R package and app that got bourne out of the ideas from the "Managing Crowds" hackathon at the Gemeente Amsterdam. This is basically a scraper for events happening around Amsterdam with extended functionality for:
- scraping of data from various sources ()
- cleaning up and enrichment of data
- persistent storage in local database
- some analysis tools as well as a jab at a predictive model
- visualisation tools
- Sample simfuny data
title | nid | img | location | lat | long | secret | attending | event_link | fb_link | fb_going | fb_interested | date_wday | date2_wday | date_conv | date2_conv | sold_out | timestamp | dist_rembrandtplein | near_rembrandtplein |
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
Solstice Festival 2017 | 66989 | http://api.simfuny.com/sites/default/files/photo/events/15895656_1272531012786388_5048735231400931615_o_0.jpg | Ruigoord 76 | 52.44437 | 4.828969 | 0 | 1706 | http://simfuny.com/event/66989/solstice-festival-2017 | https://www.facebook.com/events/604130049795847 | 1600 | 6200 | Sat | Sun | 2017-06-24 | 2017-06-25 | FALSE | 2017-06-25 01:03:48 | 9468.8851 | FALSE |
Official After Party Awakenings Festival - Saturday (SOLD OUT) | 111393 | http://api.simfuny.com/sites/default/files/photo/events/19113915_1252454411543704_9161799471844597684_n_0.jpg | Paradiso Amsterdam | 52.36228 | 4.883950 | 0 | 591 | http://simfuny.com/event/111393/official-after-party-awakenings-festival-saturday-sold-out | https://www.facebook.com/events/1493465287363922 | 631 | 894 | Sat | Sun | 2017-06-24 | 2017-06-25 | TRUE | 2017-06-25 01:03:48 | 546.4308 | FALSE |
Last minute: Vunzige Deuntjes At The Beach - Whoosah | 113285 | http://api.simfuny.com/sites/default/files/photo/events/19247802_1385427731576751_8025315936222131220_n.jpg | Whoosah Beachclub | 52.12114 | 4.292650 | 0 | 389 | http://simfuny.com/event/113285/last-minute-vunzige-deuntjes-at-the-beach-whoosah | https://www.facebook.com/events/1345309678917690 | 558 | 1400 | Sat | Sun | 2017-06-24 | 2017-06-25 | FALSE | 2017-06-25 01:03:48 | 48972.4771 | FALSE |
- Sample partyflock data
event_name | attending | event_location | date | timestamp |
---|---|---|---|---|
3x NYX | 394 | NYX | 16 september 2017 | 2017-09-16 01:22:47 |
Bongo's Bingo | 142 | Panama | 16 september 2017 | 2017-09-16 01:22:47 |
Claire & Late Night Burners | 135 | Claire | 16 september 2017 | 2017-09-16 01:22:47 |
Contact | 31 | OT301 | 16 september 2017 | 2017-09-16 01:22:47 |
Curated By Lenzman | 674 | Paradiso | 16 september 2017 | 2017-09-16 01:22:47 |
- Sample map outputs (very rough sketch)
-
set unified data model
-
change window of pulled data and increase number of pulled events
-
make all funs that call apis more safe (safely ...)
-
tests, especially the scraping stuff
-
write docs
-
check out role model pkgs for web api's testing (googledrive, osmdata)
-
consider switching to crul pkg for async requests
- dump each days data into monetdb
- parse datetime data into nicer format, separate features
- spatial visualisation prototype
- refactor code, funs in separate files
- add distance to Rembrandtplein as feature to each event (obs)
- partyflock scraper
- fake facebook account scraper
- get coordinates of all locations in amsterdam
-
spatio-temporal event analysis
-
clustering of highly-popular events in certain area
-
data mining of temporal patterns e.g. 1st Saturday of each month very busy