مقاله انگلیسی رایگان در مورد یادگیری بی نظیر نقشه برداری ثبت اختراع

مقاله سال 2016

Operationalising a company’s knowledge base in terms of its depth and breadth and creating trajectories to the future is challenging (Zhang and Baden-Fuller, 2010). The intensified complexity of emerging technologies (Breschi et al., 2003; Garcia-Vega, 2006) requires improved understanding of the nature and effect of cross-disciplinary activities in innovation processes (Wang and von Tunzelmann, 2000). Increasingly, companies must rely on broad knowledge bases covering diverse technology areas, while simultaneously having significant depth in their core competence. This creates a new type of tension for the management of technology and innovation. This is particularly problematic in highly dynamic industries. We examine the effects and potential of big data approaches in managing this increased complexity of company knowledge bases with a study on the telecommunication industry, and develop perspectives to exploit big data foresight approaches in support of strategic planning.

Previous studies on the depth and breadth of knowledge and technological trajectories have used patent information. As Moorthy and Polley (2010) point out, patents are the most feasible approach for analysing the breadth and depth of knowledge within a company as the data provides an insight to its competences. The simplest approach to quantifying the knowledge base is to use the patent classification scheme provided in the patent archive as a basis for evaluation – breadth correlating with the diversity in patent classifications and depth with the concentration of patent classifications in a company patent portfolio. This approach was used for example by Zhang and BadenFuller (2010) to analyse technology collaboration. Moorthy and Polley (2010) and SubbaNarasimha et al. (2003) use the approach to analyse the impact of breadth and depth of knowledge to company performance. Wu and Shanley (2009) operationalise the role of exploration in company knowledge stock by means of patent metrics.

Analysing classification metadata, in addition to citations, can be regarded as the de facto standard of utilizing patent metrics (e.g. Huang et al. (2015) as a case in point). This approach in analysing breadth and depth is not without limitations. Connecting patent classi- fications directly to industry sectors poses a challenge (Schmoch, 2008). Different patent classification systems have struggled to establish a tool to clearly distinguish industries into specific classes, limiting the applicability of classifications for sectoral analysis. Classifications are also of limited value in directing inventive effort (Loh et al., 2006), which is understandable due to the information retrieval nature of patent classifications. Patent classifications are a tool for the patent process, and the human process related to assigning classes is valuable in the intellectual process, even to the extent that automated classifications fall short of providing similar results (Richter and MacFarlane, 2005).