AI-based discovery of habitats from museum collections

Area of Science:

• Ecology
• Evolutionary Biology
• Museum Studies
• Artificial Intelligence

Background:

• Museum collection records provide valuable historical data on species occurrence.
• Associated textual descriptions of sample location habitats are often underutilized.
• Reconstructing past habitats is crucial for understanding ecological and evolutionary dynamics.

Purpose of the Study:

• To explore the potential of artificial intelligence (AI) methods for reconstructing past habitats.
• To leverage underutilized habitat descriptions from museum records.
• To address ecological and evolutionary questions using reconstructed historical habitat data.

Main Methods:

• Utilizing artificial intelligence (AI) techniques.
• Analyzing textual descriptions of habitat associated with museum specimen records.
• Developing methods for historical habitat reconstruction.

Main Results:

• Demonstrated the potential of AI to extract and interpret habitat information from historical museum records.
• Showcased the feasibility of reconstructing past habitats using these methods.
• Highlighted the value of previously overlooked textual data.

Conclusions:

• Artificial intelligence offers a novel approach to unlock historical ecological data within museum collections.
• Habitat reconstruction from museum records can provide new insights into ecological and evolutionary processes.
• This methodology enhances the utility of museum collections for contemporary scientific inquiry.