GEOSPATIAL SEMANTICS -- PROBLEMS AND PROJECTS
Upcoming SlideShare
Loading in...5
×
 

GEOSPATIAL SEMANTICS -- PROBLEMS AND PROJECTS

on

  • 364 views

Talk at UCSB in 2011. Lists some common challenges with respect to geo-semantics and solutions we worked between 2008 and 2011. also includes some ideas for future work.

Talk at UCSB in 2011. Lists some common challenges with respect to geo-semantics and solutions we worked between 2008 and 2011. also includes some ideas for future work.

Statistics

Views

Total Views
364
Slideshare-icon Views on SlideShare
364
Embed Views
0

Actions

Likes
0
Downloads
8
Comments
0

0 Embeds 0

No embeds

Accessibility

Categories

Upload Details

Uploaded via as Adobe PDF

Usage Rights

© All Rights Reserved

Report content

Flagged as inappropriate Flag as inappropriate
Flag as inappropriate

Select your reason for flagging this presentation as inappropriate.

Cancel
  • Full Name Full Name Comment goes here.
    Are you sure you want to
    Your message goes here
    Processing…
Post Comment
Edit your comment

    GEOSPATIAL SEMANTICS -- PROBLEMS AND PROJECTS GEOSPATIAL SEMANTICS -- PROBLEMS AND PROJECTS Presentation Transcript

