The Single-Field Indexer node creates a searchable index from a selected string column of an input data table. By default, the index is based on the TeraEdit object, which enables efficient approximate string matching using edit distance.
The Index Type configuration allows users to choose between different indexing techniques:

• TeraEdit for single-word edit-distance matching
• TeraAuto for single-word matching. Fast but less fault-tolerant than TeraEdit.

The generated index can be passed to downstream nodes, such as the Approximate Index Matcher, to quickly retrieve similar values within large datasets. By storing the indexed values in a structured format, the node enables fast and scalable similarity lookups, even in the presence of spelling variations, typos, or inconsistent text data.

Options

Select Column to Index

Selects the string column from the input table that will be used to build the index. Only string-convertible columns are available.

Index Type

Specifies the indexing technique used to build the searchable index. Different index types optimize matching for different data characteristics:

• TeraEdit – edit-distance calculations and single-word strings.
• TeraAuto - fast, but less tolerant edit-distance calculations.

Representation of Indexed Strings

tbd

Views

This node has no views

Workflows

• Index Reader and Writer – Reusing Indexed DataKNIME Hub
• Single-field Indexing and Matching Overview ExamplesKNIME Hub

Links

• No links available

Developers