Semantic Assessment and Monitoring of Crowdsourced Geographic - - PowerPoint PPT Presentation

▶
semantic assessment and
SMART_READER_LITE
LIVE PREVIEW

Semantic Assessment and Monitoring of Crowdsourced Geographic - - PowerPoint PPT Presentation

Nov 21, 2022 340 likes •964 views

Semantic Assessment and Monitoring of Crowdsourced Geographic Information Hamish McNair, Paul Goodhue University of Canterbury Christchurch, New Zealand Outline Our research Project outline FOSS framework for the project

slide-1
SLIDE 1

Semantic Assessment and Monitoring of Crowdsourced Geographic Information

Hamish McNair, Paul Goodhue

University of Canterbury Christchurch, New Zealand

slide-2
SLIDE 2

Outline

  • Our research
  • Project outline
  • FOSS framework for the project
  • Crowdsourcing information
  • Determining Trust
  • Ontologies
  • Linked Data
  • Future direction & Conclusion
slide-3
SLIDE 3

Our Research

  • Trusting Crowdsourced Geographic Information

– Improving the trust of crowdsourced geographic information

  • Crowdsourcing Spatial Data Supply Chains

– Implications of trust beyond the capture of crowdsourced geographic information.

slide-4
SLIDE 4

Project – Fruit Trees

slide-5
SLIDE 5

Project – Fruit Trees

slide-6
SLIDE 6

Project – Fruit Trees

slide-7
SLIDE 7

INPUT TRUST RATING ONTOLOGY LINKED DATA OUTPUT

SPARQL RDFLib RDFLib Folium

slide-8
SLIDE 8

INPUT

slide-9
SLIDE 9

Crowdsourcing

User Interface WFS-T Data Server Database

slide-10
SLIDE 10

Data Model

slide-11
SLIDE 11

TRUST RATING

slide-12
SLIDE 12

Conceptual Trust Model

Intrinsic: Extrinsic:

Spatio-temporal Social Semantic

Components of CGI:

Spatial: Spatial comparison to neighbours based on rules about the CGI Temporal: Temporal comparison to neighbours based on rules about the CGI Spatial: Shape metrics of the geometry based on geometry type Temporal: Assessment of feature changelog or age of feature Assessment of the trust

  • f the author as

reviewed by the crowd, e.g. through Linus’ Law, peer reviews and Consensus Crowdsorucing Assessment of the author’s trust and likely influence on the trust

  • f the CGI, e.g. through

previous trust ratings or assessments of local knowledge Assessment of internal consistency of CGI with

  • ntologies describing

the CGI Assessment of CGI to external data and

  • ntologies known to

influence the CGI Assessments of the Information Assessments of the Informations Source

slide-13
SLIDE 13

Trust Model

Python PostgreSQL/ PostGIS OWL

Feature type rules queried from OWL Features queried From PostgreSQL Trust rating written Back to database Comparisons between Features and ontology in python

slide-14
SLIDE 14

Feature Trust Rating

fruit_tree_species Lemon fruit_tree_height 2m fruit_tree_crown_diameter 1m fruit_tree_dbh 0.12m fruiting_observation Fruiting fruit_tree_trust_rating_overall 100 fruit_tree_trust_rating_metrics 100 fruit_tree_trust_rating_fruiting 100 fruit_tree_trust_rating_location 100

slide-15
SLIDE 15

Feature Trust Rating

fruit_tree_species Coconut fruit_tree_height 5m fruit_tree_crown_diameter 2m fruit_tree_dbh 0.3m fruiting_observation Fruiting fruit_tree_trust_rating_overall 66.67 fruit_tree_trust_rating_metrics 100 fruit_tree_trust_rating_fruiting 100 fruit_tree_trust_rating_location

slide-16
SLIDE 16

ONTOLOGY

slide-17
SLIDE 17

Ontologies

  • Ontologies in crowdsourcing?

– accessibility – adjustability – versatility

  • Implementation

– Protégé – OWL/RDFS/XML

slide-18
SLIDE 18

Ontology

slide-19
SLIDE 19

Ontology

hasMaxHeight

slide-20
SLIDE 20

Ontology

hasMaxHeight

10 metres

slide-21
SLIDE 21

Protégé

slide-22
SLIDE 22

Protégé

slide-23
SLIDE 23

Protégé

slide-24
SLIDE 24

Protégé

slide-25
SLIDE 25

LINKED DATA

SPARQL RDFLib

slide-26
SLIDE 26

SPARQL Query in RDFLib

  • Return reference attributes (via URIs)

SELECT ?O WHERE { <http://somethingGoesHere.org/foss4tree#appleTree> foss4tree:hasMaxHeight ?O }

slide-27
SLIDE 27

SPARQL Query in RDFLib

  • Return reference attributes (via URIs)

SELECT ?O WHERE { <http://somethingGoesHere.org/foss4tree#appleTree> foss4tree:hasMaxHeight ?O }

slide-28
SLIDE 28

SPARQL Query in RDFLib

  • Return reference attributes (via URIs)

SELECT ?O WHERE { <http://somethingGoesHere.org/foss4tree#appleTree> foss4tree:hasMaxHeight ?O }

slide-29
SLIDE 29

SPARQL Query in RDFLib

  • Return reference attributes (via URIs)

