Combine machine learning and knowledge representation to facilitate the process of assessing evidence from studies of drug-drug interaction by using an existing ontology of evidence types (e.g. DDI clinical trials vs. pharmacokinetic trials), as the backbone to develop a series of classifiers that categorize a DDI study’s evidence type based on its textual characteristics.
-
Data: contains all data files that are used/generated throughout the project.
-
Scripts: contains all Python source codes that are developed to implement the project. Python programs are used to preprocess data, generate features and develop machine learning models. Please go to Scripts folder for details of how each program works.
-
Programming Language: Python (version 3.0)
-
Please make sure that you have the following programs on your machine in order to run the script:
+ Python 3.7: https://www.python.org/downloads/ + Jupyter Notebook: http://jupyter.org/install