Int egration de connaissances ontologiques pour lapprentissage des - PowerPoint PPT Presentation

Bayesian networks Ontologies BNs and Ontologies Elements of an ontology C : Classes (concepts) Landslide P : Attributes (properties) Fire Tsunami H : Hierarchical structure (is-a, part-of relations) Catastrophes R : Other semantic relationships Name Localization Volcano Flood Date I : Instances (individuals) Nb_killed A : Axioms (logic statements) Earthquake Name Mgnitude Localization Date Nb_killed Montassar Ben Messaoud, Mouna Ben Ishak, Philippe Leray, Nahla Ben Amor Apprentissage RB et Ontologies 14/22

Bayesian networks Ontologies BNs and Ontologies Elements of an ontology C : Classes (concepts) Catastrophes P : Attributes (properties) H : Hierarchical structure (is-a, is-a is-a part-of relations) Man-made Natural R : Other semantic relationships I : Instances (individuals) is-a is-a is-a is-a is-a is-a Fire Flood Earthquake Landslide Volcano Tsunami A : Axioms (logic statements) Montassar Ben Messaoud, Mouna Ben Ishak, Philippe Leray, Nahla Ben Amor Apprentissage RB et Ontologies 14/22

Bayesian networks Ontologies BNs and Ontologies Elements of an ontology C : Classes (concepts) Catastrophes P : Attributes (properties) is-a is-a H : Hierarchical structure (is-a, Man-made Natural part-of relations) R : Other semantic relationships is-a is-a is-a is-a is-a is-a Fire Flood Earthquake Landslide Volcano I : Instances (individuals) Tsunami A : Axioms (logic statements) Causes Causes Causes Causes Causes Causes Montassar Ben Messaoud, Mouna Ben Ishak, Philippe Leray, Nahla Ben Amor Apprentissage RB et Ontologies 14/22

Bayesian networks Ontologies BNs and Ontologies Elements of an ontology C : Classes (concepts) Catastrophes P : Attributes (properties) is-a is-a H : Hierarchical structure (is-a, Man-made Natural part-of relations) R : Other semantic relationships is-a is-a is-a is-a is-a is-a Fire Flood Earthquake Landslide Volcano I : Instances (individuals) Tsunami A : Axioms (logic statements) is-instance is-instance is-instance Aleppo Shaanxi Haiti Montassar Ben Messaoud, Mouna Ben Ishak, Philippe Leray, Nahla Ben Amor Apprentissage RB et Ontologies 14/22

Bayesian networks Ontologies BNs and Ontologies Ontology learning : subtasks Ontology population Get new instances of concept(s) already present in the ontology Ontology enrichment Update (add or modify) concepts, properties and relations in a given ontology Evolution vs. revolution Evolution (ontology continuity): Add new knowledge Revolution (ontology discontinuity): Modify existing knowledge Montassar Ben Messaoud, Mouna Ben Ishak, Philippe Leray, Nahla Ben Amor Apprentissage RB et Ontologies 15/22

Bayesian networks Ontologies BNs and Ontologies Ontology learning : methods [Buitelaar & Cimiano, 2008] ”Bridging the gap between text and knowledge” Natural Language Processing [Buitelaar et al., 2003, Velardi et al., 2005] Concept extraction Taxonomy learning (is-a, part-of) Population (information extraction) Machine Learning Clustering for taxonomy learning [Bisson et al., 2000] Association rules for relation discovery [Madche & Staab, 2000] ILP for relation discovery [Rudolph et al., 2007] Montassar Ben Messaoud, Mouna Ben Ishak, Philippe Leray, Nahla Ben Amor Apprentissage RB et Ontologies 16/22

Bayesian networks Ontologies BNs and Ontologies Outline ... Bayesian networks 1 BN definition BN learning Ontologies 2 Ontology definition Ontology learning BNs and Ontologies 3 Existing work Our proposal Montassar Ben Messaoud, Mouna Ben Ishak, Philippe Leray, Nahla Ben Amor Apprentissage RB et Ontologies 17/22

