PPLib People Programming LIBrary: Easily find the best Crowdsourcing process for your application
Before setting up PPLib, please install Java 8, sbt and git on your system.
Download PPLib onto your machine. In Terminal, type git clone https://github.com/uzh/PPLib
Install PPLib on your system. In Terminal: cd PPLib && sbt publish-local
In the build.sbt
file of your SBT-enabled application, add the PPLib dependency: libraryDependencies += "pdeboer" %% "pplib" % "0.1-SNAPSHOT"
. You can find an example build.sbt
file (with other dependencies as well) here
If you don't have one yet, add an application.conf
file in your target application in the resources
folder and configure your access code to the human computation portals you'd like to use. An example configuration file can be found here
Please find a step-by-step guide in our screencast
In order to get familiar with PPLib, we recommend the following steps:
Read how to formulate simple questions to crowd workers and act upon their responses by looking at this simple code.
Learn how to aggregate simple question-answer routines into processes by following the narrated guide on Crowd Processes. This will also teach you how the parameter system works and how to create instances of existing processes for your application
Learn how to use Recombination to find the best solution to your problem by following this guide: How to run the Recombinator
Learn how to use HComp portals by following this guide: How to create a new HComp Portal
Learn how to use instruction generation by following this guide: How to use question generation
Learn how to use Auto-Execution by following this guide: How to use Auto-Execution
Write Patrick