25-01-2018, 10:32 PM
"A text known to be meaningless" is not something well defined in respect of the method used by the authors. E.g. a text with randomized letters would produce results quite different from a text with randomized words.
Apart from that, they start from the assumption that the text is meaningful (relying upon Landini and Montemurro & Zanette). I wonder, however, why computational linguistics professionals would still, over and over, argue that Zipf's law is an indicator of a meaningful text.
Apart from that, they start from the assumption that the text is meaningful (relying upon Landini and Montemurro & Zanette). I wonder, however, why computational linguistics professionals would still, over and over, argue that Zipf's law is an indicator of a meaningful text.
