About
Ollie is software that automatically identifies and extracts binary relationships from English sentences. Ollie is designed for information extraction, where the target relations cannot be specified in advance and speed is important.
To get a better idea of what Ollie does:
- Read the paper or the README.
- Download the binaries.
- Check out the source code.
Binaries
If you want to run Ollie on a small amount of text without modifying the source code, you can use an executable file that can be run from the command line. Follow these steps to get started:- Download the latest Ollie binary.
- Download the linear English MaltParser model and place it in the same directory as Ollie.
- Run
java -Xmx512m -jar ollie-app-latest.jar yourfile.txt. The input file should contain one sentence per line unless--splitis specified. Omit the input file for an interactive console.
Code
The Ollie source is available on GitHub.
Data
- Seed (sentence, extraction) pairs
- Bootstrapping set of parse patterns. Please note that this was developed for the paper. Since then, the dependency graph serialization has changed. Also, this set is reduced with some filters before the model is constructed.
- Model used for the paper. The present model included with Ollie has been enhanced.
- Annotations used for evaluation. These were used to graph the precision recall curves.
License
Ollie is licensed under a University of Washington Academic License.
Credits
Ollie was authored and developed by the following people at the University of Washington's Turing Center as part of the KnowItAll Project:
- Mausam
- Michael Schmitz
- Robert Bart
- Stephen Soderland
- Oren Etzioni
Ollie uses the following code and models, which are included in the release:
Contact
Please send all questions, comments, and bugs to the the Ollie mailing list.