doc_id
int64 | id
string | query
string | answer
string | label
sequence | token
sequence | target
string |
---|---|---|---|---|---|---|
0 | finerord0 | In the list of tokens, identify 'Person', 'Location', and 'Organisation' and label each accordingly. If the entity spans multiple tokens, use the prefix B-PER, B-LOC, or B-ORG for the first token, and I-PER, I-LOC, or I-ORG for the subsequent tokens of that entity. The beginning of each separate entity should always be labeled with a B-PER, B-LOC, or B-ORG prefix. If the token does not fit into any of the three named categories, or is not a named entity, label it as 'O'. Each line should contain one token and its corresponding label, separated by a colon. Do not combine tokens on your own. The format for each line should be: 'token:label'.
Text: H
ave
you
ever
felt
that
sudden
,
intense
dread
that
you
’re
about
to
die
?
Answer: | H:O
ave:O
you:O
ever:O
felt:O
that:O
sudden:O
,:O
intense:O
dread:O
that:O
you:O
’re:O
about:O
to:O
die:O
?:O | [
"O",
"O",
"O",
"O",
"O",
"O",
"O",
"O",
"O",
"O",
"O",
"O",
"O",
"O",
"O",
"O",
"O"
] | [
"H",
"ave",
"you",
"ever",
"felt",
"that",
"sudden",
",",
"intense",
"dread",
"that",
"you",
"’re",
"about",
"to",
"die",
"?"
] | H:O
ave:O
you:O
ever:O
felt:O
that:O
sudden:O
,:O
intense:O
dread:O
that:O
you:O
’re:O
about:O
to:O
die:O
?:O |