Bayesian networks Ontologies BNs and Ontologies Existing work Bayesian Network = ⇒ Ontology BayesOWL [Ding & Peng, 2004] OntoBayes [Yang & Calmet, 2005] PR-OWL [Costa & Laskey, 2006] Use of BNs for probabilistic modeling and reasoning (no learning) Ontology = ⇒ Bayesian Network BN ”basic” construction using ontologies [Devitt et al., 2006] Ontology-based semi-automatic construction of BN in E-health applications [Jeon & Ko, 2007] Use of ontologies to manually build a BN (without data) Montassar Ben Messaoud, Mouna Ben Ishak, Philippe Leray, Nahla Ben Amor Apprentissage RB et Ontologies 18/22

Bayesian networks Ontologies BNs and Ontologies Our proposal ”Bridging the gap between text and knowledge” ⇒ Bayesian Network = ⇒ Ontology Ontology + data = BN Structure Learning Ontology Evolution 2 variations Causal BNs 1 more semantical causal discovery SemCaDo 1.0 [Ben Messaoud & al. 2009] causal discovery for ontology evolution SemCaDo 2.0 [Ben Messaoud & al., 2009] more Object-Oriented BNs 2 OOBN structure learning and ontology evolution O2C [Ben Ishak et al., 2011] more Montassar Ben Messaoud, Mouna Ben Ishak, Philippe Leray, Nahla Ben Amor Apprentissage RB et Ontologies 19/22

Bayesian networks Ontologies BNs and Ontologies Our proposal ”Bridging the gap between data and knowledge” ⇒ Bayesian Network = ⇒ Ontology Ontology + data = BN Structure Learning Ontology Evolution 2 variations Causal BNs 1 more semantical causal discovery SemCaDo 1.0 [Ben Messaoud & al. 2009] causal discovery for ontology evolution SemCaDo 2.0 [Ben Messaoud & al., 2009] more Object-Oriented BNs 2 OOBN structure learning and ontology evolution O2C [Ben Ishak et al., 2011] more Montassar Ben Messaoud, Mouna Ben Ishak, Philippe Leray, Nahla Ben Amor Apprentissage RB et Ontologies 19/22

Bayesian networks Ontologies BNs and Ontologies Conclusion BN structure learning is NP hard, need of prior knowledge Ontology evolution is difficult, based on text-mining Cooperation can help for both tasks Originality of our proposal BN structure learning : use ontology instead of expert knowledge : separation between expert acquisition and structure learning Ontology evolution : use BN structure learning to directly discover relationships from data Difficulties No similar work or benchmark for a comparative study SemCaDo : one concept = attribute = node O2C : OOBN learning is too complex Montassar Ben Messaoud, Mouna Ben Ishak, Philippe Leray, Nahla Ben Amor Apprentissage RB et Ontologies 20/22

Bayesian networks Ontologies BNs and Ontologies Future works What next ? Creation of benchmarks Generalize SemCaDo or O2C with more general models Ideas : Multi Entity BNs [Laskey, 2006] Relational BNs [Getoor et al., 2007] Montassar Ben Messaoud, Mouna Ben Ishak, Philippe Leray, Nahla Ben Amor Apprentissage RB et Ontologies 21/22

Bayesian networks Ontologies BNs and Ontologies Thank you for your attention philippe.leray@univ-nantes.fr Montassar Ben Messaoud, Mouna Ben Ishak, Philippe Leray, Nahla Ben Amor Apprentissage RB et Ontologies 22/22

Bayesian network Structure Learning Causal Bayesian Networks SemCaDo algorithm O2C algorithm Outline ... Bayesian network Structure Learning 4 constraint-based methods score-based methods local search methods Causal Bayesian Networks 5 definition causal BN structure learning SemCaDo algorithm 6 definition experimental study O2C algorithm 7 definition algorithm Montassar Ben Messaoud, Mouna Ben Ishak, Philippe Leray, Nahla Ben Amor Apprentissage RB et Ontologies 1/49

