Hi Alessandro,
I'm so happy there is someone who's done extensive work with QAC here!
Right now, we measure nDCG via a Dynamic Bayesian Network. To break it down,
we:
- use a DBN model to generate a "score" for each query_url pair.
- We then plug that score into a mathematical formula we foun
Does "_nest_path_" come back in a normal search? I would expect that the
fields that are returned by normal searches would also work in SQL. If that
turns out to be the case you could derive the fields from performing a
search and seeing what fields are returned.
Joel Bernstein
http://joelsolr.bl
Hi Audrey,
As suggested by Erik, you can index the data into a seperate collection and
You can instead of adding weights inthe document you can also use LTR with
in Solr to rerank on the features.
Regards,
Lucky Sharma
On Fri, 24 Jan, 2020, 8:01 pm Audrey Lorberfeld - audrey.lorberf...@ibm.com,
This is a really cool idea! My only concern is that the edge case
searches, where a user knows exactly what they want to find, would be
autocomplete into something that happens to be more "successful" rather
than what they were looking for. for example, i want to know the legal
implications of ja
Hi Audrey,
As suggested by Erik, you can index the data into a seperate collection and
You can instead of adding weights inthe document you can also use
LTR(Learning to Rank) with in Solr to rerank on the documents.
And also to increase more relevance with in the Autosuggestion and making
positiona
So I haven't looked at this in a few years, but the columns should be
registered in the SQL catalog so you should be able to ask via SQL for all
the columns.
describe table or using the JDBC metadata should work.
There may be some edge cases where depending on sharding you get into a
case where t
David,
True! But we are hoping that these are purely seen as suggestions and that
people, if they know exactly what they are wanting to type/looking for, will
simply ignore the dropdown options.
On 1/24/20, 10:03 AM, "David Hastings" wrote:
This is a really cool idea! My only concern is
Hi all!
Consider a query containing fq-params like this:
"fq":["tenant_id:1",
"u_markedAsDeleted_b:false",
"u_id_s:[* TO *]",
"(u_lastLendingDate_combined_ls_ns:([8610134693 TO 8611935823]))"]
This gives me a list of users, having a last lending date (somewhat encoded as
having
fq=NOT field:value
works for me,
On Fri, Jan 24, 2020 at 10:56 AM Sebastian Riemer
wrote:
> Hi all!
>
>
>
> Consider a query containing fq-params like this:
>
>
>
> "*fq*":["tenant_id:1",
>
> "u_markedAsDeleted_b:false",
>
> "u_id_s:[* TO *]",
>
> "(u_lastLendingDa
just tried "fq":"NOT year:[1900 TO 2000]"}},
on my data et and also worked as expected, mind if i ask why:
(u_lastLendingDate_combined_ls_ns:([8610134693 TO 8611935823]))
there are ()'s around your range query?
On Fri, Jan 24, 2020 at 11:01 AM David Hastings <
hastings.recurs...@gmail.com> wrote
On 1/24/2020 9:04 AM, David Hastings wrote:
just tried "fq":"NOT year:[1900 TO 2000]"}},
on my data et and also worked as expected, mind if i ask why:
(u_lastLendingDate_combined_ls_ns:([8610134693 TO 8611935823]))
there are ()'s around your range query?
I think David is correct here about th
Click-based weights are vulnerable to spamming. Some of us fondly remember when
Google was showing Microsoft as the first hit for “evil empire” thanks to a
click attack.
For our ecommerce search, we use the actual titles of books weighted by order
volume.
Decorated titles are reduced to a base t
Greetings.
A follow-up to the below with my findings.
The problem seems to be that the binding of “concat” on the
StreamFactory.functionNames maps to ConcatEvaluator in 8.4.1, while in 8.2 it
mapped to ConcatOperation.
The mapping is defined in the file Lang.java
Branch_8_4:
Branch_8_2:
The
Hi Dhanesh,
I have also encountered the problem long back when we have 'skimmed milk'
and need to search for 'skim milk', for that we have written one filter,
such that we can customize it, and then use KStemmer, then apply the custom
ConcatPhraseFilterFactory.
You can use the link mentioned below
Erik,
Thank you! Yes, that's exactly how we were thinking of architecting it. And our
ML engineer suggested something else for the suggestion weights, actually -- to
build a model that would programmatically update the weights based on those
suggestions' live clicks @ position k, etc. Pretty co
The concat function was changed to an evaluator. An example of the new
syntax is here:
https://github.com/apache/lucene-solr/blob/visual-guide/solr/solr-ref-guide/src/loading.adoc#unique-ids
Sorry for confusion the language should be settling down now with very of
these types of changes happening
16 matches
Mail list logo
