Relevance of search results
Many factors affect the relevance of search results.
Known-item searches
For known-item searches, Vega Discover uses specialized algorithms to handle single-word and exact-match serial and monograph titles. Ideally, Vega displays the patron-entered title as the first item in the search results.
Patrons can also search for specific identifiers such as ISBNs, ISSNs, call numbers, and other standardized numbers without the need to use an advanced search.
Open-ended searches
For open-ended searches, patrons can enter terms such as
- author,
- subject,
- title,
- serial title,
- secondary fields (for example, editionStatement, genreForm, tableOfContents, supplementaryContent, and other note fields), or
- a combination of the above terms.
Vega applies stemming so that searches for words like history also return results for histories.
When a search contains three or fewer terms, Vega only returns results that contain all specified terms. For searches that contain four or more terms, Vega first displays results that contain all search terms followed by results that contain some, but not all, terms, and indicates which terms are missing from each result. Patrons who want more exact matches for searches with four or more terms can use phrase searches with parentheses or include “AND” between the search terms.
Factors that affect relevance
The following factors affect relevance:
- If you search for a word or phase and a title exactly matches your search, all exact-match titles receive a high boost and should normally appear first in the search results.
- If you search for an author name and an author name exactly matches your search, all titles by the author receive a high boost and should normally appear first in the search results.
- If you search for an author name and the author has published a book in the current or previous year, those titles receive a high boost and should normally appear first in the search results. The boost for the current year is higher than the boost for the previous year.
- If a term is found in a highly prioritized field like Title or Author, the relevance score is higher than if the term were found in a secondary field, such as Description.
- If a term is found multiple times within a work, the score is higher.
- If a term is used infrequently in the field where it was found, the score is higher.
- If there are more matched terms in a work, the score is higher.
- If a work has many copies, the score is higher.
Because of the previous points, if a record has only one match, it doesn't automatically mean that the record will have a lower score than a record with two matches that occur in other fields.
The following table lists the metadata fields that Vega boosts for searches in priority order. The order may change as we continue to improve our relevance based on customer feedback and user testing.
Priority | Description | MARC | Vega BibFrame Field | Notes |
1 | Primary agent name |
|
agent:label |
Agent names are indexed as both Last, First and First, Last. Works published in the current or previous year receive an extra boost. Agent from 1xx is indexed as primary agent no matter what their relator field says. 880 for agent is indexed. |
2 | Primary work title |
|
primaryTitle |
When a title search is an exact match for an indexed title, the title receives an extra boost. 880 for title is indexed. |
3 | Work and instance titles |
|
title | Each unique title is indexed. |
4 | Work summary |
|
summary | — |
4 | Concept |
|
concept:label | — |
4 | Contributor secondary agent names |
|
agent:label | — |
4 | Series title |
|
hasSeries | — |
4 | Identifiers |
|
identifiedBy preferredCallNumber | — |
5 | Secondary fields |
|
genreForm editionStatement provisionActivity notes genreform |
Lowest relevancy boost. |
See also