liveBook · Manning

Text tagger

As explained in AI-Powered Search

A text tagger is a search component that identifies indexed terms or entities occurring in an input text.

A text tagger identifies indexed terms or entities in incoming text, then maps matched surface forms to canonical entities for semantic query processing. section 7.4.1

A practical tagging pipeline:

  • Index domain phrases, spelling variations, cities, and semantic functions in an entities collection. section 7.4.1
  • Use an FST-backed lookup structure for compact, millisecond-scale matching across large entity lists. section 7.4.1

Tagging a location-aware query

For "top kimchi near charlotte", the tagger identifies "top", "near", and "charlotte". The resulting tagged query is "{top} kimchi {near} {charlotte}", while the typed tree maps these to popularity, location distance, and Charlotte.

How AI-Powered Search covers it

How deep it goes
Explains indexing, normalization, matching, disambiguation, and query-tree generation, but defers FST internals.
Examples
Demonstrates entity collections, Solr configuration, and tagging the query “top kimchi near charlotte.”
What it assumes
Assumes an indexed entity list and query text, while demonstrating entity schemas, lookup indexing, and query interpretation.
What you can do after
Operate/configure the Solr text tagger and transform extracted matches into tagged or typed search queries.
How it connects
Connects entity extraction with semantic search, finite state transducers, query parsing, canonicalization, and relevance functions.
Cover of AI-Powered Search
liveBookmanning.com
Read AI-Powered Search on liveBook Get the book at manning.com

Related concepts

This explainer was generated by Manning's knowledge graph from the text of AI-Powered Search. The section links point to the parts of the book it draws on.