Data and Knowledge Group

― Knowledge Representation and Reasoning

UOBM Generator

Overview

The University Ontology Benchmark (UOBM) is an extension of LUBM with a more complex ontology, which also contains disjunctive axioms and negation. UOBM provides three different data sets (for one, five and ten universities); in contrast to LUBM, no generator of data sets of varying size is provided for UOBM. To provide a more comprehensive evaluation, we have implemented a data generator for UOBM that replicates the design of LUBM's generator. Data obtained with our generator presents several differences with the default UOBM data. This is because the data in UOBM's default data sets is skewed in what we believe are rather strange ways; for example, students in the UOBM data sets are much more likely to be connected via the isFriendOf relation to faculty members than to other students. Our generator does not replicate this skewing, and thus produces what we believe is more "realistic" data.

Our UOBM Generator with a command line interface can be downloaded from here.



Download

Using UOBMGenerator

The UOBM Generator is a simple command line tool that produces data set for UOBM. See the bundled README.txt file for more information.

Researchers

Yujiao Zhou, Bernardo Cuenca Grau, and Ian Horrocks

Technical Report

Yujiao Zhou, Bernardo Cuenca Grau, Ian Horrocks, Zhe Wu and Jay Banerjee.
Making the Most of your Triple Store: Query Answering in OWL 2 Using an RL Reasoner.