LGAIMar 14, 2022

A Unified Framework for Rank-based Evaluation Metrics for Link Prediction in Knowledge Graphs

DeepMindHarvard
arXiv:2203.07544v226 citationsh-index: 30
AI Analysis

This work addresses the need for better evaluation metrics in knowledge graph research, though it is incremental as it builds upon existing rank-based methods.

The authors tackled the problem of evaluating link prediction in knowledge graphs by proposing a unified framework for rank-based metrics to improve interpretability and comparability across datasets, resulting in the introduction of several new metrics demonstrated in a benchmarking study.

The link prediction task on knowledge graphs without explicit negative triples in the training data motivates the usage of rank-based metrics. Here, we review existing rank-based metrics and propose desiderata for improved metrics to address lack of interpretability and comparability of existing metrics to datasets of different sizes and properties. We introduce a simple theoretical framework for rank-based metrics upon which we investigate two avenues for improvements to existing metrics via alternative aggregation functions and concepts from probability theory. We finally propose several new rank-based metrics that are more easily interpreted and compared accompanied by a demonstration of their usage in a benchmarking of knowledge graph embedding models.

Code Implementations2 repos
Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes