Skip to content

VICKEY: Mining Conditional Keys on RDF Knowledge Bases

Notifications You must be signed in to change notification settings

lgalarra/vickey

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

17 Commits
 
 
 
 
 
 
 
 

Repository files navigation

VICKEY: Mining Conditional Keys on Knowledge Bases

Introduction

A conditional key is a key constraint that is valid in only a part of the data. VICKEY is a system that can automatically mine conditional keys on large knowledge bases (KBs). For this, VICKEY combines techniques from key mining with techniques from rule mining. VICKEY can scale to knowledge bases of millions of facts. In addition, the conditional keys mined by VICKEY can improve the quality of entity linking by up to 47 percentage points.

Downloads

We ran two rounds of experiments to assess VICKEY. The first round, which we call runtime, aims at evaluating VICKEY's scability. For this purpose we ran the system on 9 DBpedia classes and reported their runtimes. We compared the VICKEY with a rule mining method based on the AMIE system. The second round, called linking, aims at showing the benefit of conditional keys in the task of entity linking between ontologies. For this purpose we evaluated the precision and recall of a data linking using (a) classical keys, (b) conditional keys, and (c) a combination of both.

Downloads runtime

Downloads linking

About

VICKEY: Mining Conditional Keys on RDF Knowledge Bases

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages