Скачать презентацию Some examples of ontology-based applications Oscar Corcho Скачать презентацию Some examples of ontology-based applications Oscar Corcho

458724cb529cd9402f97fdcd4b3ed582.ppt

  • Количество слайдов: 44

Some examples of ontology-based applications Oscar Corcho / Asunción Gómez-Pérez {ocorcho, asun}@fi. upm. es Some examples of ontology-based applications Oscar Corcho / Asunción Gómez-Pérez {ocorcho, asun}@fi. upm. es http: //www. oeg-upm. net/ Ontological Engineering Group Laboratorio de Inteligencia Artificial Facultad de Informática Universidad Politécnica de Madrid Campus de Montegancedo sn, 28660 Boadilla del Monte, Madrid, Spain 1

Common features in these domains • A large amount of information • Need to Common features in these domains • A large amount of information • Need to harmonize heterogeneous information sources • • • Different formats (documents, databases, etc. ) Different terms and structures Implicit knowledge hidden inside the application code 2

Procesamiento de información de satélites 28 Procesamiento de información de satélites 28

Satellite Use Case: Technical issues Space Segment Ground Segment SATELLITE FILES: DMOP files Product Satellite Use Case: Technical issues Space Segment Ground Segment SATELLITE FILES: DMOP files Product files 30

Satellite Use Case: Technical issues • Comparison between planning and product generation: DMOP_File#(n+1) DMOP#(n+1)_ Satellite Use Case: Technical issues • Comparison between planning and product generation: DMOP_File#(n+1) DMOP#(n+1)_ Start. Time File DMOP_File#n(Stop. Time) DMOP_File#n(Start. Time) Instr#1 planning Instr#n (RA_2) planning . . . DMOP_er (ORBIT_NUMBER, ELAPSED_TIME) DURATION Instr#n(RA_2) Product Generation . . . PRODUCT_FILE Start_time (SENSING_START) . . . PRODUCT_data_gap PRODUCT_FILE Stop_time (SENSING_STOP) RA 2_CAL_1 P Start_time (SENSING_START) 31 RA 2_CAL_1 P Stop_time (SENSING_STOP)

Satellite Use Case: Technical issues 32 Satellite Use Case: Technical issues 32

Satellite Use Case: Technical issues Satellite files: • DMOP (PLANNING) FILES FILE ; DMOP Satellite Use Case: Technical issues Satellite files: • DMOP (PLANNING) FILES FILE ; DMOP (generated by FOS Mission Planning System) RECORD fhr FILENAME="DMOP_SOF__VFOS 20060124_103709_0000_00001215_20060131_0140 48_20060202_035846. N 1" DESTINATION="PDCC" PHASE_START=2 CYCLE_START=44 REL_START_ORBIT=404 ABS_START_ORBIT=20498 RECORD ID RECORD parameters ENDRECORD fhr. . . . RECORD dmop_er_gen_part RECORD gen_event_params EVENT_TYPE=RA 2_MEA EVENT_ID="RA 2_MEA_000002063" NB_EVENT_PR 1=1 NB_EVENT_PR 3=0 ORBIT_NUMBER=20521 ELAPSED_TIME=623635 DURATION=41627862 ENDRECORD gen_event_params ENDRECORD dmop_er ENDLIST all_dmop_er ENDFILE 34 RECORD parameters corresponding to other RECORD structure.

Satellite Use Case: Technical issues Satellite Ontology (General view) 35 Satellite Use Case: Technical issues Satellite Ontology (General view) 35

Satellite Use Case: Technical issues Satellite Ontology (Hierarchies) 36 Satellite Use Case: Technical issues Satellite Ontology (Hierarchies) 36

Satellite Use Case: Technical issues • Planning (DMOP) RECORD parameters 37 Satellite Use Case: Technical issues • Planning (DMOP) RECORD parameters 37

