Boris Motik’s research while affiliated with University of Oxford and other places

What is this page?


This page lists works of an author who doesn't have a ResearchGate profile or hasn't added the works to their profile yet. It is automatically generated from public (personal) data to further our legitimate goal of comprehensive and accurate scientific recordkeeping. If you are this author and want this page removed, please let us know.

Publications (192)


Fig. 1. Conceptual Architecture
Fig. 2. Bridging Ontology view in UML
Fig. 3. UML representation of two small ontologies
MAFRA — A MApping FRAmework for Distributed Ontologies in the Semantic Web
  • Article
  • Full-text available

90 Reads

·

93 Citations

·

Boris Motik

·

·

Ontologies as means for conceptualiz- ing and structuring domain knowledge within a community of interest are seen as a key to realize the Semantic Web vision. However, the decentral- ized nature of the Web makes achieving this con- sensus across communities difficult, thus, hamper- ing efficient knowledge sharing between them. In order to balance the autonomy of each community with the need for interoperability, mapping mech- anisms between distributed ontologies in the Se- mantic Web are required. In this paper we present MAFRA, an interactive, incremental and dynamic framework for mapping distributed ontologies in the Semantic Web. ized" approach of mediation is probably not flexible enough, and distributed systems of mediation are more appropriate. Building on this idea and on existing work, we introduce in this paper MAFRA, an Ontology MApping FRAmework (MAFRA) for distributed ontologies in the Semantic Web. Within MAFRA we provide an approach and conceptual framework that provides a generic view onto the overall dis- tributed mapping process. The distributed nature of Semantic Web entails significant degrees of information redundancy, incoherence and constant evolution, thus changing the nature of the ontology mapping problem: instead of creating a static specification document relating entities in two ontologies, a continuous, incremental, interactive and highly dynamic pro- cess supporting mapping evolution is required to scale up to the ever-changing nature of ontologies being mapped. Estab- lishing a mapping between two ontologies is an engineering process of consensus building between two communities al- ready agreeing on common ontologies for their own respec- tive domains. This task implies negotiation, so attention is paid to providing means for cooperative mapping. Thus, pro- posed framework offers support in all parts of the ontology mapping life-cycle.

Download

HermiT: A Highly-Efficient Reasoner for Description Logics

11 Reads

HermiT is a reasoner for the Description Logic SHOIQ +. The system is based on a novel "hypertableau" calculus and includes a number of new optimizations. HermiT demonstrates significantly faster performance when classifying large and complex ontologies than other state-of-the-art reasoners. HermiT is also the first reasoner able to clas-sify a number of important knowledge bases, including the original ver-sion of the GALEN ontology.


Citations (79)


... The success of GNNs in applications has stimulated lively research into their theoretical properties such as expressive power. A landmark result is due to Barceló et al. [5] which was among the first to characterize the expressive power of GNNs in terms of logic, see [11,8,6, 23] and references therein for related results. More precisely, Barceló et al. show that a basic GNN model with a constant number of iterations has exactly the same expressive power as graded modal logic GML in restriction to properties definable in first-order logic FO. ...

Reference:

Logical Characterizations of Recurrent Graph Neural Networks with Reals and Floats
On the Correspondence Between Monotonic Max-Sum GNNs and Datalog
  • Citing Conference Paper
  • September 2023

... More recently, such reductions for Horn description logics have been implemented and evaluated Carral, González, and Koopmann 2019). Such datalog rewritings have also been studied for existential rules, for guarded (Benedikt et al. 2022), nearly guarded (Gottlob, Rudolph, and Simkus 2014), warded (Berger et al. 2022) and shy (Leone et al. 2019) rule sets. ...

Rewriting the infinite chase
  • Citing Article
  • July 2022

Proceedings of the VLDB Endowment

·

Maxime Buron

·

Stefano Germano

·

[...]

·

Boris Motik

... The canonical transformation T N induced by a max GNN can only derive new unary facts, while KG completion requires also the derivation of binary facts. To address this limitation, we use an alternative encoding/decoding scheme (Liu et al. 2021;Tena Cucala et al. 2022) where binary facts are also encoded in feature vector components and edges in the graph correspond to different types of possible joins between unary and binary atoms. As shown in (Tena Cucala et al. 2023), such scheme can be captured by fixed encoding and decoding programs P enc and P dec so that the overall transformation is given by T P dec (T N (T Penc (D))), which in turn coincides with T P dec (T L+2 P N (T Penc (D))) by Theorem 7. Benchmarks, metrics, and baselines We used the inductive KG completion benchmarks by Teru, Denis, and Hamilton (2020), based on the FB15K-237 (Bordes et al. 2013), NELL-995 (Xiong, Hoang, and Wang 2017), and WN18RR (Dettmers et al. 2018) KGs. ...

