The WebEnsemble application uses crowdsourcing to collect predictions for provided data sets and combines them into a single set of predictions for the data, representing an ensemble of the models that generated the original predictions. It is designed for classification, but may be adaptable to other tasks.
The project download contains a short paper describing the application's capabilities and use, along with a program that can be used to convert data sets between C4.5 or ARFF formats and the XML format the application uses to represent data sets.
The application was developed using NetBeans IDE, and the download contains a folder "WebEnsemble" which can be opened as a project in the IDE. The javadoc and deployable WAR file for the application can be found in that folder under the "dist" subfolder.
The application requires an SQL database, with the database requirements described in a text file included in the download. The application was designed and tested using a GlassFish application server; its compatibility with other application servers has not been tested.
- Changes to previous version:
Initial Announcement on mloss.org.
No one has posted any comments yet. Perhaps you'd like to be the first?
Leave a comment
You must be logged in to post comments.