Information diffusion between patents and scientific articles for identifying future R&D and business opportunities: a case study in nano-science and nano-technology

Full text

(1)2009:051. MASTER'S THESIS. Information Diffusion between Patents and Scientific Articles for Identifying Future R&D and Business Opportunities - A Case study in Nano-Science and Nano-Technology. Mina Ali. Luleå University of Technology Master Thesis, Continuation Courses Marketing and e-commerce Department of Business Administration and Social Sciences Division of Industrial marketing and e-commerce 2009:051 - ISSN: 1653-0187 - ISRN: LTU-PB-EX--09/051--SE.

(2) MASTER'S THESIS Information Diffusion between Patents and Scientific Articles for Identifying Future R&D and Business Opportunities “A Case study in Nano-Science and Nano-Technology” Supervisors: Dr. M. Mehdi Sepehri (TMU). Dr. Peter Naude (LTU) Referee: Prepared by: Mina Ali Tarbiat Modares University Faculty of Engineering Department of Industrial Engineering Lulea University of Technology Division of Industrial Marketing and E-Commerce Joint MSc PROGRAM IN MARKETING AND ELECTRONIC COMMERCE. 2009. 1.

(3) Abstract Scientific progress and technological innovation have become major drivers of economic progress in the emerging knowledge economies where growth, productivity, and competitiveness are increasingly based on improved technologies, novel products, upgraded processes or customized services. The more important they became, the more the need was recognized to monitor their development, to examine the conditions under which they reach an optimal performance, and to formulate and carry out policies aimed at enhancing its performance and setting its priorities. However, in an era of rapidly expanding digital content, overwhelming data available on internet and the high speed of science and technology progress makes it difficult for human beings to extract useful knowledge without powerful tools. R&D teams rarely have the time or budget to read and understand all of the documents retrieved and, as a result, are increasingly overloaded with too much information. Strategic planning for science and technology development and R&D progress require having deep insight of science and technology evolutions. Information in patents and scientific articles are good source of codified knowledge. Monitoring their evolution can help managers, policy makers, researchers, scholars and industry experts in better decision making and planning for future more efficiently. Creating the framework for monitoring and tracing science and technology developments has lots of advantages. In macro view it helps policy makers and administrators develop their strategic plan, determine the R&D priorities and specify the knowledge gaps in different fields. In micro level it aids scientists, scholars and technicians getting acquainted with science and technology trends. It enables firms to respond to ever more sophisticated consumer demands and stay ahead of their competitors, both domestically and internationally. In this research, we try to investigate the relations between science and technology in nano disciplines. For this purpose scientific articles are considered as indicators of science and patents considered as indicator of technology. After studying different approaches and methods used for investigating the linkages between science and technology in different field of studies, we investigate this linkage by using two approaches, Analyzing Co-Occurrence of inventor-author names in patents and scientific articles, and lexical analysis by using text mining techniques. Our data consists of nano articles 2.

(4) published in Science Citation Index database from 1986 to the end of September 2008 and patent records of United States Patent and Trademark Office, (USPTO), from 1790 to 2008. By finding common inventor-author’s names, we identify in which nano sub disciplines there is high probability that scientific activity leads to register patents and obtain commercial applications. In lexical approach by using text mining techniques, we develop the framework for monitoring nano-science and nano-technology evolutions based on information in scientific articles and patents. We try to determine nano sub disciplines and study the trend of each group in specified time interval. We also draw the s-curve of each cluster for analyzing their future evolution. The outcome of this research is developing practical framework for monitoring scientific articles and patent evolutions. Preparing framework for monitoring and comparing patents and papers behavior simultaneously is rather new approach and consider as our main contribution in this study.. Keywords: Science-technology Relation, Science and Technology Strategic Planning, Information Diffusion, Text Mining, Information Visualization. 3.

(5) Acknowledgement This research was done during summer 2008 and spring of 2009. There are many individuals that without their contribution and cooperation, this research wouldn’t have been accomplished. I would like to foremost thank my supervisors Dr. Peter Naude from Lulea University and Dr. Mohammad Mehdi Sepehri from Tarbiat Modares University. Their strong support, patience, careful supervision and valuable guidance were great help in completion of this study. I dedicate my special gratefulness to Mr. Babak Teimourpoor, the PhD candidate of Industrial Engineering Department in Tarbiat Modares University. His cooperation in whole steps of this research has great share in progress of this research. I would also like to express my appreciation to my colleagues in Atinegaar Think Tank. Specially, I would like to thank Mr. Amir Nazemi, my honorable colleague in Atinegaar Company and PhD candidate in Allame Tabatabai University. Without his useful comments and valuable advice, this research wouldn’t have been completed. Finally, I would like to thank my family for their consideration, support and understanding during completion of this research. I dedicate this thesis to my great parents, friends and all students and scholars that findings of this research might be useful for them.. 4.

(6) Table of Content Chapter 1 ..................................................................................................................................... 10 Introduction.................................................................................................................................. 10 1.. Introduction .......................................................................................................................... 10 1.1. Introduction and Preface ............................................................................................ 10 1.2. Background of the Study ............................................................................................ 11 1.3. Importance of Topic..................................................................................................... 13 1.4. Problem Description .................................................................................................... 15 1.5. Research Objectives and Questions Identification ................................................ 16 1.6. Outline of the Entire Thesis........................................................................................ 17. Chapter 2 ..................................................................................................................................... 19 Literature Review ....................................................................................................................... 19 2.. Literature Review ................................................................................................................ 19 2.1. Science &Technology as Source of Competitive Advantage ............................... 19 2.2. Actors in Science and Technology System ............................................................. 20 2.3. