Automatic extractions of key topics from free text files. Each text file contains a court ruling.
We would like your help to process each document, one by one. Find (**stemmed**) terms that are common in the document being considered, compared with the average frequency for the same term across all the other documents. We are not interested in names of people, companies/organisations or places, but we can strip them out after your analysis.
Here is an example case:
[url removed, login to view]
From these we might get "tire" and "accident" and a host of others. We need frequency data for each, how many mentions in the master, how many mentions per 1000 words in master, the same data for the corpus of the other cases.