dl based stream reasoning
play

DL-based Stream Reasoning Emanuele Della Valle Share, Remix, Reuse - PowerPoint PPT Presentation

How to Build a Stream Reasoning Application D. Dell'Aglio, E. Della Valle, T. Le-Pham, A. Mileo, and R. Tommasini http://streamreasoning.org/events/streamapp2017 DL-based Stream Reasoning Emanuele Della Valle Share, Remix, Reuse Legally


  1. How to Build a Stream Reasoning Application D. Dell'Aglio, E. Della Valle, T. Le-Pham, A. Mileo, and R. Tommasini http://streamreasoning.org/events/streamapp2017 DL-based Stream Reasoning Emanuele Della Valle

  2. Share, Remix, Reuse — Legally This work is licensed under the Creative Commons  Attribution 3.0 Unported License. Your are free:  • to Share — to copy, distribute and transmit the work • to Remix — to adapt the work Under the following conditions  • Attribution — You must attribute the work by inserting a credits slide stating – These slides are partially based on “ How to Build a Stream Reasoning Application 2017 ” by D. Dell'Aglio, E. Della Valle, T. Le-Pham, Mileo, and R. Tommasini available online at http://streamreasoning.org/events/streamapp2017 To view a copy of this license, visit  http://creativecommons.org/licenses/by/3.0/ http://streamreasoning.org/events/streamapp2017 2

  3. A model to describe stream processing Application Stream Processing (DSMS) Stream Processing Window merge Event Processing (CEP) Window operator Streams D. Dell’Aglio , On Unified Stream Reasoning , PhD thesis, Politecnico di Milano, 2016. http://streamreasoning.org/events/streamapp2017 3

  4. Solutions vs. requirements DSMS Sem Requirement CEP Web volume velocity variety incompleteness noise reactive answers fine-grained information access complex domain models high-level languages http://streamreasoning.org/events/streamapp2017 4

  5. Stream Reasoning • Research question – is it possible to make sense in real time of multiple , heterogeneous , gigantic and inevitably noisy and incomplete data streams in order to support the decision processes of extremely large numbers of concurrent users? Emanuele Della Valle: On Stream Reasoning . PhD thesis, Vrije Universiteit Amsterdam, 2015. Available online at http://dare.ubvu.vu.nl/handle/1871/53293 . http://streamreasoning.org/events/streamapp2017 5

  6. Is this feasible? • Proposed approach: cascading Stream Reasoning 1 Hz NEXPTIME Abstraction DL Reasoning Selection DL-Lite Querying PTIME Semantic Streams Interpretation Re-writing Raw Stream Processing 10 4 Hz AC 0 Complexity Complexity vs. Dynamics Change Frequency H. Stuckenschmidt, S. Ceri, E. Della Valle, F. van Harmelen: Towards Expressive Stream Reasoning. Proceedings of the Dagstuhl Seminar on Semantic Aspects of Sensor Networks, 2010. http://streamreasoning.org/events/streamapp2017 6

  7. A model to describe stream reasoning Application Stream Processing (DSMS) Stream Reasoning Graph-level entailment Window merge Event Processing (CEP) Window-level entailment Window operator Stream-level entailment Streams D. Dell’Aglio , On Unified Stream Reasoning , PhD thesis, Politecnico di Milano, 2016. http://streamreasoning.org/events/streamapp2017 7

  8. A model to describe stream reasoning Application Stream Processing (DSMS) Stream Reasoning Graph-level entailment Window merge Event Processing (CEP) Window-level entailment Window operator Stream-level entailment Streams D. Dell’Aglio , On Unified Stream Reasoning , PhD thesis, Politecnico di Milano, 2016. http://streamreasoning.org/events/streamapp2017 8

  9. continuous deductive reasoning • DL Ontology Stream S T – A ontology stream with respect to a static Tbox T is a sequence of Abox axioms S T (i) • A Windowed Ontology Stream S T (o,c] – A windowed ontology stream with respect to a static Tbox T is the union of the Abox axioms S T (i) where o< i≤c • Reasoning on a Windowed Ontology Stream S T (o,c] is as reasoning on a static DL KB Emanuele Della Valle, Stefano Ceri, Davide Francesco Barbieri, Daniele Braga, Alessandro Campi: A First Step Towards Stream Reasoning . FIS 2008: 72-81 http://streamreasoning.org/events/streamapp2017

  10. Example of Stream Reasoning 1/2 Query: measure the the impact of Alice's microposts  MEMO: our running example data model Post discusses For example  Bob posts p 2 . discusses discusses p 2 p 4 p 7 discusses discusses discusses discusses p 1 p 3 p 5 p 8 discusses p 6 Alice posts p 1 . 50 min ago 40 min ago 30 min ago 20 min ago 10 min ago now http://streamreasoning.org/events/streamapp2017 10

  11. Example of Stream Reasoning 2/2 What impact has been my micropost p 1 creating in the last hour? Let’s count the number of microposts that discuss it … REGISTER STREAM ImpactMeter AS SELECT (count(?p) AS ?impact) FROM STREAM <http://…/ fb> [RANGE 60m STEP 10m] WHERE { :Alice posts [ sr:discusses ?p ] } discusses discusses p 2 p 4 p 7 discusses Transitive property discusses discusses discusses p 1 p 3 p 5 p 8 Alice posts p 1 . p 6 discusses http://streamreasoning.org/events/streamapp2017 11

