Automatic content categorisation of text data files using subdivision to reduce false classification

摘要

Text data files are associated with categories by processing the text files with outline files. Large files 221 are subdivided 222 into file sections 223 of comparable length consistent with a preferred size. Each file section is categorised 224 and the original undivided file is assigned 226 to a category only if a sufficient proportion of its constituent sections have been so classified. This technique reduces false classifications.