Satellite Use Case: Technical issues Satellite files • PRODUCT FILES PRODUCT= Satellite Use Case: Technical issues Satellite files • PRODUCT FILES PRODUCT="RA 2_MW__1 PNPDE 20060131_231554_000061672044_00416_20510_0181. N 1" PROC_STAGE=N REF_DOC="PO-RS-MDA-GS-2009_3/M " SENSING_START="31 -JAN-2006 23: 15: 54. 654195" SENSING_STOP="01 -FEB-2006 00: 58: 41. 702319" PHASE=2 CYCLE=+044 REL_ORBIT=+00416 ABS_ORBIT=+20510 STATE_VECTOR_TIME="31 -JAN-2006 23: 28: 36. 484942" DELTA_UT 1=+. 323875 X_POSITION=+6637305. 306 Y_POSITION=-2700075. 034 Z_POSITION=+0000000. 000 X_VELOCITY=-0622. 619862 Y_VELOCITY=-1507. 628845 Z_VELOCITY=+7377. 140620 PRODUCT_ERR=0 TOT_SIZE=+00000087159984 SPH_SIZE=+0000006975 NUM_DSD=+000019 DSD_SIZE=+0000000280 NUM_DATA_SETS=+000004 38 Parameters to be transformed Parameters NOT needed to be transformed at the moment Parameters to be transformed at this moment

Satellite Use Case: Technical issues Satellite files: Annotated DMOP (PLANNING) FILES <? xml version='1. Satellite Use Case: Technical issues Satellite files: Annotated DMOP (PLANNING) FILES MS " GOM_OCC_0000541299" 53000 20552 2452293 [. . . ] 39

Satellite Use Case: Technical issues Satellite files: PRODUCT filename • Namefile (Product): RA 2_MW__1 Satellite Use Case: Technical issues Satellite files: PRODUCT filename • Namefile (Product): RA 2_MW__1 PNPDK 20060201_120535_000000062044_004 24_20518_0349. N 1" Corresponds to: 40

Satellite Use Case: Technical issues Satellite files: XMLed PRODUCT FILE <PRODUCT> Satellite Use Case: Technical issues Satellite files: XMLed PRODUCT FILE "RA 2_MW__1 PNPDE 20060131_231554_000061672044_00416_20510_0181. N 1“ N " PO-RS-MDA-GS-2009_3/M " " PDHS-E " " PDHS-E" " 01 -FEB-2006 01: 22: 48. 232601" " RA 2/5. 02 " " 31 -JAN-2006 23: 15: 54. 654195" " 01 -FEB-2006 00: 58: 41. 702319" CYCLE=+044 --> REL_ORBIT=+00416 --> ABS_ORBIT=+20510 --> STATE_VECTOR_TIME="31 -JAN-2006 23: 28: 36. 484942" --> DELTA_UT 1=+. 323875 --> X_POSITION=+6637305. 306 --> Y_POSITION=-2700075. 034 --> Z_POSITION=+0000000. 000 --> X_VELOCITY=-0622. 619862 --> Y_VELOCITY=-1507. 628845 --> Z_VELOCITY=+7377. 140620 --> 0 +00000087159984 +0000006975 +000019 +0000000280 +000004 41 Parameters NOT needed to be transformed at this moment. !!! FUTURE SCALABILITY IMPROVEMENT !!

Satellite Use Case: Technical issues Satellite files: Annotated PRODUCT FILE [. . . ] Satellite Use Case: Technical issues Satellite files: Annotated PRODUCT FILE [. . . ] " RA 2_MW__1 PNPDK 20060201_120535_000044792044_00424_20518_0334. N 1" 192110735 192115215 " RA 2_MW__1 PNPDK 20060202_160340_000058672044_00441_20535_0344. N 1" 192211420 192217287 [. . . ] 42

Satellite Use Case: Technical issues Satellite files (Metadata Queries): // Use to get a Satellite Use Case: Technical issues Satellite files (Metadata Queries): // Use to get a proxy class for Metadata. Service private java. lang. String Metadata. Service_address = "http: //195. 134. 67. 205: 8080/Atlas. Service/services/Metadata. Service"; public java. lang. String get. Metadata. Service. Address() { return Metadata. Service_address; } […] public eu. ist. ontogrid. ontokit. Metadata. Service get. Metadata. Service() throws javax. xml. rpc. Service. Exception { java. net. URL endpoint; try { endpoint = new java. net. URL(Metadata. Service_address); } catch (java. net. Malformed. URLException e) { throw new javax. xml. rpc. Service. Exception(e); } return get. Metadata. Service(endpoint); } […] public static void main(String[] args) { Metadata. Service. Proxy proxy = new Metadata. Service. Proxy(); String query 1 ="SELECT X FROM {X}kb: instrument_mode_id{Y} WHERE Y="STB" USING NAMESPACE kb=&http: //protege. stanford. edu/kb#"; String query 2 = "SELECT Z FROM {X}kb: plan_file_name{Y}, {Y}kb: file_id{Z}, {Y}kb: start_time{T 1}, {Y}kb: stop_time{T 2} WHERE T 1>192067200 AND T 1< 197247599 OR T 2 > 192067200 AND T 2<197247599 USING NAMESPACE kb=&http: //protege. stanford. edu/kb#"; try { System. out. println("submitting test query"); String result = proxy. query(query 2); System. out. println(result); Atlas. Result. Set results = new Atlas. Result. Set(result); 43

