SPL File IdfCorpusBuilder.spl

Toolkits > com.ibm.streamsx.nlp 1.9.2 > com.ibm.streamsx.nlp > IdfCorpusBuilder.spl

Content

Operators

• IdfCorpusBuilder: This operator trains the model (aka corpus) for TfIdfWight based on training data.

Types

• IdfCorpusBuilder.IdfCorpusType: list of terms of one document to build the corpus

Composites

composite IdfCorpusBuilder(input stream<IdfCorpusType> TermsStream)

This operator trains the model (aka corpus) for TfIdfWight based on training data. This composite can be used to prepare the corpus in a training application. The corpusFile is read by the TfIdfWeight to build the internal data structures.

Parameters

• corpusFile: The output file as input for the TfIdfWeight operator.

Input Ports

• TermsStream: Tuples containing list of terms. One tuple is one document.

Static Types

IdfCorpusBuilder.IdfCorpusType = tuple<list<rstring> terms>;

list of terms of one document to build the corpus