|
|
|
You invited to explore the two lives demos of the MemoRekall-IIIF environment ([COESO semantic analysis](https://coeso.tetras-libre.fr/) and [sandbox](https://iiif.tetras-libre.fr/)) as you please, however, we can guide your exploration by pointing you towards some specific features. We shall divide this guide into each of the live demos.
|
|
|
|
|
|
|
|
1. [COESO Semantic Analysis](#coeso-semantic-analysis)
|
|
|
|
- [Network Visualizations](#network-visualizations)
|
|
|
|
- [Bespoke Data-Driven Interface](#bespoke-data-driven-interface)
|
|
|
|
|
|
|
|
# COESO Semantic Analysis
|
|
|
|
|
|
|
|
This is a demo made specifically for browsing and viewing the results of an analysis which can be performed in the MemoRekall-IIIF environment which means that you cannot modify the content or add new annotations. The examples are based on a full analysis which is explained in detail in this article. Here we shall go over some of the features that are notably specific to features we have added to Mirador and that pertain to our concept of an IIIF Manifest Network.
|
|
|
|
|
|
|
|
## Network Visualizations
|
|
|
|
|
|
|
|
When you open the live demo, a first manifest is opened in a Mirador window. This is a network visualization of the entire collection of documents. It is a high-resolution image (which can cause the demo to run a bit slower on some machines) that was created programmatically from manually input network data. Note that the size of each node corresponds to the number of neighbors it has.
|
|
|
|
|
|
|
|
Note that each node in the network is linked to an annotation. You can run your mouse over the network image and the corresponding annotation will be selected in the list of annotations. Similarly, you can run your mouse over the list of annotations and the corresponding node will be selected on the image. Notice also that each annotation allows you to open the node’s corresponding manifest straight into another Mirador window.
|
|
|
|
|
|
|
|
When you open one of these manifests, you will find a manifest based on the corresponding document. If you consult the annotation list, you will find that each annotation corresponds to each of the network links. This allows you to navigate the network of documents either from the overview level of the network visualization, or at a more local level form each document.
|
|
|
|
|
|
|
|
If you go into the Add Resource screen, you will find another manifest called 02 Automatic Network Configuration. This is another network visualization of the same collection of documents. This time however, the links between nodes were created programmatically using a simple search term matching algorithm. This is to demonstrate how a visualization of this kind can be made in an unsupervised, automatic manner.
|
|
|
|
|
|
|
|
## Bespoke Data-Driven Interface
|
|
|
|
|
|
|
|
These kinds of network visualizations can be useful to get to know a collection of documents and examine the contextual links between them. However, for some analyses it can be useful to design more bespoke interfaces for navigating a collection of data. A goal on the MemoRekall-IIIF roadmap is to design tools that will allow the user to customize the creation of this kind of interface. Here, we can see an example of how this kind of interface might support a semantic analysis.
|
|
|
|
|
|
|
|
If you navigate to the Add Resource window, you can open the manifest called 07 Composite Collaboration Analytics Compass. You will find another overview manifest that is similar to the network visualizations. Here however, the documents are placed on a Collaboration Typology Compass (based on Boullier and Pidoux, 2021). You have the same functionalities for exploring the collection of documents, but their placement in the data-driven interface has a different semiotic meaning.
|
|
|
|
|
|
|
|
If you open one of the documents, the linked manifest will display the associated document, but also an image annotation that display’s the document’s full polygon on the Collaboration Typology Compass. This allows you to navigate not only the collection of documents at different levels (the entire collection and local level documents), but also the multidimensional semantic knowledge that organizes them (composite view of the collection, local decomposition of each document on the compass). |