Bayesian network Structure Learning Causal Bayesian Networks SemCaDo algorithm O2C algorithm Constraint-based methods How to search a good BN ? Constraint-based methods BN = independence model ⇒ find CI in data in order to build the DAG Score-based methods BN = probabilistic model that must fit data as well as possible ⇒ search the DAG space in order to maximize a scoring function Hybrid methods Montassar Ben Messaoud, Mouna Ben Ishak, Philippe Leray, Nahla Ben Amor Apprentissage RB et Ontologies 2/49

Bayesian network Structure Learning Causal Bayesian Networks SemCaDo algorithm O2C algorithm Constraint-based methods Two reference algorithms Pearl et Verma : IC, IC* Spirtes, Glymour et Scheines : SGS, PC, CI, FCI Common principle Build an undirected graph describing direct dependences between variables ( χ 2 tests) by adding edges (Pearl et Verma) by deleting edges (SGS) Detect V-structures (from previous statistical tests) Propagate some edge orientation (inferred edges) in order to obtain a CPDAG Montassar Ben Messaoud, Mouna Ben Ishak, Philippe Leray, Nahla Ben Amor Apprentissage RB et Ontologies 3/49

Bayesian network Structure Learning Causal Bayesian Networks SemCaDo algorithm O2C algorithm constraint-based methods Some inconvenients Reliability of CI test conditionally to several variables with a limited amount of data) SGS heuristic : if df < N 10 , then declare dependence Combinatorial explosion of the number of tests PC heuristic : begin with order 0 ( X A ⊥ X B ) then order 1 ( X A ⊥ X B | X C ), etc ... Montassar Ben Messaoud, Mouna Ben Ishak, Philippe Leray, Nahla Ben Amor Apprentissage RB et Ontologies 4/49

Bayesian network Structure Learning Causal Bayesian Networks SemCaDo algorithm O2C algorithm PC algorithm Step 0 : undirected complete graph Left : target BN used to generate 5000 samples A S A S T L B T L B O O X D X D Montassar Ben Messaoud, Mouna Ben Ishak, Philippe Leray, Nahla Ben Amor Apprentissage RB et Ontologies 5/49

Bayesian network Structure Learning Causal Bayesian Networks SemCaDo algorithm O2C algorithm PC algorithm Step 1a : delete all order 0 independences discovered χ 2 : S ⊥ A L ⊥ A B ⊥ A O ⊥ A X ⊥ A D ⊥ A T ⊥ S L ⊥ T O ⊥ B X ⊥ B A S A S T L B T L B O O X D X D Montassar Ben Messaoud, Mouna Ben Ishak, Philippe Leray, Nahla Ben Amor Apprentissage RB et Ontologies 5/49

Bayesian network Structure Learning Causal Bayesian Networks SemCaDo algorithm O2C algorithm PC algorithm Step 1a : delete all order 1 independences discovered χ 2 : T ⊥ A | O O ⊥ S | L X ⊥ S | L B ⊥ T | S X ⊥ T | O D ⊥ T | O ... A S A S T L B T L B O O X D X D Montassar Ben Messaoud, Mouna Ben Ishak, Philippe Leray, Nahla Ben Amor Apprentissage RB et Ontologies 5/49

Bayesian network Structure Learning Causal Bayesian Networks SemCaDo algorithm O2C algorithm PC algorithm Step 1a : delete all order 2 independences discovered χ 2 : D ⊥ S |{ L , B } X ⊥ O |{ T , L } D ⊥ O |{ T , L } A S A S T L B T L B O O X D X D Montassar Ben Messaoud, Mouna Ben Ishak, Philippe Leray, Nahla Ben Amor Apprentissage RB et Ontologies 5/49

Bayesian network Structure Learning Causal Bayesian Networks SemCaDo algorithm O2C algorithm PC algorithm Step 2 : research V-structures χ 2 : one V-structure T → O ← L is discovered A S A S T L B T L B O O X D X D Step 3 : inferred edges no one in this example Montassar Ben Messaoud, Mouna Ben Ishak, Philippe Leray, Nahla Ben Amor Apprentissage RB et Ontologies 5/49

Bayesian network Structure Learning Causal Bayesian Networks SemCaDo algorithm O2C algorithm PC algorithm From CPDAG to DAG Orientation of the remaining undirected edges (only constraint : do not create any new V-structure) A S A S T L B T L B O O X D X D Montassar Ben Messaoud, Mouna Ben Ishak, Philippe Leray, Nahla Ben Amor Apprentissage RB et Ontologies 5/49