    • MOTIVATION LINKING GEO-DATA GEO-ONTOLOGIES NEOGEOGRAPHY QUESTIONS GEOSPATIAL SEMANTICS PROBLEMS AND PROJECTS Krzysztof Janowicz GeoVISTA Center, Pennsylvania State University, USA March 10, 2011 GEOSPATIAL SEMANTICS – PROBLEMS AND PROJECTS KRZYSZTOF JANOWICZ
    • MOTIVATION LINKING GEO-DATA GEO-ONTOLOGIES NEOGEOGRAPHY QUESTIONS RETRIEVAL AND INTEGRATION RESEARCH STREAMS ANALOGY GEOSPATIAL SEMANTICS – PROBLEMS AND PROJECTS KRZYSZTOF JANOWICZ
    • MOTIVATION LINKING GEO-DATA GEO-ONTOLOGIES NEOGEOGRAPHY QUESTIONS RETRIEVAL AND INTEGRATION DIGITAL EARTH – RETRIEVAL AND INTEGRATION Digital Earth Geo-referenced Interconnected Multi-thematic Multi-perspective Multi-dimensional Knowledge engine GEOSPATIAL SEMANTICS – PROBLEMS AND PROJECTS KRZYSZTOF JANOWICZ
    • MOTIVATION LINKING GEO-DATA GEO-ONTOLOGIES NEOGEOGRAPHY QUESTIONS INTRODUCTION GEO-DATA AND SERVICES GEOSPATIAL SEMANTICS – PROBLEMS AND PROJECTS KRZYSZTOF JANOWICZ
    • MOTIVATION LINKING GEO-DATA GEO-ONTOLOGIES NEOGEOGRAPHY QUESTIONS INTRODUCTION LINKING DATA Data Silos Web services Databases Web pages hinder ad-hoc combination enforce data models limit re-usability GEOSPATIAL SEMANTICS – PROBLEMS AND PROJECTS KRZYSZTOF JANOWICZ
    • MOTIVATION LINKING GEO-DATA GEO-ONTOLOGIES NEOGEOGRAPHY QUESTIONS INTRODUCTION FROM LINKED DOCUMENTS TO LINKED DATA Use Uniform Resource Identifiers (URI) to identify entities, link them to other entities, encode information about these entities using the machine-understandable RDF, and make them available on the Web. GEOSPATIAL SEMANTICS – PROBLEMS AND PROJECTS KRZYSZTOF JANOWICZ
    • MOTIVATION LINKING GEO-DATA GEO-ONTOLOGIES NEOGEOGRAPHY QUESTIONS INTRODUCTION LINKED DATA FOR SANTA BARBARA Browsing information related to Santa Barbara using the Linked Data version of Wikipedia (DBpedia). GEOSPATIAL SEMANTICS – PROBLEMS AND PROJECTS KRZYSZTOF JANOWICZ
    • MOTIVATION LINKING GEO-DATA GEO-ONTOLOGIES NEOGEOGRAPHY QUESTIONS INTRODUCTION LINKED DATA – CREATING THE GLOBAL GRAPH Integration by searching equivalent types or same features in data sets. GEOSPATIAL SEMANTICS – PROBLEMS AND PROJECTS KRZYSZTOF JANOWICZ
    • MOTIVATION LINKING GEO-DATA GEO-ONTOLOGIES NEOGEOGRAPHY QUESTIONS PROBLEM – MOST GEO-DATA IS NOT ACCESSIBLE PROBLEM – MOST GEO-DATA IS NOT ACCESSIBLE GEOSPATIAL SEMANTICS – PROBLEMS AND PROJECTS KRZYSZTOF JANOWICZ
    • MOTIVATION LINKING GEO-DATA GEO-ONTOLOGIES NEOGEOGRAPHY QUESTIONS PROJECT – TRANSPARENT SEMANTIC ENABLEMENT PROJECT – TRANSPARENT SEMANTIC ENABLEMENT Connect both worlds without changing any standards Encapsulate SW technologies by OGC Web Processing Services (WPS) Encapsulate OGC services using RESTful proxies (e.g., RESTful SOS) GEOSPATIAL SEMANTICS – PROBLEMS AND PROJECTS KRZYSZTOF JANOWICZ
    • MOTIVATION LINKING GEO-DATA GEO-ONTOLOGIES NEOGEOGRAPHY QUESTIONS PROJECT – TRANSPARENT SEMANTIC ENABLEMENT USING SEMANTIC WEB REASONING FOR SENSOR DISCOVERY Ontologies and Semantic Web technologies can assist in discovering appropriate geo-sensors and observation data (Sensor Plug & Play). GEOSPATIAL SEMANTICS – PROBLEMS AND PROJECTS KRZYSZTOF JANOWICZ
