"wvi_corenlp" ? about nl2sql-rule HOT 7 CLOSED

wvi_corenlp -- where value index and corenlp lib cut the sentence.

What is keyword?

from nl2sql-rule.

{
"table_id": "1-1000181-1",
"phase": 1,
"question": "Tell me what the notes are for South Australia ",
"question_tok": ["Tell", "me", "what", "the", "notes", "are", "for", "South", "Australia"],
"sql": {
"sel": 5,
"conds": [
[3, 0, "SOUTH AUSTRALIA"]
],
"agg": 0
},
"query": {
"sel": 5,
"conds": [
[3, 0, "SOUTH AUSTRALIA"]
],
"agg": 0
},
"wvi_corenlp": [
[7, 8]
],
"bertindex_knowledge": [0, 0, 0, 0, 4, 0, 0, 1, 3],
"header_knowledge": [2, 0, 0, 2, 0, 1]
}

In this how did you get "wvi_corenlp": [[7, 8]], 7th & 8th words are SOUTH AUSTRALIA?
How did you extract "SOUTH AUSTRALIA"?
In my terms SOUTH AUSTRALIA is the keyword here.
What I know is CoreNLP takes the question and performs tokenization, what else does core NLP do?

from nl2sql-rule.

By string exact match.

from nl2sql-rule.

Are you searching the string(SOUTH AUSTRALIA) in the database and if the string is present in the database then you are returning the position in wvi_corenlp?

I thought the corenlp is extracting the word SOUTH AUSTRALIA?

What if SOUTH AUSTRALIA is not in the DB table?

from nl2sql-rule.

wvi_corenlp is the index in the question string.

from nl2sql-rule.

How does wvi_corenlp know to pick South Australia?

from nl2sql-rule.

wvi_corenlp is labeled by people.

from nl2sql-rule.