SELECT ?O WHERE { <http://somethingGoesHere.org/foss4tree#appleTree> foss4tree:hasMaxHeight ?O }

slide-30
SLIDE 30

SPARQL Query in RDFLib

  • Return reference attributes (via URIs)

SELECT ?O WHERE { <http://somethingGoesHere.org/foss4tree#appleTree> foss4tree:hasMaxHeight ?O }

slide-31
SLIDE 31

SPARQL Query in RDFLib

  • Return reference attributes (via URIs)

SELECT ?O WHERE { <http://somethingGoesHere.org/foss4tree#appleTree> foss4tree:hasMaxHeight ?O }

TO THE TRUST MODEL

slide-32
SLIDE 32

Linked Data

  • Structure of RDF

– Triples (Subject, Predicate, Object)

<http://somethingGoesHere.org/foss4tree#t44> <foss4tree:hasHeight> <2.5>

– Familiar (URIs), accessible, mashups

slide-33
SLIDE 33

OUTPUT

RDFLib Folium

slide-34
SLIDE 34

LINKED DATA PYTHON MODEL WUNDERGROUND FOLIUM OUTPUT

slide-35
SLIDE 35

LINKED DATA PYTHON MODEL WUNDERGROUND FOLIUM OUTPUT PYTHON MODEL TRUST RATING > 70 WINDSPEED MAP THIS

slide-36
SLIDE 36

LINKED DATA PYTHON MODEL WUNDERGROUND FOLIUM OUTPUT LINKED DATA

?id <http://somethingGoesHere.org/foss4tree#hasTR> ?tr . FILTER (?tr > 70) ?id <http://somethingGoesHere.org/foss4tree#hasSpecies> ?species . ?id <http://somethingGoesHere.org/foss4tree#hasFruiting> ?fruiting . ?id <http://somethingGoesHere.org/foss4tree#hasLat> ?lat . ?id <http://somethingGoesHere.org/foss4tree#hasLong> ?long . ?id <http://somethingGoesHere.org/foss4tree#hasHeight> ?height

slide-37
SLIDE 37

LINKED DATA PYTHON MODEL WUNDERGROUND FOLIUM OUTPUT PYTHON MODEL TRUST RATING > 70 ID i LAT i LONG i … ID ii LAT ii LONG ii … ID iii LAT iii LONG iii … ID iv LAT iv LONG iv …

slide-38
SLIDE 38

LINKED DATA PYTHON MODEL WUNDERGROUND FOLIUM OUTPUT WEATHER UNDERGROUND

http://api.wunderground.com/api/##/geolookup/q/%f,%f.json http://api.wunderground.com/api/##/conditions/q/pws:%s.json www.wunderground.com

slide-39
SLIDE 39

LINKED DATA PYTHON MODEL WUNDERGROUND FOLIUM OUTPUT PYTHON MODEL TRUST RATING > 70 WINDSPEED

slide-40
SLIDE 40

LINKED DATA PYTHON MODEL WUNDERGROUND FOLIUM OUTPUT FOLIUM

map1 = folium.Map(location = [Lat,Long], zoom_start=16) . . . For tree in trees: map1.simple_marker(treeLat, treeLong, popup = '''... https://github.com/python-visualization/folium

slide-41
SLIDE 41

LINKED DATA PYTHON MODEL WUNDERGROUND FOLIUM OUTPUT OUTPUT html ...

slide-42
SLIDE 42
slide-43
SLIDE 43
slide-44
SLIDE 44

Where to from here…

slide-45
SLIDE 45

Where to from here…

WHY? Improved credibility of crowdsourced data

slide-46
SLIDE 46

Where to from here…

WHY? Improved credibility of crowdsourced data HOW? Trust models and implementation

slide-47
SLIDE 47

Where to from here…

WHY? Improved credibility of crowdsourced data HOW? Trust models and implementation THE HERE AND NOW

slide-48
SLIDE 48

Traditional Spatial Datasets

  • Credibility from legacy
  • Provenance for tracing errors
  • Dataset-level consideration
slide-49
SLIDE 49

W3C PROV

DATASET COLLECTION wasGeneratedBy

slide-50
SLIDE 50

W3C PROV

DATASET COLLECTION wasGeneratedBy

… back to triples!

slide-51
SLIDE 51

W3C PROV

DATASET COLLECTION AGENCY wasGeneratedBy wasAttributedTo

slide-52
SLIDE 52

W3C PROV

DATASET COLLECTION AGENCY wasGeneratedBy wasAssociatedWith wasAttributedTo

slide-53
SLIDE 53

Authoritative Data

  • Dataset-level reactive provenance
slide-54
SLIDE 54

Authoritative Data

  • Dataset-level reactive provenance
slide-55
SLIDE 55

Authoritative Data

  • Dataset-level reactive provenance
slide-56
SLIDE 56

Crowdsourced Data

  • Feature level
slide-57
SLIDE 57

Crowdsourced Data

  • Feature level
slide-58
SLIDE 58

Crowdsourced Data

  • Feature level
slide-59
SLIDE 59

Crowdsourced Data

  • Feature level
slide-60
SLIDE 60

Crowdsourced Data

  • Feature level
slide-61
SLIDE 61

Trust Ratings

  • Simple indication of credibility of

Datasets Features Attributes

  • Provides proactive provenance
  • Increases usability of crowdsourced data