Bayesian network Structure Learning Causal Bayesian Networks SemCaDo algorithm O2C algorithm PC algorithm Obtained DAG versus target one χ 2 test with 5000 samples fails to discover A → T , O → X and O → D A S A S T L B T L B O O X D X D Montassar Ben Messaoud, Mouna Ben Ishak, Philippe Leray, Nahla Ben Amor Apprentissage RB et Ontologies 5/49

Bayesian network Structure Learning Causal Bayesian Networks SemCaDo algorithm O2C algorithm Score-based methods How to search a good BN ? Constraint-based methods BN = independence model ⇒ find CI in data in order to build the DAG Score-based methods BN = probabilistic model that must fit data as well as possible ⇒ search the DAG space in order to maximize a scoring function Hybrid methods Montassar Ben Messaoud, Mouna Ben Ishak, Philippe Leray, Nahla Ben Amor Apprentissage RB et Ontologies 6/49

Bayesian network Structure Learning Causal Bayesian Networks SemCaDo algorithm O2C algorithm Notion of score General principle : Occam razor Pluralitas non est ponenda sine neccesitate (La pluralit´ e (des notions) ne devrait pas ˆ etre pos´ ee sans n´ ecessit´ e) plurality should not be posited without necessity Frustra fit per plura quod potest fieri per pauciora (C’est en vain que l’on fait avec plusieurs ce que l’on peut faire avec un petit nombre) It is pointless to do with more what can be done with fewer = Parcimony principle : find a model Fitting the data D : likelihood : L ( D| θ, B ) The simplest possible : dimension of B : Dim ( B ) Montassar Ben Messaoud, Mouna Ben Ishak, Philippe Leray, Nahla Ben Amor Apprentissage RB et Ontologies 7/49

Bayesian network Structure Learning Causal Bayesian Networks SemCaDo algorithm O2C algorithm Score examples AIC and BIC Compromise between likelihood and complexity Application of AIC (Aka¨ ıke 70) and BIC (Schwartz 78) criteria S AIC ( B , D ) = log L ( D| θ MV , B ) − Dim ( B ) S BIC ( B , D ) = log L ( D| θ MV , B ) − 1 2 Dim ( B ) log N Bayesian scores : BD, BDe, BDeu S BD ( B , D ) = P ( B , D ) (Cooper et Herskovits 92) BDe = BD + score equivalence (Heckerman 94) n q i r i Γ( α ij ) Γ( N ijk + α ijk ) � � � S BD ( B , D ) = P ( B ) Γ( N ij + α ij ) Γ( α ijk ) i =1 j =1 k =1 Montassar Ben Messaoud, Mouna Ben Ishak, Philippe Leray, Nahla Ben Amor Apprentissage RB et Ontologies 8/49

Bayesian network Structure Learning Causal Bayesian Networks SemCaDo algorithm O2C algorithm Score properties Two important properties Decomposability n � ( Global ) Score ( B , D ) = ( local ) score ( X i , pa i ) i =1 Score equivalence If two BN B 1 and B 2 are Markov equivalent then S ( B 1 , D ) = S ( B 2 , D ) Montassar Ben Messaoud, Mouna Ben Ishak, Philippe Leray, Nahla Ben Amor Apprentissage RB et Ontologies 9/49

Bayesian network Structure Learning Causal Bayesian Networks SemCaDo algorithm O2C algorithm Heuristic exploration of search space Search space and heuristics B space restriction to tree space : Chow&Liu, MWST DAG with node ordering : K2 algorithm greedy search genetic algorithms, ... E space greedy equivalence search Montassar Ben Messaoud, Mouna Ben Ishak, Philippe Leray, Nahla Ben Amor Apprentissage RB et Ontologies 10/49

