Logo image
Back
A case-based methodology for computational support of professional ethics evaluation
Dissertation   Open access

A case-based methodology for computational support of professional ethics evaluation

Christopher B. Rauch
Doctor of Philosophy (Ph.D.), Drexel University
Jun 2026
DOI:
https://doi.org/10.17918/00011459
pdf
Rauch_Christopher_20263.09 MBDownloadView
PDF Open Access Open Access (License Unspecified)

Abstract

Case-based reasoning Computational ethics Defeasible reasoning Engineering ethics Large language models Epistemology Knowledge Representation
Professional fields encode their ethical norms in codes of conduct and accumulate bodies of adjudicated cases that practitioners treat as precedent. Analyzing a new case against that precedent calls for attention to roles, obligations, actions, and the temporal order of events, the same features a review board weighs. ProEthica (proethica.org) demonstrates a transferable five-part methodology for domain-specific ethical case analysis, comprising a nine-component case representation, a three-layer formal ontology, a three-step ontology-constrained extraction process that applies large language models under that ontology, a synthesis step that produces linked analytical views, and a human review loop that governs the shared vocabulary. Two studies evaluate the implementation. The first shows that retrieval over the component-aware representation recovers the precedents expert boards actually cited at higher rates than a section-text embedding baseline. The second finds that engineering-trained evaluators judge each of the five synthesis views helpful for case analysis. The result is a transferable method for converting published ethics opinions into structured, queryable cases that serve both automated precedent retrieval and human deliberation.

Metrics

1 Record Views

Details

Logo image