TReasoner: System Description Andrei Grigorev Postgraduate student Tyumen State University.

Slides:



Advertisements
Similar presentations
Charting the Potential of Description Logic for the Generation of Referring Expression SELLC, Guangzhou, Dec Yuan Ren, Kees van Deemter and Jeff.
Advertisements

Description Logic Based Ontology Languages Ian Horrocks Information Systems Group Oxford University Computing Laboratory.
Searching for the Holy Grail Ian Horrocks Information Systems Group Oxford University Computing Laboratory.
Transitive Roles in Number Restrictions Yevgeny Kazakov, Ulrike Sattler, Evgeny Zolin The University of Manchester
Knowledge Representation and Reasoning using Description Logic Presenter Shamima Mithun.
OWL - DL. DL System A knowledge base (KB) comprises two components, the TBox and the ABox The TBox introduces the terminology, i.e., the vocabulary of.
Query Answering for OWL-DL with Rules Boris Motik Ulrike Sattler Rudi Studer.
An Introduction to Description Logics
JAYASRI JETTI CHINMAYA KRISHNA SURYADEVARA
Ontological Logic Programming by Murat Sensoy, Geeth de Mel, Wamberto Vasconcelos and Timothy J. Norman Computing Science, University of Aberdeen, UK 1.
1 NP-completeness Lecture 2: Jan P The class of problems that can be solved in polynomial time. e.g. gcd, shortest path, prime, etc. There are many.
The Theory of NP-Completeness
OWL Datatypes: Design and Implementation Boris Motik and Ian Horrocks University of Oxford.
Combining the strengths of UMIST and The Victoria University of Manchester A Tableaux Decision Procedure for SHOIQ Ian Horrocks and Ulrike Sattler University.
CPSC 322, Lecture 19Slide 1 Propositional Logic Intro, Syntax Computer Science cpsc322, Lecture 19 (Textbook Chpt ) February, 23, 2009.
Computability and Complexity 9-1 Computability and Complexity Andrei Bulatov Logic Reminder (Cnt’d)
Logical Agents Chapter 7. Why Do We Need Logic? Problem-solving agents were very inflexible: hard code every possible state. Search is almost always exponential.
Description Logics. Outline Knowledge Representation Knowledge Representation Ontology Language Ontology Language Description Logics Description Logics.
Search in the semantic domain. Some definitions atomic formula: smallest formula possible (no sub- formulas) literal: atomic formula or negation of an.
Last time Proof-system search ( ` ) Interpretation search ( ² ) Quantifiers Equality Decision procedures Induction Cross-cutting aspectsMain search strategy.
Reducing OWL Entailment to Description Logic Satisfiability Ian Horrocks and Patel Schneider Presented by: Muhammed Al-Muhammed.
DL systems DL and the Web Ilie Savga
Basics of Reasoning in Description Logics
1 The Theory of NP-Completeness 2 NP P NPC NP: Non-deterministic Polynomial P: Polynomial NPC: Non-deterministic Polynomial Complete P=NP? X = P.
Daniel Kroening and Ofer Strichman 1 Decision Procedures in First Order Logic Decision Procedures for Equality Logic.
The Theory of NP-Completeness 1. Nondeterministic algorithms A nondeterminstic algorithm consists of phase 1: guessing phase 2: checking If the checking.
The Theory of NP-Completeness 1. What is NP-completeness? Consider the circuit satisfiability problem Difficult to answer the decision problem in polynomial.
An Introduction to Description Logics. What Are Description Logics? A family of logic based Knowledge Representation formalisms –Descendants of semantic.
1 The Theory of NP-Completeness 2012/11/6 P: the class of problems which can be solved by a deterministic polynomial algorithm. NP : the class of decision.
Computing & Information Sciences Kansas State University Wednesday, 08 Oct 2008CIS 530 / 730: Artificial Intelligence Lecture 17 of 42 Wednesday, 08 October.
Notes on DL Reasoning Shawn Bowers April, 2004.
