Refers to the techniques used for the quantitative examination of textual and linguistic styles, which is often used as a method for authorship attribution studies. It can also be applied to music and fine art.

Characteristics examined include word, grammar and punctuation choice, as well as average syllables per word, words per sentence and words per paragraph.

One method involves analysing a text to find the 50 most common words, dividing the text into equal parts and comparing the frequency of each of these 50 words in each part. An alternative is to construct a set of rules, which are tested against known texts and then used to determine the author for texts that are uncertain. Collocation methods can also be used.

The use of computers for statistical analysis in stylometrics enables larger quantities of data to be analysed, using algorithms, which can produce a more accurate result. It also allows comparisons using relatively common parts of speech, whereas earlier methods concentrated on the rarest elements of a text. Several pieces of specialist software are available that facilitate the process.

Related methods include Collating, Content analysis, Statistical analysis, Stemmatics and Text mining.

Syndicate content