START Conference Manager    

Artificial Intelligence and INSPIRE: Leveraging the INSPIRE Code Lists

Clinton Smyth and Jake McGregor

(Submission #95)


Abstract

Artificial Intelligence (AI) applications strive to make computer systems ever more useful to human beings by enabling them to carry out ever more complex tasks.

Human language is critical to efficient communication between people and is assuming an increasingly important role in communication between humans and computers.

The code lists embedded in the INSPIRE standards are encodings in human language (words) of information into computers, sometimes with an inherently complex structure. These words play a critical role in enabling INSPIRE-compliant maps to become the foundation upon which to build very powerful AI applications.

The INSPIRE code lists for lithology (Theme: Geology), HILUCS (Theme: Land Use) and land cover (Theme: Land Cover) are examples of complex hierarchical taxonomies reflecting the broad range (of sub-classes) of mapped features they are used to label. Querying maps attributed from these code lists can be very difficult using standard SQL. However, mapping applications which include reference to the taxonomies during query resolution can produce valuable results which would be impossible if the maps had not been attributed with the complex, reality-reflecting taxonomies in the first place.

These taxonomy-aware applications can be extended with technologies such as knowledge engineering, reasoning with uncertainty and logic programming to yield powerful artificial intelligence applications which produce compelling maps for professional users and laymen alike.

This presentation will demonstrate an artificial intelligence application developed for the management of environmental pollution from abandoned mines in the European Union, whose efficacy is entirely dependent on the accuracy and expressiveness of the three INSPIRE code lists mentioned above.

As more AI-based applications are built on the informational richness of the INSPIRE code lists, criticisms of their complexity are likely to diminish, the satisfaction of producers of new INSPIRE maps will increase, motivation for the conversion of legacy maps to INSPIRE-compliance will strengthen, and committees responsible for the curation of INSPIRE code lists are likely to be invigorated.

The presentation will close with comments on certain shortcomings of two of the INSPIRE code lists, and suggestions as to how these may be corrected.

The intended audience of this talk is anyone with a stake in, or an opinion on, the INSPIRE standards, be they data providers or users, from public or private organizations.

Categories

Topic Area:  [3.2] Tools and technologies
Abstract Type:  Oral Presentation

Additional Fields

 
Academic:   No
 
Data Provider:   Yes
 
Data User:   Yes
 
INSPIRE Implementer (IT):   Yes
 
INSPIRE newbies:   No
 
Policy Officers:   No
 
Public Administration (MS/Regional/Local):   No
 
Thematic specialists:   No
 
Comments:   This presentation would comfortably fall under a number of other topic areas, such as 1.2 Environmental monitoring and reporting; 1.6 Environmental, economic, social impact assessments; 1.12 Mitigation and management of natural and man-made hazards and disasters; 3.1 Making INSPIRE work in 2025; 3.4 Fitness for purpose


START Conference Manager (V2.61.0 - Rev. 5269)