BO-LSTM is a model based on biomedical ontologies and Long short-term memory networks. The model was developed in python, using keras. To demonstrate its utility, we trained a classification model on the DDI corpus, using the ChEBI ontology as the reference ontology. This tutorial shows how to install BOLSTM, classify any text using this model, and train new models using other documents and ontologies.
Full details
Level of knowledge: Intermediate: able to
Unless otherwise stated, all materials created by the FOSTER consortium are licensed under a CREATIVE COMMONS
ATTRIBUTION 4.0 INTERNATIONAL LICENSE.
This project has received funding from the European Union’s Seventh Framework Programme for research,
technological development and demonstration under grant agreement no 612425.