AI-driven landscape values mapping

Area of Science:

• Environmental Science
• Geospatial Intelligence
• Computational Social Science

Background:

• Traditional landscape perception methods lack scale and scope.
• Understanding human perception is crucial for conservation and planning.
• Social media data offers a vast, untapped resource for landscape studies.

Purpose of the Study:

• To introduce and validate the AI-Perceptual Landscape Mapping (AI-PLM) framework.
• To model collective human perception of landscapes using social media data.
• To assess cultural ecosystem services through AI-driven analysis.

Main Methods:

• Integrated AI-PLM framework combining geospatial intelligence, machine learning, and NLP.
• Analysis of geotagged Flickr photographs and user comments from Romania.
• AI-Spatial Cognition (Head/Tail Breaks, DBSCAN, viewshed analysis) and Affective-Semantic Intelligence (sentiment, topic modeling).

Main Results:

• Identified strong spatial hierarchies of landscape appreciation, with peaks in the Carpathians, Braşov, Bucharest, Maramureş, and the Black Sea coast.
• Sentiment analysis revealed predominantly positive emotions linked to nature-oriented regions.
• Topic modeling highlighted themes of photography, heritage, and recreation in user-generated content.

Conclusions:

• AI-PLM provides a scalable and transferable methodology for assessing cultural ecosystem services.
• The framework bridges physical geography and emotional expression for landscape analysis.
• AI-PLM offers practical tools for data-driven landscape management, conservation, and tourism planning.