Name
Abstract | ||
|---|---|---|
Building knowledge base embedding models for link prediction has achieved great success. We however argue that the conventional top-k criterion used for evaluating the model performance is inappropriate. This paper introduces a new criterion, referred to as max-k. Through theoretical analysis and experimental study, we show that the top-k criterion is fundamentally inferior to max-k. We also introduce two prediction protocols for the max-k criterion. These protocols are strongly justified theoretically. Various insights concerning the max-k criterion and the two protocols are obtained through extensive experiments.
|
Year | DOI | Venue |
|---|---|---|
2018 | 10.1145/3209978.3210029 | SIGIR |
Keywords | Field | DocType |
Knowledge Base Embedding,Link Prediction,Evaluation Metric | Data mining,Embedding,Computer science,Knowledge base | Conference |
ISBN | Citations | PageRank |
978-1-4503-5657-2 | 0 | 0.34 |
References | Authors | |
11 | 4 | |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Jiajie Mei | 1 | 2 | 0.72 |
| Richong Zhang | 2 | 232 | 39.67 |
| Yongyi Mao | 3 | 524 | 61.02 |
| Ting Deng | 4 | 149 | 12.51 |