Configuring linguistic analyzers
Linguistic analyzers manipulate raw data before it is stored in the search index. Depending on the language and nature of the attribute a fitting analyzer should be chosen.
The Overview linguistic analyzers page displays:
- Name—the name of the analyzer.
- ID—determines the analyzer: Character Removal Filter, Character Transformation Filter, Advanced Tokenizer, Lowercase Filter, Stop Filter, Stem Filter, and so on.
- Delete box for removing the analyzer.
- Add pipeline button for developing custom text analysis components in addition to the pre-configured analyzers.
Fredhopper has several pre-configured analyzers (pipelines) onboard that you can attach to Search Fields. The configuration of Analyzers is global and they can be applied in each universe/locale. You can also develop custom text analysis components if needed by clicking Add pipeline and specifying the following:
- Details
-
- ID—indicates the pipeline as a number.
- Display name—indicates the name of the pipeline.
- Components
-
- Component—choose among Russian, CJK, Simple, Greek, Dutch, and so on.
- Description—explanation for the component
- Type—character filter, Tokenizer, Token filter
- Ranking—the bigger the number, the lower the place in the order.
- Action—option to add or remove components.