LogosLink User's Manual · LogosLink version 2.0.0

Context Coverage Analytics (Corpus)

Context Coverage analytics shows the degree of overlap between elements in the corpus plus dependent models, and elements in embedded contexts. This is done by aggregating results from individual ontology, argumentation model and agency model Context Coverage analytics together with that of the corpus itself.

This is useful to evaluate how much of the context is referenced by the corpus and its dependent models, and how much of the corpus plus its dependent models represents the context.

Parameters

  • β. This allows you to weight precision and recall differently. Typical values are between 0.5 and 2.0. See the Details section below for more information.

Results

Results are given separately for the corpus itself, each kind of dependent models (ontologies, argumentation models, and agency models), plus aggregated. For each of these, results are presented as some overall data plus a tree table.

Corpus
Overall data
  • Overall. This is the degree of coverage (overlap) between the corpus and its embedded context.
  • Overall weighted. This is the degree of coverage (overlap) between the corpus and its embedded context, but applying the β parameter to weight precision and recall differently. Please see the Details section below for more information.

An overall coverage of 0% indicates that the corpus contains no elements linked to the embedded context. An overall coverage of 100% indicates that the corpus and the embedded context fully overlap, that is, all the applicable elements in the corpus are linked to the context, and all elements in the context are linked to elements in the corpus.

Tree table

The tree table shows a hierarchy of corpus and context element kinds, with some details for each one.

Data in the tree table is aggregated hierarchically. For example, the data for "Topic" and "Author" are added up and shown under "Corpus Element".

For each element kind, the following data are shown:

  • Element kind. This indicates the kind of corpus or context element for which data is provided.
  • Cover count. This shows how many elements of the kind are linked to the other model. For corpus elements, this indicates how many of them are linked to context elements; for context elements, it indicates how many of them are linked to corpus elements.
  • Total count. This shows how many elements of the kind exist.
  • Coverage. This shows the percentage of the cover count in relation to the total count.

The Coverage value for the "Corpus Element" entry in the tree table indicates how much of the corpus references the embedded context. Similarly, the Coverage value for the "Context Element" entry indicate how much of the context is referenced by the corpus.

Ontologies

The total number of ontologies used by the analytics is shown at the top as "Count". Ontologies are gathered from the following sources:

  • Dependent ontologies of active documents.
  • Embedded ontologies in topics of active documents.
  • The ontology embedded in the corpus, if it exists.
Overall data
  • Overall. This is the degree of coverage (overlap) between the ontologies and the embedded contexts.
  • Overall weighted. This is the degree of coverage (overlap) between the ontologies and the embedded contexts, but applying the β parameter to weight precision and recall differently. Please see the Details section below for more information.

An overall coverage of 0% indicates that the ontologies contain no elements linked to the embedded contexts. An overall coverage of 100% indicates that the ontologies and the embedded contexts fully overlap, that is, all the applicable elements in the ontologies are linked to the contexts, and all elements in the contexts are linked to elements in the ontologies.

Tree table

The tree table shows a hierarchy of ontology and context element kinds, with some details for each one.

Data in the tree table is aggregated hierarchically. For example, the data for "Category" and "Atom" are added up and shown under "Entity".

For each element kind, the following data are shown:

  • Element kind. This indicates the kind of ontology or context element for which data is provided.
  • Cover count. This shows how many elements of the kind are linked to the other model. For ontology elements, this indicates how many of them are linked to context elements; for context elements, it indicates how many of them are linked to ontology elements.
  • Total count. This shows how many elements of the kind exist.
  • Coverage. This shows the percentage of the cover count in relation to the total count.

The Coverage value for the "Ontology Element" entry in the tree table indicates how much of the ontologies represents the embedded contexts. Similarly, the Coverage value for the "Context Element" entry indicate how much of the contexts is represented by the ontologies.

Argumentation models

The total number of argumentation models used by the analytics is shown at the top as "Count". Argumentation models are gathered from the following sources:

  • Dependent argumentation models of active documents.
Overall data
  • Overall. This is the degree of coverage (overlap) between the argumentation models and the embedded contexts.
  • Overall weighted. This is the degree of coverage (overlap) between the argumentation models and the embedded contexts, but applying the β parameter to weight precision and recall differently. Please see the Details section below for more information.

An overall coverage of 0% indicates that the argumentation models contain no elements linked to the embedded contexts. An overall coverage of 100% indicates that the argumentation models and the embedded contexts fully overlap, that is, all the applicable elements in the argumentation models are linked to the contexts, and all elements in the contexts are linked to elements in the argumentation models.