    • MOTIVATION LINKING GEO-DATA GEO-ONTOLOGIES NEOGEOGRAPHY QUESTIONS PROJECT – TRANSPARENT SEMANTIC ENABLEMENT SEMANTIC SENSOR WEB ONTOLOGIES W3C XG SSN STIMULUS SENSOR OBSERVATION PATTERN For semantic integration and retrieval, Linked Data needs to be annotated with ontologies that describe the meaning of the used terms. GEOSPATIAL SEMANTICS – PROBLEMS AND PROJECTS KRZYSZTOF JANOWICZ
    • MOTIVATION LINKING GEO-DATA GEO-ONTOLOGIES NEOGEOGRAPHY QUESTIONS INTRODUCTION GEO-ONTOLOGIES AND HETEROGENEITY GEOSPATIAL SEMANTICS – PROBLEMS AND PROJECTS KRZYSZTOF JANOWICZ
    • MOTIVATION LINKING GEO-DATA GEO-ONTOLOGIES NEOGEOGRAPHY QUESTIONS INTRODUCTION ONTOLOGY – RESTRICTING INTERPRETATIONS STRUCTURE MATTERS GEOSPATIAL SEMANTICS – PROBLEMS AND PROJECTS KRZYSZTOF JANOWICZ
    • MOTIVATION LINKING GEO-DATA GEO-ONTOLOGIES NEOGEOGRAPHY QUESTIONS INTRODUCTION ONTOLOGY – RESTRICTING INTERPRETATIONS LOCATION MATTERS GEOSPATIAL SEMANTICS – PROBLEMS AND PROJECTS KRZYSZTOF JANOWICZ
    • MOTIVATION LINKING GEO-DATA GEO-ONTOLOGIES NEOGEOGRAPHY QUESTIONS INTRODUCTION ONTOLOGY – RESTRICTING INTERPRETATIONS TIME MATTERS GEOSPATIAL SEMANTICS – PROBLEMS AND PROJECTS KRZYSZTOF JANOWICZ
    • MOTIVATION LINKING GEO-DATA GEO-ONTOLOGIES NEOGEOGRAPHY QUESTIONS INTRODUCTION ONTOLOGY – RESTRICTING INTERPRETATIONS SOCIAL STRUCTURE MATTERS GEOSPATIAL SEMANTICS – PROBLEMS AND PROJECTS KRZYSZTOF JANOWICZ
    • MOTIVATION LINKING GEO-DATA GEO-ONTOLOGIES NEOGEOGRAPHY QUESTIONS INTRODUCTION AN ONTOLOGY OF POTHOLES Potholes are defined as cracks of more than 30mm depth [North East Somerset, UK] with a width of a ’large dinner plate’ (300mm) and the depth of a ’golf ball’ (40mm) [Gloucestershire, UK] with a width of a ’dinner plate’ (200mm) and a minimum depth of a ’fist’ (40mm) [Worcestershire, UK] if their depth is ’a pound coin and a 1p coin side by side’ [Coventry] =⇒ Due to a severe winter (T) millions of potholes need to be repaired by the local councils (S) that are legally responsible for the roads maintenance (A) within their administrative boundaries (S) . GEOSPATIAL SEMANTICS – PROBLEMS AND PROJECTS KRZYSZTOF JANOWICZ
    • MOTIVATION LINKING GEO-DATA GEO-ONTOLOGIES NEOGEOGRAPHY QUESTIONS INTRODUCTION ONTOLOGIES CANNOT FIX MEANING ACCORDING TO OUR ONTOLOGY THIS IS A POTHOLE GEOSPATIAL SEMANTICS – PROBLEMS AND PROJECTS KRZYSZTOF JANOWICZ
    • MOTIVATION LINKING GEO-DATA GEO-ONTOLOGIES NEOGEOGRAPHY QUESTIONS INTRODUCTION ONTOLOGIES CANNOT FIX MEANING REASONING: CONSTRUCTING BRIDGES CREATES POTHOLES GEOSPATIAL SEMANTICS – PROBLEMS AND PROJECTS KRZYSZTOF JANOWICZ
    • MOTIVATION LINKING GEO-DATA GEO-ONTOLOGIES NEOGEOGRAPHY QUESTIONS PROBLEM – INTEROPERABILITY VS. HETEROGENEITY PROBLEM – INTEROPERABILITY VS. HETEROGENEITY Too generic ontologies fail to restrict meaning; too restrictive ontologies exclude multi-perspective and local definitions. GEOSPATIAL SEMANTICS – PROBLEMS AND PROJECTS KRZYSZTOF JANOWICZ
    • MOTIVATION LINKING GEO-DATA GEO-ONTOLOGIES NEOGEOGRAPHY QUESTIONS PROJECT – MICROTHEORIES PROJECT – HANDLING SEMANTIC HETEROGENEITY BY MICROTHEORIES Employ space and time as fundamental structuring principles for the development of ontologies. Similar to Couclelis’ semantic contraction (without pre-defined levels). GEOSPATIAL SEMANTICS – PROBLEMS AND PROJECTS KRZYSZTOF JANOWICZ
