Since the beginning of the century, research on ontology learning has gained popularity. Automatically extracting and structuring knowledge relevant to a domain of interest from unstructured data is a major scientific challenge. We propose a new approach with modular ontology learning framework considering tasks from data pre-processing to axiom extraction. Whereas previous contributions considered ontology learning systems as tools to help the domain expert, we developed the proposed framework with full automation in mind.
Resources:
For usage :
pip install git+https://github.com/wikit-ai/olaf
For contribution :
git clone https://github.com/wikit-ai/olaf.git
cd olaf
python3 -m venv ./venv
source venv/bin/activate
pip install .
Pipelines can be run with the following command: olaf run demo_pipeline
.
Pipeline components are displayed with the following command: olaf show demo_pipeline
.
The text used can be updated in the file data/demo.txt
.
An example on how the library can be used is available in demontrators/demo_test.ipynb
.
One example of OLAF usage for LLM components evaluation is also available here : https://github.com/wikit-ai/olaf-llm-eswc2024.
When an algorithm is missing you can contribute by adding it. Please refer to the developer note in the documentation for more detailed information.
Marion Schaeffer, Matthias Sesboüé, Jean-Philippe Kotowicz, Nicolas Delestre, Cecilia Zanni-Merk, OLAF: An Ontology Learning Applied Framework, Procedia Computer Science, Volume 225, 2023, Pages 2106-2115, ISSN 1877-0509, https://doi.org/10.1016/j.procs.2023.10.201. (https://www.sciencedirect.com/science/article/pii/S1877050923013595)
This project is licensed under the Apache-2.0 License.