Linking Science to Technology - Using Bibliographic References in Patents to Build Linkage Schemes

Size: px

Start display at page:

Download "Linking Science to Technology - Using Bibliographic References in Patents to Build Linkage Schemes"

Elvin Carroll
5 years ago
Views:

1 Page 1 of 5 Paper: Linking Science to Technology - Using Bibliographic References in Patents to Build Linkage Schemes Author s information Arnold Verbeek 1 Koenraad Debackere 1 Marc Luwel 2 Petra Andries 1 Filip Deleus 3 1 Catholic University of Leuven Research and Development division INCENTIM Groot Begijnhof Leuven Belgium Tel.: Fax: GSM: arnold.verbeek@econ.kuleuven.ac.be 2 Ministry of the Flemish Community Private Office of the Flemish Minister of Education and Training Albert II-laan Brussels Belgium marc.luwel@vlaanderen.be 3 Catholic University of Leuven Medical School Department of Neuro- and Psychophysiology Herenstraat 49 B-3000 Leuven filip.deleus@med.kuleuven.ac.be

2 Page 2 of 5 Linking Science to Technology - Using Bibliographic References in Patents to Build Linkage Schemes : There is a wide consensus about the key-role of technological progress in shaping economic development (Dosi and Fabiani 1994; Silverberg and Soete 1994; Nelson 1994; Nelson and Winter 1974). Science and scientific development are hereby often viewed as catalysts of technological progress, which in interaction with other macro-level factors, may finally lead to economic growth (Freeman, 1982). However, this relationship between scientific and technological progress is a complex one that is far from linear and straightforward. As a consequence, processes of knowledge creation and the different possible modes of knowledge diffusion are central themes in the ongoing debate on science, technology, and innovation and their interaction (Freeman 1982). This interaction has been the starting point for the theoretical development and empirical testing of knowledge production functions (Griliches, 1990). However, before one will be able to model an entire knowledge production function, a more detailed understanding of the interaction between science and technology (S&T) is needed. A central issue hereby is the quantification and modelling of the complex web of linkages and interactions between S&T development. If we can arrive at better methods to understand this complex web of interactions across different fields of S&T, our more ambitious goal of further disentangling the relationship between scientific and technological progress on the one hand, and economic growth on the other hand, may one day become true. Addressing the issue of S&T linkage therefore has important policy implications. At the same time, during the past two decades wide-ranging socio-economic and technological transformations have caused European governments to reformulate their policies for government-supported scientific activity. Growing constraints on public expenditure also increased the need for greater accountability and effectiveness in all areas of the public domain. The complexity of the socio-economic environment thereby makes the identification of priorities in scientific and technological areas challenging. A thorough understanding of the S&T interaction is therefore essential for policy-makers. It is this interaction, and more precisely, the overall landscape of the S&T relations, which will be discussed in this paper. We shall present the preliminary results of a project funded by the DG Research of the European Commission, entitled Linking science to technology: bibliographic references in patents. The study is situated within the Fifth Framework Programme of the European Community for Research, Technological Development and Demonstration Activities ( ), and specifically the fifth action Support for the Development of Science and Technology policies in Europe.

