Mining Symbolic Rules to Explain Lung Cancer Treatments

authored by

Disha Purohit, Maria-Esther Vidal

Abstract

Knowledge Graphs (KGs) represent the convergence of data and knowledge as factual statements; they allow for the enrichment of decision-making semantically. Symbolic inductive learning enables uncovering relevant patterns, expressed, for example, as Horn clauses. Albeit powerful, existing symbolic inductive learning frameworks may mine many rules, being difficult for a user to extract actionable insights. This demo illustrates a pipeline to analyze mined logical rules toward discovering meaningful insights. The demo puts into perspective the role of semantic types in guiding the exploration of mined rules. Participants will observe strategies to traverse the mined logical statements and how the outcomes reveal patterns in the prescription of lung cancer treatments. A video is available online (https://www.youtube.com/watch?v=CN4a3kUjfJ4 &ab_channel=TIBSDMGroup), a Jupyter notebook executes a live demos (https://mybinder.org/v2/gh/SDM-TIB/DIGGER-ESWC2023Demo/HEAD?labpath=Mining%20Symbolic%20Rules%20To%20Explain%20Lung%20Cancer%20Treatments.ipynb), and source-code is available in GitHub (https://github.com/SDM-TIB/Mining_Symbolic_Rules_ESWC2023Demo).

Organisation(s)

L3S Research Centre

External Organisation(s)

German National Library of Science and Technology (TIB)

Type

Contribution to book/anthology

Pages

69-74

No. of pages

6

Publication date

2023

Publication status

Published

Peer reviewed

Yes

Sustainable Development Goals

SDG 3 - Good Health and Well-being

Electronic version(s)

https://doi.org/10.1007/978-3-031-43458-7_13 (Access: Closed)