Re-thinking Knowledge Graph Completion Evaluation from an Information Retrieval Perspective
This work addresses evaluation fairness for KGC researchers, highlighting a critical flaw in standard benchmarks and proposing improved methods, though it is incremental as it builds on existing IR paradigms.
The paper tackles the problem of evaluating knowledge graph completion (KGC) systems by showing that current entity ranking protocols are unreliable due to unlabeled positive examples, and finds that switching to more complete labels drastically changes model rankings, with IR-like macro metrics proving more stable.
Knowledge graph completion (KGC) aims to infer missing knowledge triples based on known facts in a knowledge graph. Current KGC research mostly follows an entity ranking protocol, wherein the effectiveness is measured by the predicted rank of a masked entity in a test triple. The overall performance is then given by a micro(-average) metric over all individual answer entities. Due to the incomplete nature of the large-scale knowledge bases, such an entity ranking setting is likely affected by unlabelled top-ranked positive examples, raising questions on whether the current evaluation protocol is sufficient to guarantee a fair comparison of KGC systems. To this end, this paper presents a systematic study on whether and how the label sparsity affects the current KGC evaluation with the popular micro metrics. Specifically, inspired by the TREC paradigm for large-scale information retrieval (IR) experimentation, we create a relatively "complete" judgment set based on a sample from the popular FB15k-237 dataset following the TREC pooling method. According to our analysis, it comes as a surprise that switching from the original labels to our "complete" labels results in a drastic change of system ranking of a variety of 13 popular KGC models in terms of micro metrics. Further investigation indicates that the IR-like macro(-average) metrics are more stable and discriminative under different settings, meanwhile, less affected by label sparsity. Thus, for KGC evaluation, we recommend conducting TREC-style pooling to balance between human efforts and label completeness, and reporting also the IR-like macro metrics to reflect the ranking nature of the KGC task.