Hi, i am beginner in elasticsearch, and have a doubt:
i am doing one search but the result not is much relevent.
I search "notebook" , but the document with " Bag black for notebook" is
not last item in result.... but this should be the last record because not
it begins with the word "notebook"
Elasticsearch is based on Lucene, so the concepts of scoring are the
same. Review them for a better understanding.
Turn on "explain" to understand why the documents are scored in the
way they are. The document for "Bag black for notebook" (6) scores
higher than the one for "Acer Notebook serie gammer". (4) Why? My
guess is term frequencies. Documents with fewer overall terms score
higher when using TDIDF scoring. "For" is a stopword, so document 6
only has 3 terms. Document 1 has the same score and number of terms.
Word position has no relevance with standard queries.
Here are some definitions for terms in Lucene:
Cheers,
Ivan
On Tue, Jul 31, 2012 at 2:51 PM, elasticSearchUserBeginner ti.honjoya@gmail.com wrote:
Hi, i am beginner in elasticsearch, and have a doubt:
i am doing one search but the result not is much relevent.
I search "notebook" , but the document with " Bag black for notebook" is
not last item in result.... but this should be the last record because not
it begins with the word "notebook"
First, send some information on your mapping and indexing configuration
(what analyzer you used for the fields)
Starting with a word does not mean that it will get a higher score (maybe
if you use some sort of query with positional information, but I'm just
guessing here). Have a look
on Apache Lucene - Scoring to have an idea how
scoring works.
Regards
On Tuesday, July 31, 2012 5:51:37 PM UTC-4, elasticSearchUserBeginner wrote:
Hi, i am beginner in elasticsearch, and have a doubt:
i am doing one search but the result not is much relevent.
I search "notebook" , but the document with " Bag black for notebook" is
not last item in result.... but this should be the last record because not
it begins with the word "notebook"
Apache, Apache Lucene, Apache Hadoop, Hadoop, HDFS and the yellow elephant
logo are trademarks of the
Apache Software Foundation
in the United States and/or other countries.