Documentation Index

Fetch the complete documentation index at: https://academy.insiderone.com/llms.txt

Use this file to discover all available pages before exploring further.

Eureka: Search Engine

Prev Next

Eureka enables you to provide an experience that helps your users find the most relevant product/content for their search terms as quickly as possible. It realizes this goal through its advanced search engine. Eureka’s advanced search engine achieves its relevance goal through query-term matching and ranking algorithms.

Eureka provides the most relevant, fastest query-term matching by processing your data during indexing and running matching algorithms when a user queries a term.

You can also refer to the video below to learn how Eureka helps:

If this video does not load in your region, you can watch it on our YouTube Channel.

How Eureka analyzes your data and product attribute words

It analyzes your data and your product attribute terms while indexing. It analyzes the word structure of the product attribute values, applying:

  • Uppercase, lowercase analysis

  • Conversion of  alphabetic, numeric, and symbolic characters that are not in the Basic Latin Characters (ASCII folding)

  • Sub-words of word analysis

  • Stop words

  • Synonyms

  • Enabled facets

Languages supported in Eureka

The languages currently supported by Eureka are English, French, Dutch, Turkish, Polish, Portuguese, Spanish, Russian, Ukrainian, Arabic, Kurdish, Romanian, Czech, Greek, Bulgarian, Danish, German, Italian, Hindi, Indonesian, Thai, Chinese, Japanese, and Korean.

How Eureka analyzes your users' search terms

Eureka analyzes your users’ search terms and finds results by applying:

  • Typo correction

  • NLP-based matching algorithms range from prefix analysis to Fuzzy search on searchable attributes such as category, brand, color, description, and name.

  • Your users' filters

How Eureka ranks search results

Eureka ranks search results depending on:

  • Intelligent ranking according to Insider One's relevancy scores

  • ML-based product scoring based on user activities

  • The rules that you provide via Insider One's merchandising editor