Annotations

The rules defined in the logx-reader grammars are run over annotated text, which is generated by a Processor to produce an annotated Document instance that can be serialized to JSON. Example JSON output for the text "How many TEUs of Frozen Meat are heading to Hamburg?" can be found in this Github Gist.

Processors

Currently there are three processors avalaible to the REST API: logx, CluProcessor, and ProxiedProcessor.

LogXProcessor

The logx processor uses a mixture of components from the clulab/processors library with custom pre- and post-processing hooks.

CluProcessor

CluProcessor comes directly from clulab/processors.

ProxiedProcessor

The ProxiedProcessor performs annotation via REST calls and allowing the user to easily mix components written in different programming languages. The ProxiedProcessor is defined in reader/src/main/scala/org/parsertongue/mr/processors. One can use SpaCy for annotation via the clu-spacy library).