@inproceedings{c4513b12bd5643afa4d6e2496b9c1ece,
title = "On link prediction in knowledge bases: Max-K criterion and prediction protocols",
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.",
keywords = "Evaluation metric, Knowledge base embedding, Link prediction",
author = "Jiajie Mei and Richong Zhang and Yongyi Mao and Ting Deng",
note = "Publisher Copyright: {\textcopyright} 2018 ACM.; 41st Annual International ACM SIGIR Conference on Research and Development in Information Retrieval, SIGIR 2018 ; Conference date: 08-07-2018 Through 12-07-2018",
year = "2018",
month = jun,
day = "27",
doi = "10.1145/3209978.3210029",
language = "英语",
series = "41st International ACM SIGIR Conference on Research and Development in Information Retrieval, SIGIR 2018",
publisher = "Association for Computing Machinery, Inc",
pages = "755--764",
booktitle = "41st International ACM SIGIR Conference on Research and Development in Information Retrieval, SIGIR 2018",
}