In 2016 , neuroscientistAdam J Calhounwondered what his pet books would look like if he absent the words and leave nothing but the punctuation mark . Theresultwas a sensational — andsurprisingly beautiful — optic current of commas , question mark , semicolon , em quad - style , and periods .
Recently , Calhoun ’s inquiry piqued the involvement of researchers in the United Kingdom , who wondered if it was possible to identify an writer from his or her punctuation mark alone .
For decades , linguists have been able-bodied to employ the quirks of written text to nail the author . The process , calledstylometric analysisorstylometry , has dozens of legal and academic applications , facilitate researchers authenticate anonymous works of lit and even nab criminals like theUnabomber . But it commonly focus on an author ’s Good Book option and grammar or the distance of his or her sentences . Until now , punctuation has been largely ignore .
But consort to arecent paperled by Alexandra N. M. Darmon of the Oxford Centre for Industrial and Applied Mathematics , an author ’s use of punctuation can be extremely revealing . Darmon ’s team assembled nearly 15,000 documents from 651 different generator and “ de - worded ” each text . “ Is it possible to distinguish literary genre based on their punctuation sequences ? ” the researchers asked . “ Do the punctuation mark styles of generator acquire over time ? ”
Apparently , yes . The researchers craft mathematical rule that could identify item-by-item writer with 72 percent accuracy . Their power to detect a specific musical genre — from repulsion to philosophy to detective fiction — was accurate more than half the time , clocking in at a 65 per centum success rate .
The results , publish on the preprint serverSocArXiv , also revealed how punctuation style has evolved . The investigator find that “ the utilization of credit marks and periods has increase over time ( at least in our [ sampling ] ) but that the use of comma has decreased over time . Less noticeably , the use of semicolon has also decrease over sentence . ”
You probably do n’t require to develop a sinewy algorithm to figure that last bit out — you just have to break through open something byDickens .