Quantified Formulas - Decision Procedure Daniel Kroening, Ofer Strichman Presented by Changki Hong 07 NOV 08.
Daniel Kroening and Ofer Strichman 1 Decision Procedures in First Order Logic Decision Procedures for Equality Logic Range Allocation.
Using the TBox to Optimise SPARQL Queries Birte Glimm Yevgeny Kazakov Ilianna Kollia and Giorgos Stamou CS 848 Paper Critique Vishnu Prathish.
INTRODUCTION TO ARTIFICIAL INTELLIGENCE COS302 MICHAEL L. LITTMAN FALL 2001 Satisfiability.
110/19/2015CS360 AI & Robotics AI Application Areas  Neural Networks and Genetic Algorithms  These model the structure of neurons in the brain  Humans.
Presented by:- Somya Gupta( ) Akshat Malu ( ) Swapnil Ghuge ( ) Franz Baader, Ian Horrocks, and Ulrike Sattler.
Logical Agents Logic Propositional Logic Summary
DRAGO: Distributed Reasoning Architecture for the Semantic Web Andrei Tamilin and Luciano Serafini Work is supported by 1 June 2005 Second European Semantic.
An Introduction to Description Logics (chapter 2 of DLHB)
Semantic web course – Computer Engineering Department – Sharif Univ. of Technology – Fall Description Logics: Logic foundation of Semantic Web Semantic.
Sound Global Caching for Abstract Modal Tableaux Rajeev Goré The Australian National University  Linh Anh Nguyen University of Warsaw CS&P’2008.
1 How to decide Query Containment under Constraints using a Description Logic Ian Horrocks, Ulrike Sattler, Sergio Tessaris, and Stephan Tobies presented.
LDK R Logics for Data and Knowledge Representation Description Logics (ALC)
NP-Complete Problems. Running Time v.s. Input Size Concern with problems whose complexity may be described by exponential functions. Tractable problems.
Logical Agents Chapter 7. Outline Knowledge-based agents Logic in general Propositional (Boolean) logic Equivalence, validity, satisfiability.
Nikolaj Bjørner Microsoft Research DTU Winter course January 2 nd 2012 Organized by Flemming Nielson & Hanne Riis Nielson.
CSE-291: Ontologies in Data Integration Department of Computer Science & Engineering University of California, San Diego CSE-291: Ontologies in Data Integration.
CSE 6311 – Spring 2009 ADVANCED COMPUTATIONAL MODELS AND ALGORITHMS Lecture Notes – Feb. 3, 2009 Instructor: Dr. Gautam Das notes by Walter Wilson.
NPC.
1 Instance Store Database Support for Reasoning over Individuals S Bechhofer, I Horrocks, D Turi. Instance Store - Database Support for Reasoning over.
An Efficient Tableau Prover Using Global Caching for the Description Logic ALC Linh Anh Nguyen University of Warsaw CS&P’2008.
The Theory of NP-Completeness 1. Nondeterministic algorithms A nondeterminstic algorithm consists of phase 1: guessing phase 2: checking If the checking.
Computing & Information Sciences Kansas State University Wednesday, 04 Oct 2006CIS 490 / 730: Artificial Intelligence Lecture 17 of 42 Wednesday, 04 October.
Daniel Kroening and Ofer Strichman 1 Decision Procedures in First Order Logic Decision Procedures for Equality Logic.
Artificial Intelligence Logical Agents Chapter 7.
The Theory of NP-Completeness
Description Logic Reasoning
Richard Anderson Lecture 26 NP-Completeness
Richard Anderson Lecture 26 NP-Completeness
Intelligent Systems (AI-2) Computer Science cpsc422, Lecture 20
Richard Anderson Lecture 25 NP-Completeness
ECE 667 Synthesis and Verification of Digital Circuits
NP-Complete Problems.
PROPOSITIONAL LOGIC - SYNTAX-
Graphs and Algorithms (2MMD30)
The Theory of NP-Completeness
Instructor: Aaron Roth
CIS Monthly Seminar – Software Engineering and Knowledge Management IS Enterprise Modeling Ontologies Presenter : Dr. S. Vasanthapriyan Senior Lecturer.
Lecture 23 NP-Hard Problems
Presentation transcript:

TReasoner: System Description Andrei Grigorev Postgraduate student Tyumen State University