Bayesian network Structure Learning Causal Bayesian Networks SemCaDo algorithm O2C algorithm Restriction to tree space Principle What is the best tree connecting all the nodes, i.e. maximizing a weight defined for each possible edge ? Answer : maximal weighted spanning tree (MWST) weight = mutual information [Chow & Liu, 1968] N ab N log N ab N � W ( X A , X B ) = N a . N . b a , b weight = any local score variation [Heckerman, 1994] W ( X A , X B ) = score ( X A , Pa ( X A ) = X B ) − score ( X A , ∅ ) Montassar Ben Messaoud, Mouna Ben Ishak, Philippe Leray, Nahla Ben Amor Apprentissage RB et Ontologies 11/49

Bayesian network Structure Learning Causal Bayesian Networks SemCaDo algorithm O2C algorithm Restriction to tree space Remarks MWST returns an undirected tree This undirected tree = CPDAG of all the directed tree with this skeleton Obtain a directed tree by (randomly) choosing one root and orienting the edges with a depth first search over this tree Montassar Ben Messaoud, Mouna Ben Ishak, Philippe Leray, Nahla Ben Amor Apprentissage RB et Ontologies 12/49

Bayesian network Structure Learning Causal Bayesian Networks SemCaDo algorithm O2C algorithm Example : obtained DAG vs. target one A S A S T L B T L B O O X D X D MWST can not discover cycles neither V-structures (tree space !) Montassar Ben Messaoud, Mouna Ben Ishak, Philippe Leray, Nahla Ben Amor Apprentissage RB et Ontologies 13/49

Bayesian network Structure Learning Causal Bayesian Networks SemCaDo algorithm O2C algorithm Greedy search Principle Exploration of the search space with traversal operators add edge invert edge delete edge and respect the DAG definition (no cycle) Exploration can begin from any given DAG Montassar Ben Messaoud, Mouna Ben Ishak, Philippe Leray, Nahla Ben Amor Apprentissage RB et Ontologies 15/49

Bayesian network Structure Learning Causal Bayesian Networks SemCaDo algorithm O2C algorithm Example : obtained DAG vs. target one A S A S T L B T L B O O X D X D start = empty graph. GS result = local optimum :-( Montassar Ben Messaoud, Mouna Ben Ishak, Philippe Leray, Nahla Ben Amor Apprentissage RB et Ontologies 16/49

Bayesian network Structure Learning Causal Bayesian Networks SemCaDo algorithm O2C algorithm Example : obtained DAG vs. target one A S A S T L B T L B O O X D X D start = MWST result. GS result is better Montassar Ben Messaoud, Mouna Ben Ishak, Philippe Leray, Nahla Ben Amor Apprentissage RB et Ontologies 16/49

Bayesian network Structure Learning Causal Bayesian Networks SemCaDo algorithm O2C algorithm What about changing our search space Preliminaries IC/PC result = CPDAG MWST result = CPDAG Score-based methods do not distinguish equivalent DAGs Search in E E = CPDAG space Better properties : YES 2 equivalent structures = 1 unique structure in E Better size : NO E size is quasi similar to DAG space asymptotic ratio is 3,7 : [Gillispie & Perlman, 2001] Montassar Ben Messaoud, Mouna Ben Ishak, Philippe Leray, Nahla Ben Amor Apprentissage RB et Ontologies 18/49

Bayesian network Structure Learning Causal Bayesian Networks SemCaDo algorithm O2C algorithm Greedy Equivalent Search Principe [Chickering, 2002] Greedy search in E Phase 1 : add edges until convergence Phase 2 : delete edges until convergence Montassar Ben Messaoud, Mouna Ben Ishak, Philippe Leray, Nahla Ben Amor Apprentissage RB et Ontologies 19/49

Bayesian network Structure Learning Causal Bayesian Networks SemCaDo algorithm O2C algorithm Add edge examples in E neighborhood of E 0 neighborhood of E 1 X 1 X 2 X 1 X 2 X 1 X 2 X 1 X 2 X 1 X 2 X 3 X 4 X 3 X 4 X 3 X 4 X 3 X 4 X 3 X 4 X 1 X 2 X 1 X 2 X 1 X 2 X 1 X 2 X 1 X 2 X 1 X 2 E 0 X 3 X 4 X 3 X 4 X 3 X 4 X 3 X 4 X 3 X 4 X 3 X 4 X 1 X 2 X 1 X 2 X 1 X 2 X 1 X 2 X 1 X 2 E 1 X 3 X 4 X 3 X 4 X 3 X 4 X 3 X 4 X 3 X 4 Montassar Ben Messaoud, Mouna Ben Ishak, Philippe Leray, Nahla Ben Amor Apprentissage RB et Ontologies 20/49

