Sole Exception Report - {{ title }}
{% if corpora %}
List of corpora
{% endif %}
{% if works %}
List of works
{% endif %}
Key to Columns
- corpus
- The name of the benchmark corpus to which the work is compared.
- corpus length
- The number of tokens in corpus, being the sum of the average number of tokens in each work (across all of its witnesses) in the corpus, rounded up to the nearest integer.
- work
- The name of the unclassified work under examination.
-
- work length
- The average number of tokens in work, across all of its witnesses, rounded up to the nearest integer.
- number of markers
- The average number of markers (n-grams exclusive to the corpus) found across all of work's witnesses, rounded to the nearest integer.
- markers:length
- Ratio of number of markers to work length. Calculated by dividing the unrounded values of the two columns.
- score
- Ratio of number of markers to work length, adjusted to normalise the length of the corpus that is the source of the markers and scaled to a value between 0 and 1. Calculated by dividing the unrounded values and multiplying by a normalisation factor of the longest corpus length divided by the corpus length. Each value is then divided by the largest value to give a final value between 0 and 1.
{% if has_extra_cols %}
Other columns, between work
and work length, are those from
the CSV file the user (who ran the script to generate this report)
supplied. These columns contain additional work-specific data
outside the scope of the raw data used to generate markers and
scores.
{% endif %}
Click on a column heading to sort by that column
or change the order of its sorting. Shift-click on a column
heading to add that column to the existing sort.
Filter rows to matching column values in the
header boxes. For numeric fields > and < may be used, optionally
joined with "or" or "&&". Eg, "< 20 or > 40" or ">20 && <40".
{{ table|safe }}
{% if corpora %}
Corpora
{% endif %}
{% if works %}
Works
{% endif %}