From 5728c3950d74d3442ab488f6a6ce00b53d3268e5 Mon Sep 17 00:00:00 2001 From: Furkan Akkurt Date: Tue, 10 Sep 2024 21:42:52 +0300 Subject: [PATCH] Fix typos ; Update README.md --- README.md | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/README.md b/README.md index 679ed5c..565029c 100644 --- a/README.md +++ b/README.md @@ -140,7 +140,7 @@ tokens[,c('doc_id','sentence','token','clause','clause_fill')] ``` In the output we see that “Mary Jane” is labeled as subject, “loves” is -labeled as verb, but also that ALL the rest is labeled as object, +labeled as verb, but also that ALL the rest is labeled as verb, including “, and Mary is loved by John”. The reason for this is that by default, rsyntax will label all children of a matched token with the same label. We call this behavior the “fill” heuristic. In the @@ -157,8 +157,8 @@ would be to control what specific nodes to fill by nesting the `fill()` function. For example, we can say that for the subject and object we only want to ‘fill’ the tokens that form a multiword expression (MWE). In Universal Dependencies this is indicated with the ‘flat’, ‘fixed’ and -‘compound’ relations (see the (Universal Dependencies Relations -table)\[\]). Here we use the +‘compound’ relations (see the [Universal Dependencies Relations +table](https://universaldependencies.org/u/dep/). Here we use the fill function to specify that we only want to fill tokens where the relation has one of these values. Note that specifying lookup values in `fill()` works in the same way as in the `children()` function.