TReasoner: Overview TReasoner is OWL Reasoner made on Java (developed in NetBeans) with usage of OWL API; Do classification and check consistency in SHOIQ logic with datatype expression support; Source code and precompiled libraries are available at 2

TReasoner: Implementation It: – implements teableau algorithm for SHOIQ (developed by Ian Horrocks and Ulrike Sattler); – uses backJumping, caching, global caching; – Implements some novel optimization techniques (SS-branching, Bron-Kerbosch algo, isomorphic concepts searching); 3

Wrapper TReasoner: Architecture 4 OWL API Knowledge base Checker Interpretation Ontology KBQuestionAnswer

About Optimizations: SS-branching ∀ R.( ∀ R1.(A ⊓ B ⊓ C)) and ∃ R.D ⊓ ∃ R.( ∃ R1.( ¬ B ⊔ ¬ C) ⊓ (T ⊓ F ⊔ G)) ⊔ ∃ R.( ∃ R1.( ¬ A ⊔ ¬ B)) For disjointness checking no tableau decision procedure needed; Unsat are determined by simple procedure named SS-branching; 5

About Optimizations: SS-branching If C and D are concepts, and C is conjunction of concepts C1, C2, …, Cn and D is conjunction of concepts D1, D2, …, Dm, then if some Ci and some Dj are disjoint then C and D are disjoint; Such conditions where determined for disjunctions, disjunction and conjunctions, exist and for all quantifiers; If Ci and Dj are complex concepts (not literals) for disjointness checking same procedure are used; It helps to determine wide class of cases of disjoint concepts. 6

About Optimizations: Bron-Kerbosch algo using For concept in DNF form: – (C11 ⊔ C12) ⊓ (C21 ⊔ C22) ⊓ (C31 ⊔ C32) ⊓ ( ¬ C11 ⊔ ¬ C21) ⊓ ( ¬ C11 ⊔ ¬ C31) ⊓ ( ¬ C21 ⊔ ¬ C31) ⊓ ( ¬ C12 ⊔ ¬ C22) ⊓ ( ¬ C12 ⊔ ¬ C32) ⊓ ( ¬ C22 ⊔ ¬ C32) Where each concept C** may be complex concept without disjunctions; Bron-Kerbosch algorithm may be used; 7

About Optimizations: Bron-Kerbosch algo using Build a n-partite graph: – Every concept proposition connection (C** ⊓ C**) is part of the graph G, C** presented as vertex in the part; – Edge between vertexes is exist iff ( ¬ C** ⊓¬ C**) are contained in concept expression; If n-clique in graph is exist then such concept is satisfiable (unsatisfiable otherwise); To date it works only for unsat concepts finding or for sat concepts without axioms in TBox; 8

(C11 ⊔ C12) ⊓ (C21 ⊔ C22) ⊓ (C31 ⊔ C32) ⊓ ( ¬ C11 ⊔ ¬ C21) ⊓ ( ¬ C11 ⊔ ¬ C31) ⊓ ( ¬ C21 ⊔ ¬ C31) ⊓ ( ¬ C12 ⊔ ¬ C22) ⊓ ( ¬ C12 ⊔ ¬ C32) ⊓ ( ¬ C22 ⊔ ¬ C32) About Optimizations: Bron-Kerbosch 9 C11 C12 C21 C22 C31 C32

About Optimizations: Isomorphic concepts New optimization technique is under development; Hypothesis: If two concepts are isomorphic, then first concept is satisfiable if and only if second concept is satisfiable; For isomorphic concepts finding algorithm Edmonds are used. 10

About Optimizations: Isomorphic concepts C1 ⊑ ∃ Role1.C2 C3 ⊑ ∃ Role1.C2 C2 ⊑ D1 D2 ⊑ C3 ⊓ C1 … 11 C3 ⊑ ∃ Role1.C2 C1 ⊑ ∃ Role1.C2 C2 ⊑ D1 D2 ⊑ C1 ⊓ C3 …

TReasoner: Evaluation 12

TReasoner: Evaluation 13

Thank you very much! 14

TReasoner: Evaluation Evaluation shows that, TReasoner may not compete yet with such strong systems like HermiT and FaCT++, but in future works reducing of worktime expected (I hope) 15