Satellite Use Case: Technical issues 44 Satellite Use Case: Technical issues 44

Generar organizaciones virtuales Precio? Garantia? Disponibilidad? Coche de sustitución ! 45 Generar organizaciones virtuales Precio? Garantia? Disponibilidad? Coche de sustitución ! 45

Business Case 1: Car Repair Business Case • Context: – Repair damaged cars – Business Case 1: Car Repair Business Case • Context: – Repair damaged cars – Negotiation between insurance and repair company • Speed, Price, Quality • Method of repair, Selection of material, Paint, Coalition • Now: – negotiation by hand – long term (yearly) • Challenge: – Automated negotiation – short term (every claim) – Include SLA 46

Business Case 2: Car. Fraud. Grid • Situation: • A lot of tricks to Business Case 2: Car. Fraud. Grid • Situation: • A lot of tricks to get money from insurance companies • Now: • Ad hoc manual techniques • Only pattern search on local or national scale • Most tricks found on accident • Challenge: • Automated fraud detection 47

Detección del fraude en seguros del coche 48 Detección del fraude en seguros del coche 48

Known trick: Berliner Model • Trick: • Buy damaged expensive car • Change some Known trick: Berliner Model • Trick: • Buy damaged expensive car • Change some features • Have stolen cars have accidents with it • Claim money from insurance company of stolen car • Search for: • Similar cars combined with similar situations combined with similar participants • National / International scale 49

Domain model • • Every insurance company uses its own database/domain model. Every claim Domain model • • Every insurance company uses its own database/domain model. Every claim database contains in some form important data about: * cars * situation Situation To find evidence we will look in claim history based on the current claim. • We look at car for: Car § * place of damage (angle of impact) § * description of accident • • • § * time of accident § * accident location § * price of damage § * damaged objects § * witnesses ? * brand, e. g. Peugeot * model, e. g. 307 * type, e. g. SW * mileage * license plate * owner * color * chassisnumber * constructionyear * countryofregistration 50

Conceptual Architecture 51 Conceptual Architecture 51

Portal Web Semántico: ODESe. W Ontology-based application that automatically generates and manages knowledge portals Portal Web Semántico: ODESe. W Ontology-based application that automatically generates and manages knowledge portals for intranets and extranets FP 6 -002006 1. 2. http: //www. agentlink. org/ Semantic Driven Permission-based IST-2001 -34373 http: //www. esperonto. net/ 3. User Oriented 4. Interoperate 5. Synchronization with the ontologies http: //kw. dia. fi. upm. es/w 3 c/ FP 6 -507482 http: //knowledgeweb. semanticweb. org/ FP 6 -027595 FP 6 -511513 http: //www. ontogrid. net/ 54 http: //www. oeg-upm. net/

Interoperability and management of Semantic portals Semantic Driven Permission-based Portal Administrators Ontologies and Software Interoperability and management of Semantic portals Semantic Driven Permission-based Portal Administrators Ontologies and Software O 1 User Oriented O 2 Oi Oj Extranet Users Agents External resources 55

Ontologías para la Gestión de Proyectos has associated event / ha s as has Ontologías para la Gestión de Proyectos has associated event / ha s as has asso soci ated ciate has / is associated with d/i s ass ocia ted is author of / has author Person Ontology is c has ontac bel con t pe ong tac rson s to t pe / is rso / for n me db y has contact person Documentation Ontology Organization Ontology 56 . Q. A ner as er art h tn p by r ant ated ing pa or icip ner uth art e a p is g ad has / le es has rat e gen is associated with r in Event Ontology has associated / leads is involved in / has p leader works in / has p p has person leader is de live in r de ks by ea or r l / w ped lo cto ner ra ve nt part de co is as ved n / / h ol s i ds inv pate i lea s ha rtic pa is WP leader / has person leader Project Ontology

