Ah OK, I thought you mentinoed the NE model because you was answering my mail .. and sorry, I never used the classifier myself.
Am 13.11.2013 21:47, schrieb Charles Martin: > Tom > I am refering to the document classifier, not the named entity extract > > It is very difficult to use this tool in a production environment > without these details > IfI can't get answers to these questions, I will be forced to stop using > the tool > and return to a standard SVM > > Charles > > > > On Wed, Nov 13, 2013 at 12:37 PM, Thomas Zastrow > <[email protected]>wrote: > >> Dear Charles, >> >> There is no more detailed description about the NE model. The raw data >> is available on the homepage and I created the models following the >> OpenNLP documentation: >> >> http://opennlp.apache.org/documentation/1.5.3/manual/opennlp.html#tools.namefind.training >> >> Best, >> >> Tom >> >> >> >> Am 13.11.2013 18:18, schrieb Charles Martin: >>> Where I can i find a detailed description of the model.bin file and the >>> scoring functions for the document classifier? >>> >>> >>> >>> >>> On Wed, Nov 13, 2013 at 1:32 AM, Thomas Zastrow <[email protected] >>> wrote: >>> >>>> Am 12.11.2013 21:19, schrieb Jörn Kottmann: >>>> >>>>> On 11/12/2013 09:11 PM, Thomas Zastrow wrote: >>>>> >>>>>> Are there any plans to merge the repositories? Any other sources for >>>>>> models? >>>>>> >>>>> >>>>> There are a few people here and there publishing models on their sites, >>>>> we should add a model link page to our wiki, so people can easily find >>>>> them. >>>>> >>>>> Jörn >>>>> >>>> I think this is a good idea. The main point here is in my oppinion, that >>>> most models were created by using copyright protected material, which >> puts >>>> them into a grey zone. Thats the reason why many people are not >>>> distributing them. >>>> >>>> But anyway, having a list of models would be great. >>>> >>>> Best, >>>> >>>> Tom >>>> >>>> -- >>>> Dr. Thomas Zastrow >>>> Riemerfeldring 7a >>>> >>>> 85748 Garching >>>> Tel.: 0162 422 8029 >>>> www.thomas-zastrow.de >>>> >>>> >>> >>> >> >> > >
