A multiple regression model for patent appraisal

[[abstract]]Purpose: Concerning the general patent trading mechanism, this paper proposes a systematic patent appraisal model to provide patent transferors and transferees a reasonable price suggestion of the target patent. Design/methodology/approach: Based on the appraisal factors (including the patent transferor, patent transferee, patent features, and patent trading specifications) and regression model, a patent trading system is developed with an automatic patent appraisal function. Findings: Based on the case study, it is found that the system performance is better if the non-critical factors can be identified and dropped out from the appraisal model. Practical implications: The proposed model and platform can enhance patent trading performance and, therefore, the enterprise R and D tasks can be accomplished more efficiently. Originality/value: This study proposes quantitative models of patent appraisal factors and a multiple regression model for patent appraisal to present an automatic patent price determination mechanism.[[fileno]]2020420010002[[department]]工工