Welcome to the AI WitchLab
... where cutting-edge AI developments and real-world research and practice meet the playful spirits of cyber mischief. Enter a world where the seriousness of progress coexists under one roof with the snark digital devilry, keeping things irreverently real.
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​​
No Witchwork.
Millions of undetermined specimen vouchers need identification.
Millions of undetermined specimen vouchers in the global herbaria meet the lack of skilled scientific personnel.
However, this is a complex problem, yet no comfortable solution to automatically identify herbarium specimens due to:
- Lack of skilled scientific personnel to resolve the problem
- Possible species gaps due to yet unknown specimens (see: Hortal et al., 2015)
- Plant Functional Traits concept may support problem-solving, but it has not yet been explored regarding environmental plant traits
- Use of AI / deep learning / CNNs has not yet been developed to resolve the problem
​
Topography as a set of predictors in SDM.
This study considers topography as an overarching concept for different aspects that can serve as predictors in species distribution models.
- How can a particular key aspect pattern influence species distribution?
- How can a particular key aspect pattern in topography be measured in a GLMM?
​
