LogosLink User's Manual · LogosLink version 2.0.1

Analysis Progress Analytics (Corpus)

Analysis Progress analytics show how much effort you have invested in analysing documents in a corpus, and how much effort is necessary to finish.

This is useful to keep track of your work, report your effort and performance, and estimate when you will be done.

Parameters

(none)

Results

Results are given separately for each kind of analysis (ontology, argumentation and agency). For each of these, results are presented in the same way, which involves two lists and a summary table and chart.

Pending

This list shows the documents for which analysis hasn't started yet. For each document, the following data are shown:

Key and Title. These are the basic details of the document.
Words. This is the number of words in the document.
Expected time. This is an estimation of the time that will be needed to complete analysis for the document. See the Details section below for additional information.

Done / Pending / Total

The table shows a summary of the anaysis for done, pending and total documents. See the Details section below for additional information on how done and pending documents are determined.

For each columns, the following data are shown:

Document count. This indicates how many documents there are for this category.
Word count. This indicates how many words there are in total for documents in this category.
Time. For the "Done" column, this is the actual time invested in analysing the documents. For the "Pending" column, this is an estimation of how much time it will take to analyse the pending documents. For the "Total" column, this is the sum of the previous too, which is an estimation of how much time it will have taken to complete analysis for all the documents once you are done.
Words/hour. This is the analysis performance obtained for the done documents. This indicates how fast analysis has been carried out.

In addition, a ring chart is shown, displaying the done and pending effort as number of documents (outer ring) and number of words (inner ring).

Done

This list shows the documents for which analysis is done. For each document, the following data are shown:

Key and Title. These are the basic details of the document.
Words. This is the number of words in the document.
Time. This is the time that was spent to analyse the document.
Expected time. This is an estimation of the time that should be necessary to analyse the document.
Deviation. This is the amount of time that anaysis actually needed over or under the estimated time. Positive times indicate that the document needed more time than expected, whereas negative times indicate that the document needed less time than expected.
Relative. This shows the time deviation relative to the expected time. Positive numbers indicate that the document needed more time than expected, whereas negative numbers indicate that the document needed less time than expected.
Words/hour. This is the number of words that were analysed per hour of work for the document.
Deviation. This is the amount of words per hour that was actually obtained for the document over or under the average words/hour for all the done documents. Positive numbers indicate that the analysis performed faster for the document as compared to the average, whereas negative numbers indicate that the analysis performed slower for the document as compared to the average.
Relative. This shows the words/hour deviation relative to the average. Positive numbers indicate that the analysis performed faster for the document as compared to the average, whereas negative numbers indicate that the analysis performed slower for the document as compared to the average.

Details

Since analysis performance is measured in words/hour, this analytics only considers active documents in the corpus that contain text in a known language. Active documents with no text, or with text in an unknown language, are not considered by the analytics.

The selected documents are classified into "done" and "pending" groups for each kind of analysis (ontology, argumentation and agency):

The "done" group includes those documents that have an associated file of the necessary kind. For example, if a document has an associated ontology file, then it is considered to be done as far as ontology analysis is concerned.
Similarly, the "pending" group includes those documents that do not have an associated file of the necessary kind. For example, if a document has no associated agency model file, then it is considered to be pending as far as agency analysis is concerned.

The "done" group is used to calculate the average words/hour performance. In turn, this average is the used to esimate how much time is expected to be used to analyse each "pending" document.