3 Page 3 of 5 The main objective of this project is the elaboration of a methodology enabling us to model the linkage patterns between scientific and technological fields over longer periods of time. The methodology developed is based on the use of self-organising clustering methods that are applied to the European patent data (EPO), the United States patent data (USPTO) and the Science Citation Index (SCI) bibliographic databases, produced by the Institute for Scientific Information (ISI). The methodology is applied to a number of specific technological fields by way of a prototype test. Subsequent to this prototype test, the concrete implementation of this methodology for the USPTO and EPO data will follow. The central units of analysis in this clustering approach are the bibliographic citations in the patent records (Narin and Noma 1985; Narin and Olivastro 1992; 1998). In developing this approach, we of course have to acknowledge its limitations (Meyer, 2000). The most frequently used indicators for measuring technological and scientific output are based on patent and publication data. This same approach is followed in our research. Both, patents and publications, are distant measures, proxies, of technological and scientific activity. Despite their shortcomings, these indicators are the best available and therefore the most widely used. The qualitative understanding of the linkage patterns between science and technology has grown. In this respect we refer to and discuss the linear model versus the network model of knowledge transfer applied to the S&T interaction and linkage, and the implications these models have on the interpretation of interconnected scientific and technological domains. Until the 1990 s the amount of quantitative data, especially high level data, available to characterise and substantiate this relationship and to elaborate the subject as well as the national, international and temporal patterns in the coupling between science and technology, was rather limited. In order to further enhance our understanding on modelling the link between S&T fields we present a methodology that is based on a direct linkage approach. Hereby, we examine the sets of non-patent references (NPR s) present in patent documents, more specifically the references to scientific publications, as a direct criterion to connect scientific domains (represented by scientific publications) to technological domains (represented by patent documents). Scientific publications cited in patents are being traced and inventoried both in the EPO and in the USPTO databases over the period These citations are then processed and matched with the original journal publications covered by the SCI. The matching of citations to serial literature and the source papers in the SCI is achieved using a key based on the combination of the bibliographic fields {lead author name}, {year}, {volume}, {number/issue}, {starting page}. These fields are both present in the citations and the original source papers. The process of standardisation and unification of these fields, an essential building block in the linkage approach, is also discussed as well as the accuracy of the matching mechanism. This is an essential step since individual patents and publications have to be aggregated into broader and relevant domain-level clusters before useful linkage matrices can be developed.

4 Page 4 of 5 Once these domain clusters are identified and validated, the linkage patterns between technometric clusters as they arise out of the patent databases and scientometric clusters as derived from the publication database, can be modelled. This modelling is done using both traditional statistical clustering algorithms as well as self-organising neural-net based maximum entropy clustering methods. In this paper we discuss and illustrate this methodology on sets of patents extracted from the EPO and USPTO databases covering the period We subsequently demonstrate the creation of a static linkage scheme providing an extensive overview of technological domains (IPC 4-digits), and their science-based associations, based on combinations of SCI subject categories. In a further step, we discuss the results of the neural network clustering to identify groups of interrelated technological and scientific domains based on the referencing pattern between patents and publications. As a consequence, a form of meta-classification of interlinked S&T domains might be achieved. To further support policy makers in the elaboration of a S&T policy framework, the linkage methodology and its application need to be integrated in a decision support tool. We finally present and discuss further enhancements consisting of a number of related and value-creating steps towards policy-making on specific technological and scientific areas. These enhancements, among others, encompass the analysis of the S&T linkage scheme and its temporal evolution, the selection of technological domains of interest and subsequent analysis in terms of patenting and publication activity of the different actors.

5 Page 5 of 5 References Dosi, G. and S. Fabiani (1994), Convergence and Divergence in the Long-term Growth of Open Economics, In: Silverberg & Soete (eds.) Freeman, C. (1982), The Economics of Industrial Innovation, Pinter, London Griliches, Z. (1990), Patent Statistics as Economic Indicators: A Survey, Journal of Economic Literature, 28, pp Meyer, M. (2000), Does science push technology? Patents citing scientific literature, Research Policy, 29 (3), pp Narin, F., and D. Olivastro (1992), Status Report: Linkage Between Technology and Science, Research Policy, 21, pp Narin, F., and D. Olivastro (1998), Linkage Between Patents and Papers: An Interim EPO/US Comparison, Scientometrics, 41, pp Narin, F., and E. Noma (1985), Is technology becoming science?, Scientometrics, 7 (3-6), pp Nelson, R.R. (1994), What has been the Matter with Neoclassical Growth Theory?, In: Silverberg & Soete (eds.) Nelson, R.R. and S.G. Winter (1974), Neo-classical vs Evolutionary Theories of Economic Growth: Critique and Prospectus, Economic Journal, pp Silverberg, G. and L. Soete (eds.), (1994), The economic growth and technical change: technologies, nations, agents, Elgar Aldershot Keywords: S&T linkage, S&T policy, patents, publications, non-patent references

Science and technology interactions discovered with a new topographic map-based visualization tool

Science and technology interactions discovered with a new topographic map-based visualization tool Filip Deleus, Marc M. Van Hulle Laboratorium voor Neuro-en Psychofysiologie Katholieke Universiteit Leuven