Neo4j

hedefalk · ‎05-25-2020

I'm trying to use search fuzzy and with proximity in combination. I found an answer for how to do it with Lucene here, but it doesn't seem to work:

Trying something like

CALL db.index.fulltext.queryNodes('myIndex',  '“bill~ killl~”~2')

doesn't work for me. I get Lucene parse errors:

Failed to invoke procedure `db.index.fulltext.queryNodes`: Caused by: org.apache.lucene.queryparser.classic.ParseException: Encountered " <FUZZY_SLOP> "~5 "" at line 1, column 22.
Was expecting one of:
    <EOF> 
    <AND> ...
    <OR> ...
    <NOT> ...
    "+" ...
    "-" ...
    <BAREOPER> ...
    "(" ...
    "*" ...
    "^" ...
    <QUOTED> ...
    <TERM> ...
    <PREFIXTERM> ...
    <WILDTERM> ...
    <REGEXPTERM> ...
    "[" ...

Is it at all possible to tune the fuzzy search within a proximity search? Because fuzziness seems to be there anyways but I can't add any score limits manually. For instance, this works and get matches:

CALL db.index.fulltext.queryNodes('myIndex',  '“bill killl”~2')

but I can't do

CALL db.index.fulltext.queryNodes('myIndex',  '“bill~0.8 killl~0.7”~2')

without getting parse errors.

lingvisa · ‎04-16-2021

I am having the same issue due to some special characters in the query:

CALL db.index.fulltext.queryNodes("NameIndex", 'killl~~2')

If I remove one of the '~', it won't report the problem. Some other characters that cause the same issue: '!'. Is there a solution to fix this?

lingvisa · ‎04-16-2021

This is a report here:

Can't the Neo4j Team fix this as suggested in the question by using the 'escape' function. Some special characters or punctuations marks are common in text.

Neo4j

How do I combine fuzzy and proximity search against Lucene?