LifeWatch ERIC Metadata Catalogue
Maxent Service
This service aims at running a species distribution modelling (sdm) service based on Maximum Entropy (maxent) algorithm. It is intended to predict the potential geographic distribution of a species based on environmental variables.
Maxent algorithm is a probabilistic modelling technique used for making predictions based on incomplete information or limited data. It is based on the principle of maximum entropy, which states that when making predictions, one should choose the probability distribution that is the least informative, or maximally uncertain, while still satisfying a set of known constraints.
The application calculates the distribution model of a species by taking as input files a vector of presence and a group of raster files that represent the background or environmental layers and providing five output files: elapid_object.ela, maxent_prediction_model.tif, maxent_prediction_plot.png, dependency_plots.png, and auc_score.txt.
The application can take as input also a set of parameters (optional), which have default values in case the user does not want to configure them.
Default
Identification
- Date ( Creation )
- 2023-06-01
- Date ( Publication )
- 2023-06-28
- Status
- On going / operational
- Version
- 1.0.0
- Keywords
- maxent
- Keywords
- maximum entropy
- Keywords
- sdm
- Keywords
- species distribution modelling
- Keywords
- artificial intelligence
- Access constraints
- License
- Use limitation
- Component has MIT License and it's based on a external library that needs to be referenced: Elapid, that has MIT License.
- Protocol
- DOI
- Operation name
- Maxent vector of presence
- Description
- This is a single file that represents the presence points of the species. The formats allowed are the following: bna, dxf, csv, shp, dbf, json, geojson, geojsonl, geojsons, gpkg, gml, xml, gpx, gtm, gtz, tab, mif, mid, dgn.
- Function
- Input file
- Operation name
- Maxent raster files
- Description
- This is a group of files that represent the background or environmental layers. These files should be in the following formats: tif, tiff, asc, hdr, h5. And they must be compressed in a zip file.
- Function
- Input file
- Operation name
- Maxent parameters
- Description
- The application can take as input also a set of parameters (optional), which have default values in case the user does not want to configure them.
- Function
- Optional input parameters
- Operation name
- elapid_object.ela
- Description
- ELAPID object containing the results of the Maxent model. Final filename will be determined by the name of the species.
- Function
- Output file
- Operation name
- maxent_prediction_model.tif
- Description
- Raster layer with the probability of presence of the species. Final filename will be determined by the name of the species.
- Function
- Output file
- Operation name
- maxent_prediction_plot.png
- Description
- Plot with the probability of presence of the species. Final filename will be determined by the name of the species.
- Function
- Output file
- Operation name
- dependency_plots.png
- Description
- Plots with the percentage of participation of each variable in the final model. Final filename will be determined by the name of the species.
- Function
- Output file
- Operation name
- auc_score.txt
- Description
- Area under the curve score of the model. Final filename will be determined by the name of the species.
- Function
- Output file
- Topic Category
- Species distribution
- Service Category
- data analysis
- Service Language
- eng
- Service TRL
- TRL 6 – Technology demonstrated in relevant environment
- Service Helpdesk
- https://helpdesk.lifewatch.eu
