Skip to content

Latest commit

 

History

History
71 lines (50 loc) · 3.52 KB

README.md

File metadata and controls

71 lines (50 loc) · 3.52 KB
Raw

climatemind-ontology-visualization

The ontology is visualized using cytoscape.js and Dash.

image

Setup

The easiest way to run this visualization is via docker with the command

docker run -p 8050:8050 henry2833/climatemind-ontology-visualization

It contains the ontology downloaded and processed at Mar 23. Then, go to http://127.0.0.1:8050/

Setup without Docker

  1. Follow the instructions at ClimateMind/climatemind-ontology-processing.git and download their requirements.txt and install it as a pip module.

The requirements.txt file for this project doesn't include climatemind-ontology-processing. To run this project without Docker, you need to have installed the ontology-processing pip package on your machine.

  1. Clone the repository
git clone https://github.com/ClimateMind/climatemind-ontology-visualization.git
  1. Install the requirements
pip install -r requirements.txt
  1. Run the script
python visualize.py
  1. Visit http://127.0.0.1:8050/

Usage Notes

  • Select the specific subgraph (a view) of the ontology that you want from the top-left dropdown.
    • downstream nodes are caused by increase in greenhouse effect
    • upstream nodes cause increase in greenhouse effects
    • mitigation solutions affect upstream nodes
    • adaptation solutions affect downstream nodes
    • specific views of only downstream nodes causing specific personal values are also availabl
  • Search for nodes in the currently selected graph using the "search for nodes" dropdown top right.
  • Select filters (right now only three) with the bottom right dropdown ("Select filter")
    • "personal value nodes/solutions without long description": highlights the personal (PV) associated nodes or solution nodes without a schema_longDescription property - Only 1 solution node doesn't have long descriptions. View it in upstream_mitigations graph.
    • "personal value nodes/edges without sources": highlights the PV nodes without any sources. highlights the edges (in red) without any sources. Edges connected to solutions are ignored in this search.
    • "all personal values": highlights all personal value nodes
    • Nodes are only highlighted when they are visible in the graph. For example, applying the "all personal values" filter when viewing the "upstream" graph won't do anything as there are no PV nodes upstream.
    • For more filters to be added in, please send a message in Slack :) !

Hovering over any downstream node causes the edge path that leads to increase in greenhouse effect highlighted.

Clicking on a node causes its JSON properties to be shown in the right grey rectangle.

Double clicking on that right grey rectangle leads you to WebProtege.

So to go to WebProtege you have to click once on the node, then double click on the grey rectangle

Even though nodes might show up as having sources on WebProtege, these sources may not show up in the visualization (and will be flagged as missing source). This is because a source about an edge is added to both the start and end node of that edge. Therefore, just because a node has a source on WP doesn't mean the source is for that node (it may be for an edge linked to that node).

Notes

ontology.owl file is included in this repository. However, if you want to see an updated view of the repository, then download the OWL from WebProtge and replace the ontology.owl file.

If you get an error when installing pygraphviz, then try sudo apt-get install graphviz libgraphviz-dev pkg-config.