Bayesian network Structure Learning Causal Bayesian Networks SemCaDo algorithm O2C algorithm Score-based methods How to search a good BN ? Constraint-based methods BN = independence model ⇒ find CI in data in order to build the DAG Score-based methods BN = probabilistic model that must fit data as well as possible ⇒ search the DAG space in order to maximize a scoring/fitness function Hybrid methods Montassar Ben Messaoud, Mouna Ben Ishak, Philippe Leray, Nahla Ben Amor Apprentissage RB et Ontologies 21/49

Bayesian network Structure Learning Causal Bayesian Networks SemCaDo algorithm O2C algorithm Hybrid methods = local search methods Local search and global learning Search one local neighborhood for a given node T Reiterate for each T Learn the global structure with these local informations which neighborhood ? PC ( T ) : Parents and Children T (without distinction) MB ( T ) : Markov Blanket of T - Parents, children and spouses Montassar Ben Messaoud, Mouna Ben Ishak, Philippe Leray, Nahla Ben Amor Apprentissage RB et Ontologies 22/49

Bayesian network Structure Learning Causal Bayesian Networks SemCaDo algorithm O2C algorithm Local search identification Identification of MB(T) or PC(T) IAMB [Aliferis et al., 2002] MMPC [Tsamardinos et al., 2003], ... Hybrid structure learning algorithms MMHC [Tsamardinos et al., 2006] = MMPC + Greedy search back Montassar Ben Messaoud, Mouna Ben Ishak, Philippe Leray, Nahla Ben Amor Apprentissage RB et Ontologies 23/49

Bayesian network Structure Learning Causal Bayesian Networks SemCaDo algorithm O2C algorithm A BN is not a causal model Usual BN A → B does not imply direct causal relationship between A and B , Only edges from the CPDAG represent causal relationships ∗ Confusion When the DAG is given by an expert, this graph is very often causal When the DAG is learnt from data, no reason to be causal ! Causal BN Each A → B represents one direct causal relationship, i.e. A is the direct cause which generates B Montassar Ben Messaoud, Mouna Ben Ishak, Philippe Leray, Nahla Ben Amor Apprentissage RB et Ontologies 25/49

Bayesian network Structure Learning Causal Bayesian Networks SemCaDo algorithm O2C algorithm Intervention vs. Observation Probabilistic inference : we observe B = b , we compute P ( A | B = b ) Causal inference [Pearl 00]: we manipulate/intervene upon B : do ( B = b ) example with A → B P ( A | do ( B = b )) = P ( A ), P ( B | do ( A = a )) = P ( B | A = a ) example with A ← B P ( A | do ( B = b )) = P ( A | B = b ), P ( B | do ( A = a )) = P ( B ) Montassar Ben Messaoud, Mouna Ben Ishak, Philippe Leray, Nahla Ben Amor Apprentissage RB et Ontologies 26/49

Bayesian network Structure Learning Causal Bayesian Networks SemCaDo algorithm O2C algorithm Causal structure learning Usual situation : observational data whatever the method, the right result is the CPDAG partial determination of the causal structure How to find a full causal graph ? Use only experimental data, and decide at every step the more interesting experiment to realize ( active learning [Murphy, 2001], ...) Use only observational data, for a very specific distribution ( LiNGAM models [Hoyer et al., 2008]) Another idea: MyCaDo algorithm [Meganck et al., 2006] Use (already existing) observational data to find the CPDAG Complete the orientation with experimental data Montassar Ben Messaoud, Mouna Ben Ishak, Philippe Leray, Nahla Ben Amor Apprentissage RB et Ontologies 27/49

