xLP: Explainable Link Prediction for Master Data Management
This work addresses the need for interpretable AI in enterprise settings where user trust and time costs are critical, though it appears incremental by combining existing methods.
The paper tackled the problem of explaining neural model predictions for link prediction in master data management by developing multiple explainability solutions, resulting in a demo that allows users to select preferred explanations to enhance trust and adoption.
Explaining neural model predictions to users requires creativity. Especially in enterprise applications, where there are costs associated with users' time, and their trust in the model predictions is critical for adoption. For link prediction in master data management, we have built a number of explainability solutions drawing from research in interpretability, fact verification, path ranking, neuro-symbolic reasoning and self-explaining AI. In this demo, we present explanations for link prediction in a creative way, to allow users to choose explanations they are more comfortable with.