Faithful Approaches to Rule Learning
  • Citing Conference Paper
  • July 2022

... RSP-QL [29] is a reference model that unifies the semantics of the existing RSP approaches. RSP has been extended to support ROT in various ways: (i) solutions incorporating efficient incremental maintenance of materializations of the windowed ontology streams [10,44,54,73], (ii) solutions for expressive Description Logics (DL) [47,68], and (iii) a solution for Answer Set Programming (ASP) [52]. More central to ROT is the logic-based framework for analyzing reasoning over streams (LARS) [13] that extends ASP for analytical reasoning over data streams. ...

Incremental Update of Datalog Materialisation: the Backward/Forward Algorithm
  • Citing Article
  • February 2015

Proceedings of the AAAI Conference on Artificial Intelligence

... In this approach, no redundancy is allowed, i.e. updated views reflect only currently true logical assertions: this differs from the overgrounding idea, which aims to materialize bigger portions of logic programs which can possibly support true logic assertions. Hu et al. (2022) extended further the idea, by proposing a general method in which modular parts of a Datalog view can be attached to ad-hoc incremental maintenance algorithms. For instance, one can plug in the general framework a special incremental algorithm for updating transitive closure patterns, etc. ...

Modular materialisation of Datalog programs
  • Citing Article
  • April 2022

Artificial Intelligence

... A Datalog engine then executes these rules to a fixed point, materializing a database for subsequent querying. While early systems such as OWLIM (Kolovski, Wu, and Eadon 2010) and WebPIE (Urbani et al. 2010) were challenged by slow performance, modern Datalog engines enable scaling useful queries to internet-scale datasets (Ajileye and Motik 2022). ...

Materialisation and data partitioning algorithms for distributed RDF systems
  • Citing Article
  • March 2022

Journal of Web Semantics

... They have indeed been studied to enrich the expressive power of rules expressed in Datalog [Consens and Mendelzon, 1993], or in disjunctive Datalog with a notable implementation in the DLV system [Dell'Armi et al., 2003]. More recently, restrictions of Datalog Z , an extension of Datalog which captures many data aggregation tasks by allowing arithmetic functions over integers at the cost of undecidability, have been studied to regain decidability, resulting in the fragment Limit Datalog Z [Cuenca Grau et al., 2020] whose expressive power has been further studied [Kaminski et al., 2021]. ...

The Complexity and Expressive Power of Limit Datalog
  • Citing Article
  • February 2022

Journal of the ACM

... Due to data richness and high coverage, the MAKG has been employed in several research fields and scenarios, including bibliometrics and scientific impact [27][28][29], recommender systems [30], data analytics (i.e., Nesta business intelligence tools 26 ), and benchmarking [31]. Further, the MAG, to which the MAKG originally derives, has been extensively investigated [32][33][34] and used for scientific ethics and mobility networks [24,25], and in COVID-19 related studies [35,36]. ...

Streaming Partitioning of RDF Graphs for Datalog Reasoning
  • Citing Chapter
  • May 2021

Lecture Notes in Computer Science

... They have indeed been studied to enrich the expressive power of rules expressed in Datalog [Consens and Mendelzon, 1993], or in disjunctive Datalog with a notable implementation in the DLV system [Dell'Armi et al., 2003]. More recently, restrictions of Datalog Z , an extension of Datalog which captures many data aggregation tasks by allowing arithmetic functions over integers at the cost of undecidability, have been studied to regain decidability, resulting in the fragment Limit Datalog Z [Cuenca Grau et al., 2020] whose expressive power has been further studied [Kaminski et al., 2021]. ...

Limit Datalog: A Declarative Query Language for Data Analysis
  • Citing Article
  • February 2020

ACM SIGMOD Record

... Specifically, we first use the pp-ocrv3 model to realize the invoice text recognition in the Chinese scene, then the extracted information is utilized to classify into commodities based on the BERT-TextCNN model. We next construct a knowledge graph and rules for financial reimbursement, and finally the compliance judgment of the invoice is realized via reasoning on the knowledge graph [5]. The experimental results show that our Chinese invoice audit framework can accurately and efficiently replace the manual invoice audit. ...

Datalog Reasoning over Compressed RDF Knowledge Bases
  • Citing Conference Paper
  • November 2019