how to find most common words in text by matlab

how to tag POS on nouns and verbs in MATLAB, Is it related to regular expressions? I know that regular expressions find a pattern in a text, but I want to find the most common words in texts and tag POS on them( I mean the words are nouns or verbs) and then exchange that POS and make an unfamiliar pair of words. how can I find the most common words in texts by MATLAB?is there any solution for that or I should use another software?

Answer 1

[Christopher Creutzig]

Finding the most common words is easy with Text Analytics Toolbox:

>> sonnets = extractFileText("sonnets.txt");
>> sonnets = erasePunctuation(sonnets);
>> tokenizedSonnets = tokenizedDocument(lower(sonnets));
>> bag = bagOfWords(tokenizedSonnets);
>> topkwords(bag, 10)

ans =

    10×2 table

       Word     Count
      ______    _____

      "and"      490 
      "the"      436 
      "to"       409 
      "my"       371 
      "of"       370 
      "i"        344 
      "in"       321 
      "that"     320 
      "thy"      281 
      "thou"     234

You probably want to remove some words (check out removeWords and stopWords). POS tagging is supported in release R2018b and later, see addPartOfSpeechDetails.

[bita hallajian]

With great thanks for your help I'll try the helpful points you directed me I created a text document with the name "sonnets.txt" and loaded it in command window but there is an error : "Number of columns on line 2 of ASCII file sonnets.txt must be the same as previous lines." Can you give me useful advice to eliminate this error?

[Christopher Creutzig]

What command(s) did you try to read that file? The error message looks like you tried to read it as a table; try using the commands listed above instead.

Answer 2

[Sarah Palfreyman]

See Text Analytics Toolbox

[IORUNDU GABRIEL]

Which version of matlab is the least that supports the Text analytic toolbox?

[Rik]

R2017b

Answer 3

[Charmaine Tan]

Hi, after finding my topkwords (most frequent words), how can I plot a histogram of these?

[Christopher Creutzig]

txt = extractFileText('sonnets.txt');
td = tokenizedDocument(lower(txt));
td = erasePunctuation(td);
bow = bagOfWords(td);
top = topkwords(bow,20);

bar(top.Count)
set(gca,'XTick',1:size(top,1),'XTickLabel',top.Word,'XTickLabelRotation',45)

(In general, it's a good idea not to ask a new question as an “answer,” but to open a new question instead. It helps other people searching MATLAB Answers in the future.)

[Charmaine Tan]

Noted, I'll do that. Thanks a lot!