How is the relevance ranking of articles calculated?

In Embase the relevance ranking of an article is based on an algorithm that combines the relevance scores for each search term. A query score is calculated per search term for each search.

The relevance score for each term is calculated from its term frequency and inverse document frequency, where the term frequency is its frequency of occurrence.

• Term frequency is normalized so that unless a field is specified, all article fields have the same weight
• It takes into account the size of the article (i.e., terms that occur 10 times score higher in a 100-word document than in a 1,000-word document).

The inverse document frequency is defined as log(1/d), where d (document frequency) is the number of articles in which the term occurs.

Therefore, the score is calculated based on the following formula:
score = log(term frequency) x (inverse document frequency).

Scores for each term are added together to provide a relevance ranking for each article .