I have a large number of medical reports. I am trying to determine sentences that show a future action will be taken eg 'I will prescribe a medication'
I am using english-ewt model from udpipe and I have also tried english-gum but neither give me a future tense of the verb- just Tense Past/Pres
How get I determine a future sentence as the one above with udpipe (I am using this specifically as I have trouble installing rjava which openNLP and NLP require). If there is no future tense form of a verb given via udpipe are there other ways I can determine what I want using the POS tags etc that udpipe outputs?
I think this is a duplicate to question answered at determine the temporality of a sentence with POS tagging Let's clarify this further.
The verb will is a modal auxiliary, and it does not have a tense. English has 2 morphological tenses (https://en.wikipedia.org/wiki/Grammatical_tense#English), the present and the past. There is no future Tense.
In general, the tense concept is about the sentence, not about the individual words.
Future tense is formed by some conventions: e.g. a modal will/shall followed by an infinitive verb.
Summary: so you will need to combine POS tags with words themselves. So look to verbs where the dependency parsing output of udpipe links to an AUX term.
library(udpipe)
x <- udpipe('I will prescribe medication in the future', "english")
x[, c("token", "token_id", "upos", "xpos", "feats", "head_token_id", "dep_rel")]
token token_id upos xpos feats head_token_id dep_rel
I 1 PRON PRP Case=Nom|Number=Sing|Person=1|PronType=Prs 3 nsubj
will 2 AUX MD VerbForm=Fin 3 aux
prescribe 3 VERB VB VerbForm=Inf 0 root
medication 4 NOUN NN Number=Sing 3 obj
in 5 ADP IN <NA> 7 case
the 6 DET DT Definite=Def|PronType=Art 7 det
future 7 NOUN NN Number=Sing 3 obl