Kreuchauff F., Korzinov V.

Scientometrics, 2017, Vol. 111, 743–772.

Abstract

Emerging technologies are in the core focus of supra-national innovation policies. These strongly rely on credible data bases for being effective and efficient. How- ever, since emerging technologies are not yet part of any official industry, patent or trademark classification systems, delineating boundaries to measure their early devel- opment stage is a nontrivial task. This paper is aimed to present a methodology to automatically classify patents as concerning service robots. We introduce a synergy of a traditional technology identification process, namely keyword extraction and veri- fication by an expert community, with a machine learning algorithm. The result is a novel possibility to allocate patents which (1) reduces expert bias regarding vested in- terests on lexical query methods, (2) avoids problems with citational approaches, and (3) facilitates evolutionary changes. Based upon a small core set of worldwide service robotics patent applications we derive apt n-gram frequency vectors and train a sup- port vector machine (SVM), relying only on titles, abstracts and IPC categorization of each document. Altering the utilized Kernel functions and respective parameters we reach a recall level of 83% and precision level of 85%.