To include hourly, minute and quaterly.
Ideally, allow numeric columns to be selected instead of datetime as a special mode, with the option to turn the group by time options to customise bin size, or value interval.
There will be many different scenarios with the numeric plot and options, so we can start with one usecase where the data is recorded with timestamp in seconds. And the purpose is to check certain keyword/term usage over time. The workflow is:
- Extract the original script with concordancer on each term, detach them.
- Concat the detached concodancers.
- Use the timeline tool to load the new corpus, select timestamp as X-axis, set interval as 10 seconds.
- Use the "match" column as additional group.
This should display the usage/count of selected terms over time.
This brings another possible option for concordancer, shall we allow multiple matches to be done in one go besides RegEx?
To include hourly, minute and quaterly.
Ideally, allow numeric columns to be selected instead of datetime as a special mode, with the option to turn the group by time options to customise bin size, or value interval.
There will be many different scenarios with the numeric plot and options, so we can start with one usecase where the data is recorded with timestamp in seconds. And the purpose is to check certain keyword/term usage over time. The workflow is:
This should display the usage/count of selected terms over time.
This brings another possible option for concordancer, shall we allow multiple matches to be done in one go besides RegEx?