Generate a Tumblr blog on the fly that pulls down interesting Hunch results. People could follow it and the recommendations would evolve based on the followers.

  • A Foursquare hack that pulls down all the places you've been and recommends similar spots, similar items, or similar people.

  • A "stereotype generator" that shows things that people like that frequent particular Foursquare places (Cufflinks and golf clubs for the Four Seasons, Birkenstocks and anti-nuclear proliferation for Whole Foods).

  • A turntable.fm mashup that recommends rooms based on your similarity to people in the room.

  • A clone of EchoFi, powered by Hunch.

Links Worth Reading

Good luck! Go out there and crunch some data!