Part-of-speech tagging takes a text and marks grammatical information about all the words (and sometimes associated elements, like punctuation). This is a key step in enabling you to answer questions specific to language use in the text.
TINT (The Italian NLP Tool) is based on Stanford CoreNLP but includes training data for Italian.
The current stable build (v0.2) is no longer compatible with Stanford CoreNLP, and throws errors when you try to run it. Instead, go to the TINT downloads page and download tint-runner-1.0-SNAPSHOT-bin.tar.gz (latest nightly build, tar.gz). Save the file, and unzip it.
TINT is built using the ISTD (Italian Stanford Dependency Treebank), and uses the following tags.
| Value | Description | Examples | Contexts of use |
|---|---|---|---|
| A | adjective | bello, buono, pauroso, ottimo | una bella passeggiata\ un ottimo attaccante\ una persona paurosa |
| AP | possessive adjective | mio, tuo, nostro, loro | a mio parere\ il tuo libro |
| B | adverb | bene, fortemente, malissimo, domani | arrivo domani\ sto bene |
| BN | negative adverb | no, non, nemmeno | non arrivo nemmeno domani\ non sto bene |
| CC | coordinate conjunction | e, o, ma | i libri e i quaderni |
| CS | subordinate conjunction | mentre, quando | quando ho finito vengo |
| DD | demonstrative determiner | questo, codesto, quello | questo denaro\ quella famiglia |
| DE | exclamative determiner | che, quale, quanto | che disastro!\ quale catastrofe! |
| DI | indefinite determiner | alcuno, certo, tale, parecchio, qualsiasi | alcune telefonate\ parecchi giornali\ qualsiasi persona |
| DR | relative determiner | cui, quale | i cui libri |
| DQ | interrogative determiner | che, quale, quanto | che cosa\ quanta strada\ quale formazione |
| E | preposition | di, a, da, in, su, attraverso, verso, di | a casa del poeta\ prima di giorno\ verso sera |
| EA | articulated preposition | del, alla, dal | a casa del poeta\ all' alba |
| FB | balanced punctuation | ( ) - | frutta (pere e banane) |
| FC | clause boundary punctuation | , ; | mele, pere e banane**.** |
| FF | comma | , - | mele**,** pere e banane |
| FS | sentence boundary punctuation | ; . ? ! | cosa vuoi**?** |
| I | interjection | ahimè, beh, ecco, grazie | Beh, che vuoi? |
| N | cardinal number | uno, due, cento, mille, 28, 2000 | due partite\ 28 anni |
| NO | ordinal number | primo, secondo, centesimo | secondo posto |
| PC | clitic pronoun | ci, vi, ne | vi piace? |
| PD | demonstrative pronoun | questo, quello, costui | quello di Roma\ costui uccide |
| PE | personal pronoun | stressed: io, tu, egli, lui, noi, voi, essi\ unstressed: lo, la, mi, ci, vi | io parto\ lo mangio |
| PI | indefinite pronoun | chiunque, ognuno, molto | chiunque venga\ i diritti di ognuno |
| PP | possessive pronoun | mio, tuo, suo, loro, proprio | il mioè qui\ più bella della loro |
| PR | relative pronoun | che, cui, quale | ciò che dice\ il quale afferma\ a cui parlo |
| PQ | interrogative pronoun | che, chi, quanto | non so chi parta\ quanto costa?\ che ha fatto ieri? |
| RD | determinative article | il, lo, la, i, gli, le | il libro\ i gatti |
| RI | indeterminative article | uno, un, una | un amico\ una bambina |
| S | common noun | amico, insegnante, verità | l'amico\ la verità |
| SP | proper noun | Monica, Pisa, Fiat, Sardegna | Monica scrive |
| SW | foreign noun | fazenda, mulieris dignitatem, weekend | una fazenda in Brasile\ il prossimo weekend |
| T | predeterminer | tutti i giorni | tutti i giorni\ entrambi i genitori |
| V | verb | mangio, avere, passato | il tempo passa\ le scriverò una lettera\ vengo domani |
| VA | auxiliary verb | sono stato, ho mangiato | il peggio è passato\ ho scritto una lettera |
| VM | modal verb | posso, devono | il peggio deve venire\ dovrò scriverle una lettera |
| X | residual class: it includes formulae,\ unclassified words, alphabetic symbols\ and the like | distanziare di piacce |
To use the segmenter, you need to have the text you're working with saved as a Unicode (UTF-8) text file. An example text file you can use for this exercise is a few lines from "Una questione privata" by Beppe Fenoglio.
You should also name the file without any spaces; e.g. instead of my example file.txt, name it my-example-file.txt, my_example_file.txt, myexamplefile.txt. Spaces mean something else when you're working with the command line, and while there's workarounds for using a file with spaces in its name, it's easiest to just avoid it.
It will be simplest for you to run the part-of-speech tagger if you copy your text file into the tint folder that you unzipped. If you're comfortable navigating file paths in the command line, you can skip this step.
On a Mac
Open a new terminal window, and navigate to the tint folder. Unless you saved it somewhere else, you should be able to enter:
cd Downloads/tint
Once you're in the folder, you can run ls and check for the name of your text file if you want to make sure you moved it into the right folder.
Run the following command, replacing yourtext.txt with the name of your text file:
cat yourtext.txt | ./tint.sh -o yourposoutput.txt -f textpro
A few lines of text should appear, ending with Tint is ready, before returning you to the command line. If everything worked correctly, you should see a new file, yourposoutput.txt, in the directory.
You can open the output in a plain text editor, which will display a series of columns.
To be able to sort the results more easily, you can use Excel -- but don't try to open the file directly in Excel, because Excel will mess up the encoding and non-ASCII characters will be jibberish.
Instead, open a new, blank document in Excel. Using hte top navigation bar, go to Data > Get eternal data > Import text file. Select the output (look in the folder with Tint), and in the "File origin" dropdown, choose Unicode (UTF-8). Your text should become readable in the preview. The rest of the default settings are fine, so click Finish.
When it asks where to put the data, choose the top left column (selected by default) and hit OK.
Your text will appear in a series of columns. The most important are:
- tokenid: this gives a unique ID to each word or piece of punctuation
- parserid: this gives an ID number indicating order within a sentence
- pos: the part of speech, using the tags described above
- entity: indicates whether the word has been flagged as a named entity (PER for person, LOC for location)
- deprel: universal dependency relation, further linguistic information about the word
You can sort your sheet by pos (column E) to see, for instance, all the nouns in the text. Note, though, that this will lose the original sentence-based structure of the file -- although you can always restore this by sorting by tokenid (column A).