  12. MEMO: forms of reasoning for Q/A Data-driven (a.k.a. forward reasoning)  RDF Inferred Reasoner SPARQL data data ontology Query-driven – backward reasoning  RDF Reasoner SPARQL data ontology Query-driven – query rewriting (a.k.a. ontology based data access)  Rewritten SPARQL data Reasoner query ontology http://streamreasoning.org/events/streamapp2017 12

  13. Naïve Stream Reasoning Data-driven (a.k.a. forward reasoning)  RDF Inferred SPARQL S2R Reasoner data data ontology Query-driven – backward reasoning  RDF S2R Reasoner SPARQL data ontology http://streamreasoning.org/events/streamapp2017 13

  14. Backward and forward naïve Stream Reasoners Streaming knowledge base  • Ref: O. Walavalkar, A. Joshi, T. Finin and Y. Yesha, Streaming knowledge bases , in: In International Workshop on Scalable Semantic Web Knowledge Base Systems, 2008. C-SPARQL  • Ref: D.F. Barbieri, D. Braga, S. Ceri, E. Della Valle, Y. Huang, V. Tresp, A. Rettinger and H. Wermser, Deductive and inductive stream reasoning for semantic social media analytics , IEEE Intelligent Systems 25(6) (2010), 32 – 41. Sparkwave  • Ref: Sparkwave: Continuous Schema-Enhanced Pattern Matching over RDF Data Streams . Komazec S, Cerri D. DEBS 2012 Dynamite  • Ref: J. Urbani, A. Margara, C.J.H. Jacobs, F. van Harmelen and H.E. Bal, DynamiTE: Parallel materialization of dynamic 25 RDF data , in: International Semantic Web Conference (1) , Lecture Notes in Computer Science, Vol. 8218, Springer, 2013, pp. 657 – 672. Yasper  • Ref.. R. Tommasini, E. Della Valle, Challenges and issues of an RSP-QL implementation , in: Web Stream Processing workshop, 2017 • You will use it in the next hands on session http://streamreasoning.org/events/streamapp2017 14

  15. Naïve Stream Reasoning Data-driven (a.k.a. forward reasoning)  RDF Inferred SPARQL S2R Reasoner data data ontology Query-driven – backward reasoning  RDF S2R Reasoner SPARQL data ontology Query-driven – query rewriting (a.k.a. ontology based data access)  Rewritten SPARQL Reasoner data S2R query ontology http://streamreasoning.org/events/streamapp2017 15

  16. Naïve query-driven stream reasoning by query rewriting MEMO  Rewritten SPARQL Reasoner data S2R query ontology It is not that straight forward :-(  • Lack of a standard query language for DSMS and CEP • Lack of a well-understood operational semantics for DSMS and CEP (cf. SECRET by I. Botan et al., PVLDB 3(1), 2010) • Lack of expressiveness in OWL2QL http://streamreasoning.org/events/streamapp2017 16

  17. Query rewriting naïve Stream Reasoners Jean-Paul Calbimonte, Óscar Corcho, Alasdair J. G. Gray:  Enabling Ontology-Based Access to Streaming Data Sources . International Semantic Web Conference (1) 2010: 96-111 J-P, Calbimonte, J. Mora, O. Corcho, Query rewriting in  RDF stream processing , in: Extended Semantic Web Conference, 2016 http://streamreasoning.org/events/streamapp2017 17

  18. Not so naïve stream reasoning Naïve data-driven approach  RDF Inferred SPARQL S2R Reasoner data data ontology From snapshots to changes  Incremental!!! • What has just been inserted? • What has just been deleted? insertions Inferred SPARQL S2R Reasoner data deletions ontology http://streamreasoning.org/events/streamapp2017 18

  19. Not so naïve stream reasoning MEMO  Incremental !!! insertions Inferred SPARQL S2R Reasoner data deletions ontology The problem is that materialization (the result of data-  driven processing) are very difficult to decrement efficiently. • State-of-the-art: DReD algorithm – Over delete – Re-derive – Insert Ceri, S., Widom, J.: Deriving production rules for incremental view maintenance. In: Lohman,G.M., Sernadas, A., Camps, R. (eds.) VLDB, pp. 577 – 589. Morgan Kaufmann, San Francisco (1991) http://streamreasoning.org/events/streamapp2017 19

  20. The Intuition of DRed Algorithm Let’s assume that we have the following materialized graph  discusses p 2 discusses discusses discusses discusses p 1 p 4 p 3 While inserts are not problematic, deletion are difficult to  handle. If we delete p 2 discusses p 1 (p 2 ->p 1 ), we have • overestimate the impact of the deletion and mark for deletion p 4 ->p 1 that can be derived by p 4 ->p 2 and p 2 ->p 1 discusses p 2 discusses discusses discusses discusses p1 p 4 p 3 • look for alternative derivation of p 4 ->p 1 and eventually find the chain p 4 ->p 3 and p 3 ->p 1 discusses discusses discusses p 1 p 4 p 3 http://streamreasoning.org/events/streamapp2017 20

Recommend


More recommend