Term Neighborhood Extractor

This node extracts the neighborhood for each term of the document. The neighborhood is defined by the N left and N right neighbors that can be set in the dialog. The neighbors can be extracted as term cells or string cells. Each neighbor is stored in a column, however it's also possible to store the right and left neighbors in a collection cell. Additionally, there is an option to extract the sentence in which the term occurs.

Options

Document column

Select the document column that is used for term neighborhood extraction.

Neighbors to extract

The number of (left/right) neighbors to extract.

Extract sentences

If checked, the second column contains the sentence of the document containing the term.

Neighbor Terms as Strings

If checked, neighbors will be extracted as String columns, otherwise as Term columns.

Neighbors as Collection

If checked, the neighbors will be all encapsulated in a collection (one collection for left neighbors, one collection for right neighbors), otherwise there will be one column for each neighbor. If no neighbor exists, the value is a missing value.

Input Ports

The input table containing the documents.

Output Ports

An output table containing documents, terms and their N neighbors.

Best Friends (Incoming)

• Strings To Document (14 %) Streamable
• String To Term (11 %) 
• Column Filter (6 %) Streamable
• Row Filter (5 %) Streamable
• Stop Word Filter (5 %) Streamable
• Show all 47 recommendations

Best Friends (Outgoing)

• Term To String (10 %) 
• Column Filter (10 %) Streamable
• Row Filter (10 %) Streamable
• Term Co-Occurrence Counter (8 %) 
• TF (5 %) 
• Show all 46 recommendations

Workflows

• 24_​Analyzing_​Twitter_​Posts_​with_​Custom_​Tagging (KNIME Hub) 
• POS_​Tag_​Sequences_​Extraction (KNIME Hub) 
• Simple Text Autocompletion (KNIME Hub) 

Installation

To use this node in KNIME, install KNIME Textprocessing from the following update site:

A zipped version of the software site can be downloaded here. Read our FAQs to get instructions about how to install nodes from a zipped update site.

Developers

You want to see the source code for this node? Click the following button and we’ll use our super-powers to find it for you.