Bibliographically coupled patents: Their temporal pattern and combined relevance

Abstract Bibliographically coupled patents reveal a temporal pattern associated with their ages (how long ago they have been issued) and spans (their distance in time). The coupling strength of the aged or long-spanned patent pairs may be inherently limited, especially for fields that reveal patent and reference expansions. This study proposes a simple measure, combined relevance (CR), to counter such impact. Unlike traditional measures, CR provides more balanced treatment to short- and long-spanned but favors more aged patents pairs. A fixed CR threshold may be more safely applied with a reduced possibility of erroneously removing patent pairs that are truly related. For observing long-term knowledge dissemination or tracing overall development trajectory, CR may be an alternative.