Relevancy is very context dependent. Only you have the knowledge to
determine if a result is relevant for the query or not.
Lucene/elasticsearch's default TFIDF algorithm should theoretically give
you the relevant results. Do the results change if you tweak your query
slightly?
Recall is even harder than precision, because you would need to define
ahead of time what your golden set of records will be for a specific query.
Creating these golden records can be a challenge.
Ivan what I am doing is we have a set of good results and best results,if
our output is coming from those good or best result will be consider as
relevant record,
but what about those results which are not in good or best but can be
relevant, how can we calculate the relevancy to find the rest of the
result..?
On Wednesday, 5 February 2014 19:02:05 UTC+5:30, Navneet Mathpal wrote:
Hi,
This is actually not very related to elasticsearch, if i want to calculate
the precision and recall how can i do it .....following is the scenario.
the formula for precision is :- No of relevant document retrieved /
total no of document retrieved
how can we know that which of retrieved are relevant.
Is there is any approach we can follow to calculate which are relevant
result.
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.