Bayesian network Structure Learning Causal Bayesian Networks SemCaDo algorithm O2C algorithm MyCaDo algorithm Données d'observation Système Algorithme d'apprentissage de structure Données d'un RB expérimentales Réseau Bayésien causal CPDAG Choix Réalisation Analyse de l'exp. de l'exp. des résultats Algorithme MyCaDo (1) Choice of the experiment = what variable M manipulate ? the one potentially orienting more edges by taking into account experiment/observation cost back Montassar Ben Messaoud, Mouna Ben Ishak, Philippe Leray, Nahla Ben Amor Apprentissage RB et Ontologies 28/49

Bayesian network Structure Learning Causal Bayesian Networks SemCaDo algorithm O2C algorithm MyCaDo algorithm Données d'observation Système Algorithme d'apprentissage de structure Données d'un RB expérimentales Réseau Bayésien causal CPDAG Choix Réalisation Analyse de l'exp. de l'exp. des résultats Algorithme MyCaDo (2) Experimentation do ( M = m ) for all possible values m observe all candidate variables C ( C – M ) back Montassar Ben Messaoud, Mouna Ben Ishak, Philippe Leray, Nahla Ben Amor Apprentissage RB et Ontologies 28/49

Bayesian network Structure Learning Causal Bayesian Networks SemCaDo algorithm O2C algorithm MyCaDo algorithm Données d'observation Système Algorithme d'apprentissage de structure Données d'un RB expérimentales Réseau Bayésien causal CPDAG Choix Réalisation Analyse de l'exp. de l'exp. des résultats Algorithme MyCaDo (3) Result analysis : P ( C | M ) (obs.) ≃ P ( C | do ( M )) (exp.) ? if equal C ← M else M ← C orient some other edges by applying specific rules back Montassar Ben Messaoud, Mouna Ben Ishak, Philippe Leray, Nahla Ben Amor Apprentissage RB et Ontologies 28/49

Bayesian network Structure Learning Causal Bayesian Networks SemCaDo algorithm O2C algorithm SemCaDo assumptions BN ⇔ Ontology general assumptions Nodes ⇐ ⇒ Concepts Random variables ⇐ ⇒ Concept attributes Causal dependencies ⇐ ⇒ Semantic causal relations Data ⇐ ⇒ Concept-attribute instances SemCaDo specific assumptions Causal relations concern concepts sharing the same semantic type Ontology continuity (evolution) Montassar Ben Messaoud, Mouna Ben Ishak, Philippe Leray, Nahla Ben Amor Apprentissage RB et Ontologies 30/49

Bayesian network Structure Learning Causal Bayesian Networks SemCaDo algorithm O2C algorithm Example Catastrophes is-a is-a Man-made Natural is-a is-a is-a is-a is-a is-a Fire Flood Earthquake Landslide Volcano Tsunami Causes Causes Causes Causes Causes Causes Our causal BN will represent the grey part of the ontology Montassar Ben Messaoud, Mouna Ben Ishak, Philippe Leray, Nahla Ben Amor Apprentissage RB et Ontologies 31/49

Bayesian network Structure Learning Causal Bayesian Networks SemCaDo algorithm O2C algorithm SemCaDo three main steps Domain ontology Inter. Data 1 st step 3 rd step Choice of experience Choice of experience Structure learning Ontology evolution Perform the experience Obs. CBN PDAG Analyse the results Data 2 nd step Causal discovery Enriched ontology Montassar Ben Messaoud, Mouna Ben Ishak, Philippe Leray, Nahla Ben Amor Apprentissage RB et Ontologies 32/49

Bayesian network Structure Learning Causal Bayesian Networks SemCaDo algorithm O2C algorithm 1 st step: initial BN structure learning Extraction of the causal relationships ( R ⌋ ) from the ontology Integration of these edge as constraints in the structure learning algorithm [De Campos & al., 2007] Continuity : these edges will not be ”questioned” during learning Interest Ontology helps in reducing the search task complexity Montassar Ben Messaoud, Mouna Ben Ishak, Philippe Leray, Nahla Ben Amor Apprentissage RB et Ontologies 33/49

