A Rule-Based Framework for Assessing Industrial Applicability

Patent searches are conducted in multiple independent databases. In addition, it is necessary to mention that patent search identifies the novelty and inventive step criteria among the criteria of patentability. Hence, the model of patent search cannot define the industrial application criterion. As can be seen, the proposed structure clearly illustrates how the industrial applicability evaluation may be performed based on a series of predefined rules. In contrast to machine learning algorithms, which require large amounts of data, the offered approach may rely solely on legal and scientific criteria already used by patent offices in practice. The proposed framework enables different types of automation in conducting patent evaluations: Fully Automated systems; Semi-Automated systems; Partially automated systems. Using rule-based algorithms enables providing an explanation for each step made by the system.