#MammalWeb

This project is about adding functionality to the system to ensure reliability of user classifications and provide detailed analysis and reports on this data. On user end: increase user engagement, which will result in more classifications and therefore better data to work with.

#Features

Swanson Algorithm

Each image was circulated to multiple users and retired after meeting the following criteria:

• the first five classifications were “nothing here” (blank)
• ten non-consecutive “nothing here” classifications (blank_consensus)
• ten matching classifications of species or species - combination, not necessarily consecutive (consensus).

If none of these criteria were met, the image was circulated until accumulating 25 species classifications (complete).

Administrator Dashboard

• Filters - Narrow down searches with filters. Includes support for searching by location.
• Photo/Sequence selection - Show only sequences to remove repeated data.
• User Management- Scrollable list of user ratings in a range of areas. Quickly find users who are causing problems.
• Graphs - Graph your current search using a range of variables. Uses GoogleCharts
• CSV - Download any search in CSV format compatible with R or Python based analyses software

User-facing Dashboard

• Map - Displays the locations of animals on a map using icons. Uses Leaflet
• Timeline - Timeline of classifications, linked with map.
• Slideshow - Slideshow of images, includes basic information.

Usage

Run the algorithm from root using:

php swansonAlgorithm/classify.php

Unit tests written using PHPUnit and are located in swansonAlgorithm/tests/algorithmTest.php. Run them from root using:

phpunit swansonAlgorithm/tests --bootstrap vendor/autoload.php

The node backend is available at mammalWebNode

git clone https://github.com/durhamteam7/mammalWebNode.git
cd mammalWebNode
npm install
npm start

Documentation

The project is documented using Angular-JSDoc and phpDocumentor

A browsable HTML version of this documentation for all parts of the project can be created by running the ./createDocs.sh bash file. (Note: requires npm to get dependancies). Docs will be found in the /docs folder.

License

Credit to team7 (Thomas Hudson, Tom Robson, Linus Ericsson, Mat Pardoe, Ram Gupta, Caitlin McArdle)

This project is licensed under the terms of the GNU AGPLv3 license. You can check out the full license here