Bayesian network Structure Learning Causal Bayesian Networks SemCaDo algorithm O2C algorithm 2 nd step: serendipitous causal discovery Experimentations are needed in order to find the causal orientation of some edges Our solution : MyCaDo [Meganck & al., 2006], iterative causal discovery process Adaptation to take into account ontological knowledge : Rada distance on H between one set of concepts and their most specific common subsumer Interest Ontology helps in potentially orienting the more unexpected (serendipitous) links Montassar Ben Messaoud, Mouna Ben Ishak, Philippe Leray, Nahla Ben Amor Apprentissage RB et Ontologies 34/49

Bayesian network Structure Learning Causal Bayesian Networks SemCaDo algorithm O2C algorithm 3 rd Step: Ontology evolution process [Stojanovic et al., 2002] Interest BN structure learning from data helps in discovering new relations in the ontology Montassar Ben Messaoud, Mouna Ben Ishak, Philippe Leray, Nahla Ben Amor Apprentissage RB et Ontologies 35/49

Bayesian network Structure Learning Causal Bayesian Networks SemCaDo algorithm O2C algorithm Experimental study (1) Benchmark : no existing real benchmark & system :-( BN graph : random generation (50 to 200 nodes) Ontology : Causal relationships : BN edges Hierarchy of concept : generation by clustering BN nodes Data is generated by using BN as a generative model Experimental protocol Hierarchy of concepts and 10% to 40% of existing causal relationships are given as inputs Semantic gain : cumulative Rada distance of the discovered relationships Montassar Ben Messaoud, Mouna Ben Ishak, Philippe Leray, Nahla Ben Amor Apprentissage RB et Ontologies 36/49

Int egration de connaissances ontologiques pour lapprentissage des - PowerPoint PPT Presentation

Bayesian networks Ontologies BNs and Ontologies Int egration de connaissances ontologiques pour lapprentissage des r eseaux bay esiens Montassar Ben Messaoud 12 , Mouna Ben Ishak 12 , Philippe Leray 2 , Nahla Ben Amor 1 1 LARODEC,

PRESENTATION OF THE ASSOCIATION 20 October 2016 changer connaissances et techniques sur les

BASTA B oost A pplied munition detection through S mart data in T egration and A I workflows

Dveloppem ent dun systm e m ixte innovant de Dveloppem ent dun systm e m ixte

PRESENTATION OF THE ASSOCIATION 8 March 2017 changer connaissances et techniques sur les routes

ISO TC67 WG10 ISO TC252 WG2 Normalisation internationale pour les installations et

Yes We Can Yes We Can Yes We Can Yes We Can From biomedical informatics to translational

Descripteurs divers niveaux de concepts pour la classification concepts pour la classification

Simulations Thermo-Hydro- Mcaniques pour le stockage profond des dchets nuclaires

pour le dveloppement ? Serge Tomasi, Directeur Adjoint Direction de la Coopration pour le

Solid Int egration avec le mat eriel sans utiliser daspirine K evin Ottens 26

M ISSION AND VALUES AVAID - Association de Volontaires pour lAide au Dveloppement,

Systme lectronique de monitoring de d donnes pour sportifs de haut niveau if d h i

Integrity Under Pressure Systmes intgrs pour conduites en pression Integrated systems for

OCCIware un cadre formel et ou2ll pour la ges2on de

GFC Missions GFC Missions Elaboration of Performance Tests Le Consulat 147 avenue Paul

codage pour ados CAMP DT 2018 Prsent par Axiom Academy Offert par lAcad mie

Unified Patent Court Mock Trial Paris April, 2 nd 2015 U.J.U.B. - Union pour la Juridiction

E XPLORATION DE DONN EES POUR L OPTIMISATION DE TRAJECTOIRES A ERIENNES C edric

Apport des modlisations ab initio pour la comprhension des proprits structurales et

In egalit es spectrales pour le contr ole des EDP lin eaires : groupe de Schr

M ethodologies and outils pour l evaluation des protocoles de transport dans les r

in developing countries Jol Cariolle Fondation pour les tudes et recherches sur le

De labstraction des modles de composants logiciels pour la programmation dapplications

Techniques et outils pour la vrification de systmes-sur-puces au niveau transactionnel