    • MOTIVATION LINKING GEO-DATA GEO-ONTOLOGIES NEOGEOGRAPHY QUESTIONS PROJECT – MICROTHEORIES INSPIRE EXAMPLE– COMPUTING A COMMON GROUND INSPIRE Watercourse: ’A natural or man-made flowing water course or stream.’ Instead of a top-down ontology for the EU that violates local definitions, compute a top-level based on local definitions of the member states. *)INSPIRE: Infrastructure for Spatial Information in the European Community GEOSPATIAL SEMANTICS – PROBLEMS AND PROJECTS KRZYSZTOF JANOWICZ
    • MOTIVATION LINKING GEO-DATA GEO-ONTOLOGIES NEOGEOGRAPHY QUESTIONS PROJECT – MICROTHEORIES INSPIRE EXAMPLE– LCS AND SIMILARITY Germany and Spain are administratively contained in Europe. Hence, the European definition has to satisfy the local definitions. Least Common Subsumer (LCS) reasoning to compute a common EU definition. Similarity reasoning to ensure that the result does not end up as the compromise of nobody. GEOSPATIAL SEMANTICS – PROBLEMS AND PROJECTS KRZYSZTOF JANOWICZ
    • MOTIVATION LINKING GEO-DATA GEO-ONTOLOGIES NEOGEOGRAPHY QUESTIONS PROJECT – MICROTHEORIES PRESERVING LOCAL CONCEPTUALIZATIONS The INSPIRE definition is too restrictive GEOSPATIAL SEMANTICS – PROBLEMS AND PROJECTS KRZYSZTOF JANOWICZ
    • MOTIVATION LINKING GEO-DATA GEO-ONTOLOGIES NEOGEOGRAPHY QUESTIONS INTRODUCTION INCORPORATING VGI GEOSPATIAL SEMANTICS – PROBLEMS AND PROJECTS KRZYSZTOF JANOWICZ
    • MOTIVATION LINKING GEO-DATA GEO-ONTOLOGIES NEOGEOGRAPHY QUESTIONS INTRODUCTION SANTA BARBARA IN VGI "[Various VGI sources and duplication] make [the] interpretation of a situation more complicated as multiple sources can provide conflicting versions of the built and natural environments." (Zook et al; 2010) "[...] an important item [...] is the formalization of rules that permit contributed geographic information to be assessed against its geographic context, and [...] tools that would implement these rules." (Goodchild and Glennon; 2010) GEOSPATIAL SEMANTICS – PROBLEMS AND PROJECTS KRZYSZTOF JANOWICZ
    • MOTIVATION LINKING GEO-DATA GEO-ONTOLOGIES NEOGEOGRAPHY QUESTIONS INTRODUCTION LINKEDGEODATA SCHOOLS IN SANTA BARBARA School ≡ University; ∃x(School(x) ∧ ¬ManMade(x)) GEOSPATIAL SEMANTICS – PROBLEMS AND PROJECTS KRZYSZTOF JANOWICZ
    • MOTIVATION LINKING GEO-DATA GEO-ONTOLOGIES NEOGEOGRAPHY QUESTIONS INTRODUCTION LINKEDGEODATA SCHOOLS IN NEW YORK ∃x(School(x) ∧ University(x))? GEOSPATIAL SEMANTICS – PROBLEMS AND PROJECTS KRZYSZTOF JANOWICZ
    • MOTIVATION LINKING GEO-DATA GEO-ONTOLOGIES NEOGEOGRAPHY QUESTIONS INTRODUCTION WIKIMAPIA SCHOOLS IN SANTA BARBARA ∃x(School(x) ∧ University(x)); ∃x(University(x) ∧ NightClub(x)) GEOSPATIAL SEMANTICS – PROBLEMS AND PROJECTS KRZYSZTOF JANOWICZ
    • MOTIVATION LINKING GEO-DATA GEO-ONTOLOGIES NEOGEOGRAPHY QUESTIONS PROBLEM – INTEGRATION BY CLASS EQUIVALENCE PROBLEM – INTEGRATION BY CLASS EQUIVALENCE LGDSB : School ≡ LGDNY : School ≡ WikimapiaSB : School? GEOSPATIAL SEMANTICS – PROBLEMS AND PROJECTS KRZYSZTOF JANOWICZ
