You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@lucene.apache.org by "thakkar.aayush" <th...@gmail.com> on 2015/01/23 04:53:23 UTC
Normailzing Job Title
I have master list of known job titles and associated job counts.
I am looking for ways to extract the same from the searched term. For
example:
Searched job title: Senior Digital Marketing Specialist
Extracted to: Senior Digital Marketing
Searched job title: Retail In-Store Sales Assistant; Full Time
Extracted to: Retail Sales Assistant
For boosting up the term search I have planned to tokenize the searched job
title and calculate:
1) The occurrence of the tokens taken 2 at a time and matching it with
master list of job titles.
2) The occurrence of token taken 1 at time and finding the job count.
Please suggest ideas for cleaning or extracting the job title. Should I look
for some other parameters as well for better results.
--
View this message in context: http://lucene.472066.n3.nabble.com/Normailzing-Job-Title-tp4181388.html
Sent from the Lucene - Java Developer mailing list archive at Nabble.com.
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@lucene.apache.org
For additional commands, e-mail: dev-help@lucene.apache.org
Re: Normailzing Job Title
Posted by "thakkar.aayush" <th...@gmail.com>.
Thanks for your suggestion Alex.
I am currently using lucene and found good results using search on title
alone. I will just boost a little based on job count to further improve on
the results.
--
View this message in context: http://lucene.472066.n3.nabble.com/Normailzing-Job-Title-tp4181388p4181402.html
Sent from the Lucene - Java Developer mailing list archive at Nabble.com.
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@lucene.apache.org
For additional commands, e-mail: dev-help@lucene.apache.org
Re: Normailzing Job Title
Posted by Alexandre Rafalovitch <ar...@gmail.com>.
Why can't it be just a Solr search against the collection of titles
(and counts)? It will manage all the approximate matches and ranking
for you, you just need to tell it the rules (e.g. with mm parameter).
Regards,
Alex.
----
Sign up for my Solr resources newsletter at http://www.solr-start.com/
On 22 January 2015 at 22:53, thakkar.aayush <th...@gmail.com> wrote:
> I have master list of known job titles and associated job counts.
> I am looking for ways to extract the same from the searched term. For
> example:
>
> Searched job title: Senior Digital Marketing Specialist
> Extracted to: Senior Digital Marketing
>
> Searched job title: Retail In-Store Sales Assistant; Full Time
> Extracted to: Retail Sales Assistant
>
> For boosting up the term search I have planned to tokenize the searched job
> title and calculate:
> 1) The occurrence of the tokens taken 2 at a time and matching it with
> master list of job titles.
> 2) The occurrence of token taken 1 at time and finding the job count.
>
> Please suggest ideas for cleaning or extracting the job title. Should I look
> for some other parameters as well for better results.
>
>
>
> --
> View this message in context: http://lucene.472066.n3.nabble.com/Normailzing-Job-Title-tp4181388.html
> Sent from the Lucene - Java Developer mailing list archive at Nabble.com.
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: dev-unsubscribe@lucene.apache.org
> For additional commands, e-mail: dev-help@lucene.apache.org
>
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@lucene.apache.org
For additional commands, e-mail: dev-help@lucene.apache.org