Tree table

The tree table shows a hierarchy of argumentation model and context element kinds, with some details for each one.

Data in the tree table is aggregated hierarchically. For example, the data for "Speaker" and "Proposition" are added up and shown under "Argumentation Model Element".

For each element kind, the following data are shown:

  • Element kind. This indicates the kind of argumentation model or context element for which data is provided.
  • Cover count. This shows how many elements of the kind are linked to the other model. For argumentation model elements, this indicates how many of them are linked to context elements; for context elements, it indicates how many of them are linked to argumentation model elements.
  • Total count. This shows how many elements of the kind exist.
  • Coverage. This shows the percentage of the cover count in relation to the total count.

The Coverage value for the "Argumentation Model Element" entry in the tree table indicates how much of the argumentation models references the embedded contexts. Similarly, the Coverage value for the "Context Element" entry indicate how much of the contexts is referenced by the argumentation models.

Agency models

The total number of agency models used by the analytics is shown at the top as "Count". Agency models are gathered from the following sources:

  • Dependent agency models of active documents.
Overall data
  • Overall. This is the degree of coverage (overlap) between the agency models and the embedded contexts.
  • Overall weighted. This is the degree of coverage (overlap) between the agency models and the embedded contexts, but applying the β parameter to weight precision and recall differently. Please see the Details section below for more information.

An overall coverage of 0% indicates that the agency models contain no elements linked to the embedded contexts. An overall coverage of 100% indicates that the agency models and the embedded contexts fully overlap, that is, all the applicable elements in the agency models are linked to the contexts, and all elements in the contexts are linked to elements in the agency models.

Tree table

The tree table shows a hierarchy of agency model and context element kinds, with some details for each one.

Data in the tree table is aggregated hierarchically. For example, the data for "Speaker" and "Response Part" are added up and shown under "Agency Model Element".

For each element kind, the following data are shown:

  • Element kind. This indicates the kind of agency model or context element for which data is provided.
  • Cover count. This shows how many elements of the kind are linked to the other model. For agency model elements, this indicates how many of them are linked to context elements; for context elements, it indicates how many of them are linked to agency model elements.
  • Total count. This shows how many elements of the kind exist.
  • Coverage. This shows the percentage of the cover count in relation to the total count.

The Coverage value for the "Agency Model Element" entry in the tree table indicates how much of the agency models references the embedded contexts. Similarly, the Coverage value for the "Context Element" entry indicate how much of the contexts is referenced by the agency models.

Aggregated
Overall data
  • Overall. This is the degree of coverage (overlap) between the corpus plus its dependent models and the embedded contexts.
  • Overall weighted. This is the degree of coverage (overlap) between the corpus plus its dependent models and the embedded contexts, but applying the β parameter to weight precision and recall differently. Please see the Details section below for more information.

An overall coverage of 0% indicates that the corpus plus dependent models contain no elements linked to the embedded contexts. An overall coverage of 100% indicates that the corpus plus dependent models and the embedded contexts fully overlap, that is, all the applicable elements in the corpus plus dependent models are linked to the contexts, and all elements in the contexts are linked to elements in the corpus plus dependent models.

Tree table

The tree table shows a hierarchy of abstract and context element kinds, with some details for each one.

Data in the tree table is aggregated hierarchically.

For each element kind, the following data are shown:

  • Element kind. This indicates the kind of element for which data is provided.
  • Cover count. This shows how many elements of the kind are linked to the other model. For corpus plus dependent model elements, this indicates how many of them are linked to context elements; for context elements, it indicates how many of them are linked to corpus plus dependent model elements.
  • Total count. This shows how many elements of the kind exist.
  • Coverage. This shows the percentage of the cover count in relation to the total count.

The Coverage value for the "Total Element" entry in the tree table indicates how much of the corpus plus dependent models references the embedded contexts. Similarly, the Coverage value for the "Context Element" entry indicate how much of the contexts is referenced by the corpus plus dependent models.

Details

This analytics employs a precision and recall approach. Precision is interpreted as how many corpus (or dependent model) elements are linked to context elements (i.e. how "precise" the corpus and dependent models is are in capturing the context), whereas recall is intepreted as how many context elements are linked to corpus or dependent model elements (i.e. how much context is "recalled" by the corpus and its dependent models).

Similarly, the overall and weighted overall coverage measures are calculated as F1 and Fβ scores , respectively. Values of β between 0 and 1 make precision more important than recall in calculating the weighted overall coverage. Contrarily, values of β above 1 make recall more important than precision.

See Also

Contents distributed under a Creative Commons Attribution 4.0 International License · About · Terms of Use · Contact Us · last updated on 02/01/2025 11:53