Document Similarity Predictor

The Document Similarity Predictor applies the model obtained by the Document Similarity Learner to a test document. It computes the cosine similarity between the original corpus of documents table and the test document.

Options

Document column from Similarity Learner

The document column to which the similarity is calculated. It is part of the output from the "Document Similarity Learner" Component.

Document Vector column associated with documents

Select the document vector column which is associated to the documents selected beforehand. It is also the output from the "Document Similarity Learner" Component.

Preprocessed document column for similarity prediction

Select the column containing preprocessed documents that will be used to predict the similarity to the original documents.

Term column associated with preprocessed documents

Select the column containing the terms associated with the preprocessed documents.

Minimum similarity

Selects the minimum similarity values you would like to output.

Neighbor count

Selects the number of similar neighboring documents you would like to output.

Nodes

• Column Selection Configuration4 ×
• Breakpoint3 ×
• Column Rename3 ×
• Merge Variables2 ×
• Split Collection Column2 ×
• CASE Switch Data (End)1 ×
• CASE Switch Data (Start)1 ×
• Column Filter1 ×
• Component Input1 ×
• Component Output1 ×
• Document Vector Applier1 ×
• Double Configuration1 ×
• Integer Configuration1 ×
• Rule Engine1 ×
• Similarity Search1 ×
• String Manipulation (Variable)1 ×
• Table Row to Variable1 ×

Extensions

• FeatureKNIME Base nodes
• FeatureKNIME Distance Matrix
• FeatureKNIME Javasnippet
• FeatureKNIME Quick Forms
• FeatureKNIME Textprocessing

Links

• No links available