mahout-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Dan Brickley <dan...@danbri.org>
Subject Re: Named Entity Extraction.
Date Sun, 10 Jun 2012 01:35:26 GMT
On Sunday, 10 June 2012, Lance Norskog wrote:

> OpenNLP has a Hidden-Markov-Model NER implementation. Stanbol seems to
> be a semantics server.


Yes, seems to lean heavily on OpenNLP but there are some other engines
listed that might be worth a look:
http://incubator.apache.org/stanbol/docs/trunk/enhancer/engines/list.html

Dan




> Aho-Corasick is string search, not NER.
>
> Is there something about NER that would require distributed computation?
>
> On Sat, Jun 9, 2012 at 5:45 AM, Suneel Marthi <suneel_marthi@yahoo.com<javascript:;>>
> wrote:
> > Also take a look at Apache OpenNLP (http://opennlp.apache.org) for NER
> implementations.
> >
> >
> >
> > ________________________________
> >  From: Dan Brickley <danbri@danbri.org <javascript:;>>
> > To: "user@mahout.apache.org <javascript:;>" <user@mahout.apache.org<javascript:;>
> >
> > Sent: Saturday, June 9, 2012 7:07 AM
> > Subject: Re: Named Entity Extraction.
> >
> >
> >
> >
> >
> > On 9 Jun 2012, at 12:09, Gaurav Sehgal <gaurav_sehgal@yahoo.com<javascript:;>>
> wrote:
> >
> >> Hello,
> >>         We are planning to deploy Mahout in our systems; to use named
> entity extraction; Is there a specific algorithm such as Aho-Corasick
> Algorithm; which Mahout implements for NER. I went through the Mahout
> website and it dosen't seem to have any NER related algorithm.
> >
> > Try Apache Stanbol for NER. Collocations are the closest thing in Mahout
> afaik.
> >
> > Dan
> >
> >>
> >> Thanks for your help,
> >> Gaurav
>
>
>
> --
> Lance Norskog
> goksron@gmail.com <javascript:;>
>

Mime
  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message