KB User's Guide - Semantic Search FAQ

This document answers some common questions about the new "semantic" search engine for the KB and how it compares to the original native search.

How does the new hybrid semantic search work, and how does it differ from the original search engine?

For an overview of how the two search engines compare, please see KB User's Guide - Understanding the New Semantic Search Engine.

What makes the new search engine a hybrid search?

When semantic search is enabled, the first set of results are a combination of semantic results and keyword results. By default, this first set will include up to 20 results, though this limit can be adjusted site-by-site.

When the results of the two engines are combined, any duplicates are removed. From there, the combined set of results are sorted for relevance using the following criteria (in order):

  1. Search priority ("top" priority docs are returned first, followed by "high", "normal", and "low" priority)
  2. Device matching (if the client OS / browser sort option is enabled, results matching the users device are prioritized)
  3. Keyword similarity (results in the combined set whose keywords match the search terms are prioritized)

Any additional docs matching the site's default search mode that were not included in the initial set of results, e.g. those that have exact matches in the title and keywords fields, will then be appended.

Are there certain types of searches where semantic search won't perform as well?

Because the semantic search engine is trained on a large language model, it will not perform as well when searching unique acronyms or other proprietary terms, such as product names. For example, it does not recognize words like "NetID" or "WiscVPN".

In other cases, it may understand a word, but in a different context from how it may be used in documentation. So, while it recognizes the word "canvas" as a word in the English language, it does not know that "Canvas" is the name of a learning management system and that it has a relationship with words related to online courses.

We are continuing to investigate ways to address this directly. In the meantime, results for queries containing these types of words will still appear after the initial set of semantic search results, i.e. they will be included as normal title/keyword results.

How do I enable semantic search for my site?

Currently, semantic search can only be enabled by members of the KB Team. Please contact us at kb-team@doit.wisc.edu if you would like us to enable semantic search for your external and/or internal site(s).

Can I disable semantic search for my site once it has been enabled?

Yesif you decide that you want to switch back to the traditional keyword+title search, please let us know, and we can disable semantic search for you.

How can I tell if semantic search is enabled for a site I am searching?

To avoid confusion for end users, we do not have any explicit indicators that semantic search is enabled. However, you can generally tell by observing one or both of the following:

  • A larger number of search results, with some results lacking exact search terms in the title and keywords
  • A slightly different search success/failure message (does not apply if you have customized this message)
    • Standard message: "Not finding what you are looking for? Suggest a new document be created, or try adjusting your search criteria."
    • Semantic search message: "Not finding what you are looking for? Suggest a new document be created, or try searching again with multiple search terms."

Can a user switch back to the basic search if semantic search is enabled on a site?

At this time, end users cannot disable semantic search while searching.

Will changing the search mode dropdown have an effect on my results?

Currently, interacting with the "search mode" dropdown will only affect the second set of results, i.e. those that load after the initial set of semantic search results. 

How can I report a possible issue with the semantic search engine?

If you are getting unexpected search results from the semantic search engine, or if you believe it is otherwise not functioning as expected, please contact us at kb-team@doit.wisc.edu

Do I need to change my existing documents to make them work with semantic search?

You do not need to, but there are some good things to keep in mind when authoring or updating docs:

  • Follow formatting and structure best practices, such as using headings and lists to organize your content
  • Use plain language in your documentation; avoid jargon and overly technical terms whenever possible
  • Make sure you write out acronyms at least once
  • Avoid overly long documents; split up long content into multiple documents where possible


Keywords:
nlp natural language processing searches differences between native hybrid new engine results filters
Doc ID:
142333
Owned by:
Leah S. in KB User's Guide
Created:
2024-09-11
Updated:
2024-12-12
Sites:
KB User's Guide