An evaluation of models for runtime approximation in link discovery

Abstract

Time-efficient link discovery is of central importance to implement the vision of the Semantic Web. Some of the most rapid Link Discovery approaches rely internally on planning to execute link specifications. In newer works, linear models have been used to estimate the runtime the fastest planners. However, no other category of models has been studied for this purpose so far. In this paper, we study non-linear runtime estimation functions for runtime estimation. In particular, we study exponential and mixed models for the estimation of the runtimes of planners. To this end, we evaluate three different models for runtime on six datasets using 400 link specifications. We show that exponential and mixed models achieve better fits when trained but are only to be preferred in some cases. Our evaluation also shows that the use of better runtime approximation models has a positive impact on the overall execution of link specifications.

Complex Event Recognition with Allen Relations

“…We compared RTEC A with AEGLE and D 2 IA. AE-GLE (Georgala et al 2016) is a state-of-the-art system computing Allen relations that has been used in the link discovery framework LIMES (Ngomo et al 2021). AEGLE reduces each Allen relation into a subset of eight atomic comparisons between interval endpoints.…”

A Probabilistic Algorithm to Predict Missing Facts from Knowledge Graphs

“…Even if KGs could reach a perfect representation of the world, such world changes over time, and then link prediction still remains important. Furthermore, link prediction can be easily mapped to not only adding missing data to KGs [30,36], but also discovering links in the context of linked data [15], predicting new relationships over professional networks [11], and even applied to predicting drug-target and protein-protein interactions in biomedical graphs [13].…”