Europeana-TwitterBot
Switch branches/tags
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Failed to load latest commit information.
Twitter-EuropeanaBot
.gitignore
.travis.yml
CHANGES.md
README.md

README.md

Build Status

twitter-europeanabot

the aim of this twitter bot is to explore the diverse contents of Europeana.

The programm takes a list of "seed terms" from a file, searches for corresponding images and posts the URL of the first result to the Twitter-Account EuropeanaBot.

In the current implementation I used [Austrian place names] (http://www.statistik.at/web_de/klassifikationen/regionale_gliederungen/gemeinden/index.html) and a list of Nobel Prize winners.

What's New?

Please refer to the CHANGES file

How To Deploy

  1. get API keys

you'll need an Europeana API-key and Twitter OAuth credentials.

  1. get seed file

I used a list of Austrian place names which I got from [Statistik Austria] (http://www.statistik.at/web_de/klassifikationen/regionale_gliederungen/gemeinden/index.html) as a seed file for location searches. A CSV file with all Nobel Prize winners was found at http://console.apihq.com/nobel-prize-api.

  1. edit config files

Rename logging.conf_example to logging.conf and edit if necessary. Rename èuropeana.json_example' to 'europeana.json' and edit the parameters.

  • pidbase location of PID-file, should be user-writable

  • debug if set to 1, all messages are only written into the logfile and not posted on twitter, default is 0

  • europeana_api_key see step 1

  • twitter_* see step 1

  • url_shortener generally the result urls are too long, here you can user also any other service

  • location_file list of possible location search terms in a csv file. If you use a different source file, you may have to adjust the subroutine createSeed in EuropeanaBot.pm

  • nobel_file CSV file with a list of Nobel Prize winners

  • user_agent the Wikipedia API requires setting a user agent for requests

  • wikipedia_base which Wikipedia version to use, defaults to http://en.wikipedia.org

  • sleep_time interval between searches

  • sleep_time_margin add a little random variation, to maybe fool Twitters bot detection algorithm ;-)

More Information

Talks

Blog posts