HyLAR: Hybrid Location-Agnostic Reasoning
Mehdi Terdjimi, Lionel Médini, Michael Mrissa
LIRIS - Université Lyon 1
Anatomy of a SPARQL endpoint
Well, that's Web 1.0!
Scalability? Availability?
Actual need
"Bridge the gap between the web and the semantic web"
(Phil Archer, W3C, Semweb.Pro Paris, Nov. 2014)
- Classical Web solutions
- Static data caching, code execution deferring
- Limitations
- Heterogeneous, resource-constrained clients
- Requirements
- Client-side reasoning
- Adaptivity and flexibility
Our proposition
An architecture for locating the different steps of a reasoning process on either server or client side
Objectives
- Provide an adaptable reasoning task
- Use SemWeb standards (DL reasoner)
State of the art
So first, grab a reasoner...
Mobile reasoners
- KRHyper (Sinner and Kleemann, 2005), FOL
- Embedded EL+ Reasoner (Grimm et al., 2012)
- Mine-ME 2.0 (Ruta et al., 2014), ALCN
JavaScript reasoners
Implementation
...Then, cut it into pieces and embed it in a modern webapp architecture...
Modularization of OWLReasoner steps
- Server-side as Node / Express modules
- Client-side as AngularJS services
Client queries
Sent by an agnostic angular service
Evaluation
...And test
3 configurations x 3 steps x 2 ontologies
- Ontology loading
- Ontology parsing and classification
- SPARQL query processing
Evaluation results
...
Network delays
Classification [P2] and reasoning [P3] times
- Qx: time for the query to reach the server
- Rx: time for the response to reach the client
- Px: processing time
- Times expressed in ms
Discussion
...And talk
Time calculation for M users sending N queries
Server-side:
P2server + M x N x (Q3 + P3server+ R3)
Client-side:
M x (Q0 + R0 + Q1 + R1) + P2client + N x P3client
Hybrid:
P2server + M x (Q0 + R0 + Q2 + R2) + N x P3client
P2 not linear, P3 linear
Perspectives
- Improve the reasoner
- Adaptive reasoning in view of different parameters:
- Client's computing resources
- Ontology size
- Network conditions
- Study the impact of more complex queries
References
-
(Grimm et al., 2012) Grimm, S., Watzke, M., Hubauer, T., Cescolini, F.: Embedded EL+ reasoning on programmable logic controllers. In: The Semantic Web - ISWC 2012, pp. 66-81. Springer (2012)
-
(Kollia and Glimm, 2014) Kollia, I., Glimm, B.: Optimizing sparql query answering over owl ontologies. arXiv preprint arXiv:1402.0576 (2014)
-
(Médini et al., 2013) Médini, L., Bâcle, F., Nguyen, H.D.T.: DataConf: Enriching conference publications with a mobile mashup application (May 2013), http://liris.cnrs.fr/publis/?id=6032, lIME'2013 Workshop at WWW'2013 conference
References
-
(Ruta et al., 2014) Ruta, M., Scioscia, F., Loseto, G., Gramegna, F., Ieva, S., Di Sciascio, E.: Mini- me 2.0: powering the semantic web of things. In: 3rd OWL Reasoner Evaluation Workshop (ORE 2014) (2014)
-
(Sinner and Kleemann, 2005) Sinner, A., Kleemann, T.: Krhyper - in your pocket. In: Automated Deduction - CADE-20, pp. 452-457. Springer (2005)
-
(Verborgh et al., 2014) Verborgh, R., Vander Sande, M., Colpaert, P., Coppens, S., Mannens, E., Van de Walle, R.: Web-scale querying through linked data fragments. In: Proceedings of the 7th Workshop on Linked Data on the Web (2014)
HyLAR: Hybrid Location-Agnostic Reasoning
Mehdi Terdjimi, Lionel Médini, Michael Mrissa
LIRIS - Université Lyon 1