Ontología de Documentos Documentation Additional Documentation Management Documentation Technical Documentation Publication Templates Manual Article Ontología de Documentos Documentation Additional Documentation Management Documentation Technical Documentation Publication Templates Manual Article Slides Thesis Master Thesis Deliverable Book . . . Agenda Cost Statement EC Templates Fax Mail Minutes. . . 57 Periodic Report Project Proposal . . . Ph. D Thesis

Ontología de Eventos Event International Conference Review International Workshop EPMB Meeting Management Project Meeting Ontología de Eventos Event International Conference Review International Workshop EPMB Meeting Management Project Meeting KW Area Meeting KW Plennary Meeting Research Area Meeting Education Area Meeting PMB Meeting Industry Area Meeting 58

Relaciones entre personas, proyectos y documentos has associated WP workload Milestone has participant with Relaciones entre personas, proyectos y documentos has associated WP workload Milestone has participant with workload Workpackage is made up of is deliver in has Task leads works in is WP leader is involved in is author of Person has contact person 59 Deliverable

Relaciones entre personas, proyectos y documentos has associated WP workload Milestone has participant with Relaciones entre personas, proyectos y documentos has associated WP workload Milestone has participant with workload Workpackage is made up of is deliver in has Project Ontology Task leads works in Documentation Ontology is involved in Person Ontology is WP leader is author of Person has contact person 60 Deliverable

Proceso de Generación del Portal Semántico 1. Construir las ontologías 2. Importar las ontologías Proceso de Generación del Portal Semántico 1. Construir las ontologías 2. Importar las ontologías en ODESe. W 3. Crear tipología de usuarios y usuarios concretos 4. Definición de permisos para los distintos grupos de usuarios 5. Definir la visualización personalizada para cada grupo de usuario 61

http: //knowledgeweb. semanticweb. org/ Descripción del proyecto Otros links a información útil Acceso directo http: //knowledgeweb. semanticweb. org/ Descripción del proyecto Otros links a información útil Acceso directo a las instancias de los conceptos más importantes 62

External User • Visuliazation – Project description • • Objectives Workpackages Task Management Board External User • Visuliazation – Project description • • Objectives Workpackages Task Management Board – List of deliverables • Description • Download the public documents – – List of partnets List of participants List of publications Calendar of events 63

Visualización basada en Semántica Workpackage has associated Deliverable 64 Visualización basada en Semántica Workpackage has associated Deliverable 64

Visualización basada en semántica Workpackage has associated Deliverable is generated by has Q. A. Visualización basada en semántica Workpackage has associated Deliverable is generated by has Q. A. partner Organization 65

Visualización de la instancia Tipos de atributos Nombres de los atributos Atributos Nombres Relaciones Visualización de la instancia Tipos de atributos Nombres de los atributos Atributos Nombres Relaciones de las relaciones Rango de las relaciones 66

KW Users • Kweb members Insert/update new information: – – Events Upload deliverables Personal KW Users • Kweb members Insert/update new information: – – Events Upload deliverables Personal information Publications • No access to – effort reports – management documentation 67

Edición de los contenidos del portal (Atributos) 68 Edición de los contenidos del portal (Atributos) 68

Edición de los contenidos del portal (Relaciones) Inserted values Potencial values 69 Edición de los contenidos del portal (Relaciones) Inserted values Potencial values 69

Contenidos del portal Knowledge. Web Ontologies Ontology Instances 70 Contenidos del portal Knowledge. Web Ontologies Ontology Instances 70

Managing director • Monitoring progress report – Area overviews – WP progress reports • Managing director • Monitoring progress report – Area overviews – WP progress reports • Effort report – Kweb year effort report – Partner effort report • Progress report draft generation 71

Progress Report Draft Generation HTML Format of the draft of the Progress Report WP Progress Report Draft Generation HTML Format of the draft of the Progress Report WP Progress & Effort Report Form MSWord Format of the draft of the Progress Report 72

Reporting performed by Organization performed by is workload of Annual Period workpackage workload Annual Reporting performed by Organization performed by is workload of Annual Period workpackage workload Annual Effort is workload on workpackage Workpackage performed in has associated Organization Deliverable. Contractual Date of Delivery [Annual Period. Start date, Annual. Period. End date] Annual Period 73