    • MOTIVATION LINKING GEO-DATA GEO-ONTOLOGIES NEOGEOGRAPHY QUESTIONS PROJECT – SEMANTIC SIGNATURES PROJECT – SEMANTIC SIGNATURES Bottom-up ontology personalization by Semantic Signatures GEOSPATIAL SEMANTICS – PROBLEMS AND PROJECTS KRZYSZTOF JANOWICZ
    • MOTIVATION LINKING GEO-DATA GEO-ONTOLOGIES NEOGEOGRAPHY QUESTIONS PROJECT – SEMANTIC SIGNATURES INTRODUCING SEMANTIC SIGNATURES Figures by ASTER Spectral Library at http://speclib.jpl.nasa.gov Combine numerical (statistical) models and data with ontologies to derive local (personal) primitives (reifications). Analogy to Spectral Signatures used in Remote Sensing Multiple bands → multiple Semantic Signatures GEOSPATIAL SEMANTICS – PROBLEMS AND PROJECTS KRZYSZTOF JANOWICZ
    • MOTIVATION LINKING GEO-DATA GEO-ONTOLOGIES NEOGEOGRAPHY QUESTIONS PROJECT – SEMANTIC SIGNATURES SEMANTIC SIGNATURES – LOCATION-BASED SOCIAL NETWORKS Locations types and log-in patterns from Whrrl, New York, USA Local Reifications: e.g., Weekend and Evening Used to automatically compute missing types of Whrrl POIs GEOSPATIAL SEMANTICS – PROBLEMS AND PROJECTS KRZYSZTOF JANOWICZ
    • MOTIVATION LINKING GEO-DATA GEO-ONTOLOGIES NEOGEOGRAPHY QUESTIONS PROJECT – SEMANTIC SIGNATURES SEMANTIC SIGNATURES – SPATIAL-SEMANTIC-PATTERNS POIs plotted by similarity to bar and recycling in OSM data, London, UK Reifications: e.g., Uniform and Clumped (local signatures) Bars (and similar features) tend to clump together Recycling (and similar features) are rather uniformly distributed GEOSPATIAL SEMANTICS – PROBLEMS AND PROJECTS KRZYSZTOF JANOWICZ
    • MOTIVATION LINKING GEO-DATA GEO-ONTOLOGIES NEOGEOGRAPHY QUESTIONS PROJECT – SEMANTIC SIGNATURES SEMANTIC SIGNATURES – SPATIAL-SEMANTIC INTERACTION Dzero measures the likeliness of features of a certain type to co-occur within a specific semantic and spatial range. User support: generate recommendations, and clean up data based on type likelihood. ’How likely is a recycling center directly next to an existing one?’ GEOSPATIAL SEMANTICS – PROBLEMS AND PROJECTS KRZYSZTOF JANOWICZ
    • MOTIVATION LINKING GEO-DATA GEO-ONTOLOGIES NEOGEOGRAPHY QUESTIONS THANKS, QUESTIONS? THANKS, QUESTIONS? GEOSPATIAL SEMANTICS – PROBLEMS AND PROJECTS KRZYSZTOF JANOWICZ
    • MOTIVATION LINKING GEO-DATA GEO-ONTOLOGIES NEOGEOGRAPHY QUESTIONS THANKS, QUESTIONS? SOCIAL IMPLICATIONS OF SEMANTIC TECHNOLOGIES Different conceptualizations of geographic space may lead to conflicts. Can we use semantic technologies such as similarity reasoning to make the hidden differences explicit and guide the negotiation process. GEOSPATIAL SEMANTICS – PROBLEMS AND PROJECTS KRZYSZTOF JANOWICZ
    • MOTIVATION LINKING GEO-DATA GEO-ONTOLOGIES NEOGEOGRAPHY QUESTIONS THANKS, QUESTIONS? THANKS TO... [I am very thankful to the contributors of Wikimedia Commons where the pothole pictures were taken from. The icons and clip-arts are based on images from OpenClipArt.org. The LOD diagram was designed by Richard Cyganiak. The collage of rivers, forests, and estuaries has been composed by my former student Stephanie Duce. The Dzero plots and the Whrll log-in figures have been created by my students Christoph Mülligann and Mao Ye.] GEOSPATIAL SEMANTICS – PROBLEMS AND PROJECTS KRZYSZTOF JANOWICZ