"Linguistic Fingerprints" are a concept arising from the field of linguistic forensics. This discipline is divided into the examination of the written word and the spoken word. For the written word, we are concerned with forensic stylistics or stylometry, a sub-discipline within forensic linguistics.
While this discipline can have accuracy rates as high as 70-95% in some cases, it is doubtful that scriptural analysis would yield results this accurate due to a number of limitations.
Firstly, the National Clearinghouse for Science Technology & the Law1 notes that in courtrooms, forensic linguists typically would not conclude that someone had authored a particular work, but instead that an author had not authored a particular work:
In most cases, the expert can say with confidence that ... it is not likely that a certain person was the author of a particular written piece. ...[however,] it is easier to eliminate a person than to prove beyond a reasonable doubt that the person is the speaker or author.
In his masters thesis, Colin Mitchel of the University of South Africa2 notes the difficulty of establishing a means of determination:
Grant (2010) adduces that even if we accept the notion that “every native speaker has their own idiolect” (Coulthard 2004, 432) it is not necessarily true that an individual’s idiolect will be measurable in every text written by that individual, irrespective of the length of the text. Moreover, it would require a fairly substantial length of text before any measurable idiolect could be discerned, and in order to be useful in a forensic analysis, the idiolectal features would have to be repeated either in one text, or in a range of texts by the same author. Grant (2010) talks of the need to make a distinction between observation and theory when discussing idiolect. Even though the theory of an idiolectas distinctive variety of language may be necessary for authorship identification, the practical applications require the ability to detect consistent patterns of usage.
The typical way of determining the authorship of a piece is by comparing a work to another known work of the author using writer invarients. These are characteristics or distinctive which are consistently found in their writing. In English, many of these distinctive are centered around punctuation, though others can center on word choice and sentence structure. Because Biblical Hebrew has traditionally lacked many punctuation features of English and other modern languages, this would make comparisons comparatively more difficult.
For a very short work like Jude, there simply isn't a great body of writing to base this comparison on while something like Isaiah has a significantly larger body of writing to work with. Furthermore, there are often few other writings for comparison. Due to the nature of this discipline, a reference sample is generally required for comparison, otherwise opinions are based merely on an expert's intuition and "feel". But as Koppel and Schler3 note,
the examiner’s intuition about the significance of a
stylistic feature can lead to methodological subjectivity and bias
They also go on to point out that for edited works, many distinctive features used in the identification process can be normalized. This means that scribal editing done to Biblical manuscripts removes idiosyncrasies needed for "fingerprinting"
Another concern is that in Forensic Stylistics, the assumption is typically that works can be clearly distinguished as to their beginning and end and that works consist of a sole author. Documentary Hypotesis dictates however that many sections of scripture have 2 or more authors and finding the boundaries between works is nearly impossible. For example, perhaps the reason Computer software referenced in the OP thought the authorship change of Isaiah started 6 chapters earlier could be due to a splicing in which works of both authors were redacted together and fingerprints of both authors are seen in those 6 chapters as a result.
While we might be able to determine if Paul did not author a purported Pauline epistle using these techniques with relative certainty, it would be nearly impossible to determine an exact number of authors in total. What works in one case for one book or passage of the Bible will not always be applicable in another. This doesn't mean that we cannot make educated guesses about texts and that this analysis would not be insightful, but to boil this down to an exact number of authors or to correctly divide texts into an exact number of John Doe authors would firstly be unverifiable and secondly be difficult to do with a high degree of confidence.
1 Angela Lack, July 2008, "It's Evident" newsletter, National Clearinghouse for Science Technology & the Law
3 Colin Mitchell, University of South Africa "Investigating The Use Of Forensic Stylistic And Stylometric Techniques Inthe Analyses Of Authorship On A Publicly Accessible Social Networkingsite (Facebook)"
3 Moshe Koppel and Jonathan Schler; Dept. of Computer Science Bar-Ilan University Exploiting Stylistic Idiosyncrasies for Authorship Attribution