Onto Grid Paving the way for Knowledgeable Grid

Onto. Grid: Paving the way for Knowledgeable Grid Services and Systems www. ontogrid. net FP 6 -511513 KOPE: Knowledge-Oriented Provenance Environment Jose Manuel Gómez-Pérez, Francisco Javier García (i. SOCO), Rafael González (UPM), Chris Van Aart (Y’All) Monterey, July 26 th 2007 1

Motivation Provenance (Source: Uo. S micro-site) Provenance Pyramid (Source: my. Grid) § Goals q Increase understanding of process execution q Explain provenance in a way closer to how domain experts reason on a given problem § Problem Solving Methods [Mc. Dermott, 1988] 2

Problem Solving Methods (PSM) § PSM are knowledge templates that q. Establish and control the sequence of actions required to perform a task q. Define the kind of knowledge necessary at each task step q. Hierarchically specify how tasks decompose into subtasks down to the level of primitive actions q. Describe tasks at several levels of refinement § PSM are domain-independent q. PSM inputs and outputs modelled as generic roles q. Reusable across different domains 3

Problem Solving Methods Visualization Paradigm Decomposition view Interaction view Knowledge Flow view 4

Applications of Problem Solving Methods § Knowledge Engineering q Knowledge acquisition: Guidelines to acquire problem solving knowledge q Reasoning: Enable flexible reasoning by selecting methods during problem solving q Process analysis: Description of the main rationale of (reasoning) processes § Provenance Interpretation q Explain the results of queries on process documentation 5

Who defines Problem Solving Methods? § Ideally, collaboratively defined by a community of domain experts q. Canonical specifications of domain processes q. Agreed throughout the community q. Examples: • Regulations for good medical praxis • Diagnosis • Reasoning (Common. KADS) § Also possible: knowledge engineer with a little domain knowledge, e. g. population-based brain atlas 6

Provenance Interpretation Workflow 7

Semantic Resources Domain ontology PSM meta-model § Bridge: Explicit definition of mappings between domain and PSM entities § Refiner: Specification of task decomposition into subtasks Roles (catalogue task) 8

Annotation of Process Documentation PASOA interaction p-assertion Automatically annotated against the domain ontology during process execution <content> <items xmlns=""> <item 1> http: //www. ontogrid. net/#Atlas. Graphic. X </item 1> <item 2> http: //www. ontogrid. net/#Atlas. Graphic. Y </item 2> <item 3> http: //www. ontogrid. net/#Atlas. Graphic. Z </item 3> </items> </content> 9

Twig Matching Algorithm § twig_join(D, i(P), o(P)) is a boolean function which checks whether a twig exists that connects i(P) and o(P) in D, where: q q § § § P is a problem solving method i(P) is the set of input roles of P o(P) is the set of output roles of P D is the provenance DAG of the documented process, returned by a provenance query We consider P as an interpretation of a process if twig_join(D, i(P), o(P)) = true Bridges allow detecting occurrences of PSM roles in D Refiners allow applying the algorithm recursively across the PSM hierarchy 10

Twig Matching for Brain Atlas workflow and Catalogue Brain Atlas Provenance Data Flow Brain Atlas Workflow Prime Catalogue Method Knowledge Flow 11

Demo 12

Thanks for your attention! Jose Manuel Gómez-Pérez jmgomez@isoco. com #T +34 91 334 9778 #M +34 609 077 103 i. SOCO http: //www. isoco. com i. SOCO Barcelona +34 93 5677200 Edifici Testa A C/ Alcalde Barnils 64 -68 St. Cugat del Vallès 08190 Barcelona i. SOCO Madrid +34 91 3349797 C/Pedro de Valdivia, 10 28006 Madrid i. SOCO Valencia +34 96 3467143 Oficina 107 C/ Prof. Beltrán Báguena 4, 46009 Valencia 13