ClearNLP semantic Role Labelling + Entity Co referencing

[ssetty]

Hello, I have used clearNLP for semantic role labelling and entity coreferencing.

(1)Entity corefence - resolves pron-nouns with proper nouns I am not clear about how to interpret output of coreferencing - basically I require entity names and set of sentences or sentence numbers associated with this. What are clusters here?

rain@27.17 COMMON UNKNOWN SINGULAR Clusters: [-1, -1, -1, 1, -1, -1, -1, -1, 7, 1, 9, -1, 10, -1, 9, -1 Confidence: [1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0,

===== Clusters ===== 0: morning (Singleton) 3: He -> 1: Stewart

(2) How do interpret dependency parser output? Is there any documentation about labels used? Sometimes we see output with 7 slots and sometimes with 9 slots why? How TVRSReader recognise this?

(3) Adding semantic role annotation adds an extra slot - how do link semantic role outputs?

Thanks

[jdchoi77]

I think this issue is redundant to the thread opened in our group:

https://groups.google.com/forum/#!topic/emorynlp/yIlK0LW4A6A

Please follow up the thread. Thanks.