A Re-evaluation of Knowledge Graph Completion Methods
This work addresses evaluation inconsistencies in KGC research, which is crucial for researchers in data mining, machine learning, and NLP, though it is incremental as it focuses on protocol refinement rather than new methods.
The paper identifies that inflated performance in Knowledge Graph Completion (KGC) methods is due to flawed evaluation protocols and proposes a robust protocol to address this bias, leading to corrected performance metrics for existing methods.
Knowledge Graph Completion (KGC) aims at automatically predicting missing links for large-scale knowledge graphs. A vast number of state-of-the-art KGC techniques have got published at top conferences in several research fields, including data mining, machine learning, and natural language processing. However, we notice that several recent papers report very high performance, which largely outperforms previous state-of-the-art methods. In this paper, we find that this can be attributed to the inappropriate evaluation protocol used by them and propose a simple evaluation protocol to address this problem. The proposed protocol is robust to handle bias in the model, which can substantially affect the final results. We conduct extensive experiments and report the performance of several existing methods using our protocol. The reproducible code has been made publicly available