Presentation is loading. Please wait.

Presentation is loading. Please wait.

Cyprus Map Through Ontologies CMPE 583 - Web Semantics Project Prepared by Gizem OLGU Ali TÜZEL.

Similar presentations


Presentation on theme: "Cyprus Map Through Ontologies CMPE 583 - Web Semantics Project Prepared by Gizem OLGU Ali TÜZEL."— Presentation transcript:

1 Cyprus Map Through Ontologies CMPE 583 - Web Semantics Project Prepared by Gizem OLGU Ali TÜZEL

2 Cyprus Map Through Ontologies2 Gizem OLGU and Ali TUZEL Outline Introduction Cyprus Map Ontologies  CyprusMapOnto_vGizem  CyprusMapOnto_vAli Ontology Combining  Prompt Approach  Similarity Approach Demo  Protege Demo  MATLAB Demo Conlusions & Future Works

3 Cyprus Map Through Ontologies3 Gizem OLGU and Ali TUZEL Introduction Semantic Web  Is an extension of the current web, in which information on the web can processed by machines, in other words data becomes in machine-readable form on the web. Ontology  Is content theories about sort of object, properties of the object and relation between objects that are possible in a specified domain of knowledge.  Ontology has identified with computer as a machine-readable vocabulary. The OWL (Web Ontology Language)

4 Cyprus Map Through Ontologies4 Gizem OLGU and Ali TUZEL Introduction - Case The Purpose  Combining the different ontologies using similarity approach. The Subject  Proposed case study is based on Cyprus Map. Two different Cyprus Map ontologies  CyprusMapOnto_vGizem  CyprusMapOnto_vAli Ontology Combining  Different similarity measures combined to find mapping candidates between two ontologies.

5 Cyprus Map Through Ontologies5 Gizem OLGU and Ali TUZEL Introduction - Case Rules  Different rules are determined in calculating similarity measures and mapping the entities.  We have tried to obtain the most appropriate rule combination for our ontology combination case. Comparison  Prompt versus Similarity Approach

6 Cyprus Map Through Ontologies6 Gizem OLGU and Ali TUZEL Introduction - Platforms Protégé 2.1.2 Protégé is an ontology editor it provides graphical user interface (GUI) that models classes (domain concepts) and their attributes and relationships in ontology.  OWL plug-in  PROMPT plug-in Prompt is an ontology merging method for merge ontologies. MatLab  To perform our similarity approach algorithm.

7 Cyprus Map Through Ontologies7 Gizem OLGU and Ali TUZEL Cyprus Map Ontologies Coverage  Location oriented Coordinates Connections Location Specific Properties  No graphical information

8 Cyprus Map Through Ontologies8 Gizem OLGU and Ali TUZEL CyprusMapOnto_vGizem

9 Cyprus Map Through Ontologies9 Gizem OLGU and Ali TUZEL CyprusMapOnto_vGizem Properties

10 Cyprus Map Through Ontologies10 Gizem OLGU and Ali TUZEL CyprusMapOnto_vGizem Relations

11 Cyprus Map Through Ontologies11 Gizem OLGU and Ali TUZEL CyprusMapOnto_vAli location

12 Cyprus Map Through Ontologies12 Gizem OLGU and Ali TUZEL CyprusMapOnto_vAli Properties location Example 1: city Example 2:

13 Cyprus Map Through Ontologies13 Gizem OLGU and Ali TUZEL CyprusMapOnto_vAli Relations historicalPlace historicalPlaceType isHistoricalPlaceTypeOf Example 1: cityvillage hasConnectionTo Example 2:

14 Cyprus Map Through Ontologies14 Gizem OLGU and Ali TUZEL Ontology Combination Prompt Approach Similarity Approach

15 Cyprus Map Through Ontologies15 Gizem OLGU and Ali TUZEL PROMPT Approach PROMPT starts with the similarity matches for the initial comparison. After the user selects an operation to perform PROMPT not only points to the places where changes should be made, but also presents a list of actions to the user. Summarizing, PROMPT gives iterative suggestions for concept merges and changes, based on linguistic and structural knowledge, and it points the user to possible effects of these changes.

