Knowledge Diffusion and the Dynamics of Citation Accrual

<p>The diffusion of knowledge through society proceeds like an invisible ripple that moves between agents through multiple information channels. However, some types of knowledge are recorded, systematised and digitised for the benefit of everyone. Patents and academic articles are examples of such codified knowledge. These documents also contain a common element that is utilised for linking new and established knowledge: citations. This thesis harnesses citations in patents and scientific articles as proxies for signifying the existence of knowledge flows between cited and citing documents, focusing primarily on the dynamics of citation accumulation and the mechanisms governing these dynamics. For this purpose, it is helpful to think of patents and their citations as nodes and links, respectively, in a network where new nodes join the network and distribute their citations among existing nodes. This mode of thinking leads directly to the question: How does the citation network grow? This thesis addresses that question both empirically and theoretically. Two mechanisms that can explain much of the observed citation dynamics are preferential attachment and node ageing. The former mechanism reflects the tendency for successful nodes (by citation count) to become even more successful, while the latter captures the propensity for knowledge to become obsolete over time. The independence of these phenomena is nontrivial, but has generally been assumed. We put this assumption to the test for both patent and scientific-article citation networks and found it to be generally true if precautions are taken to account for important context surrounding the meaning of citations. Achieving a clear separation of these mechanisms is found to be very useful both mathematically and empirically, as they can now be studied independently. Patents are particularly sophisticated documents, with various components holding specific legal meanings. Associating certain properties of these components with popularity in the form of citation accrual creates a rare opportunity to build a framework that can identify ex-ante node fitnesses and examine their effect on the growth of a citation network. We find that a significant portion of the preferential-attachment process observed in the patent-citation network can be attributed to basic properties of patents determined by their time of grant. Besides suggesting novel approaches towards estimating patent quality, the results of our work also provide a platform for gaining a deeper understanding of the various mechanisms that underpin the success-breeds-success dynamics ubiquitously observed in complex systems.</p>

Knowledge Diffusion and the Dynamics of Citation Accrual

<p>The diffusion of knowledge through society proceeds like an invisible ripple that moves between agents through multiple information channels. However, some types of knowledge are recorded, systematised and digitised for the benefit of everyone. Patents and academic articles are examples of such codified knowledge. These documents also contain a common element that is utilised for linking new and established knowledge: citations. This thesis harnesses citations in patents and scientific articles as proxies for signifying the existence of knowledge flows between cited and citing documents, focusing primarily on the dynamics of citation accumulation and the mechanisms governing these dynamics. For this purpose, it is helpful to think of patents and their citations as nodes and links, respectively, in a network where new nodes join the network and distribute their citations among existing nodes. This mode of thinking leads directly to the question: How does the citation network grow? This thesis addresses that question both empirically and theoretically. Two mechanisms that can explain much of the observed citation dynamics are preferential attachment and node ageing. The former mechanism reflects the tendency for successful nodes (by citation count) to become even more successful, while the latter captures the propensity for knowledge to become obsolete over time. The independence of these phenomena is nontrivial, but has generally been assumed. We put this assumption to the test for both patent and scientific-article citation networks and found it to be generally true if precautions are taken to account for important context surrounding the meaning of citations. Achieving a clear separation of these mechanisms is found to be very useful both mathematically and empirically, as they can now be studied independently. Patents are particularly sophisticated documents, with various components holding specific legal meanings. Associating certain properties of these components with popularity in the form of citation accrual creates a rare opportunity to build a framework that can identify ex-ante node fitnesses and examine their effect on the growth of a citation network. We find that a significant portion of the preferential-attachment process observed in the patent-citation network can be attributed to basic properties of patents determined by their time of grant. Besides suggesting novel approaches towards estimating patent quality, the results of our work also provide a platform for gaining a deeper understanding of the various mechanisms that underpin the success-breeds-success dynamics ubiquitously observed in complex systems.</p>

Patent citation network analysis: A perspective from descriptive statistics and ERGMs

Patent Citation Analysis has been gaining considerable traction over the past few decades. In this paper, we collect extensive information on patents and citations and provide a perspective of citation network analysis of patents from a statistical viewpoint. We identify and analyze the most cited patents, the most innovative and the highly cited companies along with the structural properties of the network by providing in-depth descriptive analysis. Furthermore, we employ Exponential Random Graph Models (ERGMs) to analyze the citation networks. ERGMs enables understanding the social perspectives of a patent citation network which has not been studied earlier. We demonstrate that social properties such as homophily (the inclination to cite patents from the same country or in the same language) and transitivity (the inclination to cite references’ references) together with the technicalities of the patents (e.g., language, categories), has a significant effect on citations. We also provide an in-depth analysis of citations for sectors in patents and how it is affected by the size of the same. Overall, our paper delves into European patents with the aim of providing new insights and serves as an account for fitting ERGMs on large networks and analyzing them. ERGMs help us model network mechanisms directly, instead of acting as a proxy for unspecified dependence and relationships among the observations.