Probability PR details
----------------------

1. Introduction

The probability PR is used to obtain the probabilities for the occurances of each of the named entity in the corpus.

Following describe the basic steps to use it

First build the corpus either manually adding documents or querying google or crawling the WWW

Once the corpus is build load ANNIE

perform any processing required using any JAPE Transducers, etc.

Now load the ProbabilityPR via the manage plugins menu

Once loaded create a pipeline or add to the existing pipeline.

In the runtime parameters you would see the following options:

a) the Entity Type to be used. (Note that you can use only entities in the Default Annotation Set)
   eg Person, Location, Organization.

b) The processed corpus to be used.

Once run the output in the messages tab would correspond to the basic probabilistic outcomes of occurance for each entity.

Please note that there is no co-reference resolution done.