16 Cyprus Map Through Ontologies16 Gizem OLGU and Ali TUZEL PROMPT Results CyprusMapOnto_vAli: Class IDCyprusMapOnto_vGizem: Class ID Location mountainMountain City Lake Village Bay Beach Cape

17 Cyprus Map Through Ontologies17 Gizem OLGU and Ali TUZEL Similarity Approach Similarity Measure i: Ontology index j: Entity index e ij : j th entity of i th ontology k: Number of rules w k : Weight of the specific k th similarity rule (%0) 0 < w k < 1 (%100) (No similarity) 0 < sim < 1 (Identical) Decision Treshold

18 Cyprus Map Through Ontologies18 Gizem OLGU and Ali TUZEL Similarity Rules RuleExplanationEstimated weight Rule 1Common class IDs rdf:ID %50 Rule 2Common properties rdf:ID, rdfs:domain %20 Rule 3Common sub classes rdfs:subClassOf, rdf:resource, rdf:about %10 Rule 4Common super classes rdfs:subClassOf, rdf:resource, rdf:about %10 Rule 5Common instances rdf:ID, property value %10

19 Cyprus Map Through Ontologies19 Gizem OLGU and Ali TUZEL Similarity Calculation Using MATLAB RETURNING VALUES Class Pairs Having Similarity Over Treshold Matched Class Similarity Measures PARAMETERS Classes Class properties Sub classes Class relations Rule weights Treshold

20 Cyprus Map Through Ontologies20 Gizem OLGU and Ali TUZEL Similarity Results Similarity MeasureCyprusMapOnto_vAliCyprusMapOnto_vGizem 0.62109locationLocation 0.50836mountainMountain 0.5072cityCity 0.50697lakeLake 0.50697villageVillage 0.50396bayBay 0.50396beachBeach 0.50396capeCape

21 Cyprus Map Through Ontologies21 Gizem OLGU and Ali TUZEL Demo Prompt approach using Protege Similarity approach using MATLAB

22 Cyprus Map Through Ontologies22 Gizem OLGU and Ali TUZEL Conclusions Two different ontologies can be combined using different approaches Prompt versus Similarity  Comparable  Future Works RDB versus Ontology  Data: Unstructured

23 Cyprus Map Through Ontologies23 Gizem OLGU and Ali TUZEL Future Works Optimization of rule weights  Training Stage Smart string comparison  Thesaurus Synonyms Singular – Plural etc.

24 Cyprus Map Through Ontologies24 Gizem OLGU and Ali TUZEL References 1. Marc Ehrig and York Sure, “Ontology Mapping - An Integrated Approach”, ESWS 2004, vol. 3053/2004, p. 76-91, April 21, 2004. Available at: http://www.aifb.uni- karlsruhe.de/WBS/ysu/publications/2004_esws_mapping.pdf.http://www.aifb.uni- karlsruhe.de/WBS/ysu/publications/2004_esws_mapping.pdf 2. N. F. Noy & M. A. Musen, “PROMPT: Algorithm and Tool for Automated Ontology Merging and Alignment”, Seventeenth National Conference on Artificial Intelligence (AAAI-2000), Austin, TX,. 2000. Available at: http://smi-web.stanford.edu/pubs/SMI_Reports/SMI- 2000-0831.pdfhttp://smi-web.stanford.edu/pubs/SMI_Reports/SMI- 2000-0831.pdf 3. Otology Editor, Protégé 2.1.2, http://protege.stanford.eduhttp://protege.stanford.edu

25 Cyprus Map Through Ontologies25 Gizem OLGU and Ali TUZEL References 4. Macromedia Fireworks MX, http://www.macromedia.com/software/fireworks/ http://www.macromedia.com/software/fireworks/ 5. Bener Hakkı Hakeri, “Kıbrıs Türk Ansiklopedisi”, Book, vol. I, 1992. 6. Bener Hakkı Hakeri, “Kıbrıs Türk Ansiklopedisi”, Book, vol. II, 1992. 7. Zekai Altan, “Kuzey Kıbrısı Tanıyalım Gezelim Görelim”, Book, December 2000.

26 Thank You Any Questions??

27 Cyprus Map Through Ontologies27 Gizem OLGU and Ali TUZEL Question 1 Smart string comparison?

28 Cyprus Map Through Ontologies28 Gizem OLGU and Ali TUZEL Question 2 Standalone application instead of using MATLAB?

29 Cyprus Map Through Ontologies29 Gizem OLGU and Ali TUZEL


Download ppt "Cyprus Map Through Ontologies CMPE 583 - Web Semantics Project Prepared by Gizem OLGU Ali TÜZEL."

Similar presentations


Ads by Google