EBSCO’s relevance ranking methodology is the foundation of our premier search and discovery platforms.
Delivering Relevant Results
EBSCO’s vendor neutral relevance ranking strategy uses numerous criteria, including term frequency, field weighting, exact title matching and content attribute boosting, to provide users with the most relevant results for each search query. The major contributing factor in relevance scoring is the frequency of the user’s search terms in matching database metadata and full-text records. EBSCO’s goal is to display the most relevant results on the first page.
The most influential fields used in our relevance ranking calculations are listed in order of influence below:
- Matches on subject headings
- Term appearance in the title
- Author-supplied keywords
- Keywords within abstracts
- Match on keywords in the full text
As an area of active and continuous development, we diligently tune and improve relevance ranking to give users the results they want for every search, in every context. A key aspect of our research is that it is tested and validated with human users. We realize that humans are the ultimate judge of relevance, and they vet changes we make to our algorithms.
Recognizing that the value of a search result is not derived only from search terms, EBSCO takes into account certain attributes when determining relevance. Specific attributes that are considered include:
- Currency in publication means that more recent articles will rank higher
- Type of publication in relation to the search terms
- Peer-reviewed status assists the academic researcher in quickly locating key information in their field
- Length of article ensures that when other factors are equal, a shorter article is considered less valuable