"what is formula for Schur orthogonality relations?"

[kaushal2161]

I getting following:

Request {'measures': {'relevance': 0.5, 'accuracy': 0.5}, 'id': '6e7ab78d-4a1c-47d6-9c31-80e2bedfe459', 'trace': [], 'tree': PPP node "triple" {'object': PPP node "missing" {}, 'predicate': <PPP node "resource" {'value': 'formula'}>, 'inverse_predicate': <PPP node "list" {'list': []}, 'subject': PPP node "resource" {'value': 'Schur orthogonality relation'}}, 'language': 'en'}

But in subject it should be Schur orthogonality relations instead of Schur orthogonality relation.
removing 's from the subject don't know why?

One more thing related to different kind of questions having "–" in the question.
What is formula for Lippmann–Schwinger equation?" gives,

Request {'tree': PPP node "intersection" {'list': (PPP node "triple" {'inverse_predicate': PPP node "list" {'list': []}, 'predicate': PPP node "resource" {'value': 'formula'}, 'object': PPP node "missing" {}>, 'subject': PPP node "resource" {'value': 'Lippmann'}}, PPP node "resource" {'value': 'Schwinger equation'})}, 'id': '721d0843-dadc-4e79-b6ba-75941ae60380', 'measures': {'accuracy': 0.5, 'relevance': 0.5}, 'trace': [], 'language': 'en'}

partitioned the subject. Is there any way to get this together in subject like:
'subject': <PPP node "resource" {'value': 'Lippmann–Schwinger equation'}>

And last:
Could not parse the question like:
"what is formula for Wiman's sextic?" gives
WARNING:root:('case', 'unknown dependency')

[Tpt]

But in subject it should be Schur orthogonality relations instead of Schur orthogonality relation.
removing 's from the subject don't know why?

It's because we set nouns to singular in order to simplify lookup for Wikidata properties because usually predicates labels are singular nouns.

One more thing related to different kind of questions having "–" in the question. What is formula for Lippmann–Schwinger equation?" gives,...

what is formula for Wiman's sextic? WARNING:root:('case', 'unknown dependency')

It's probably because of wrong output from CoreNLP. We could try to create heuristics to fix this issues but a better way should be to improve CoreNLP/switch to an other parser.

[kaushal2161]

Ok thanks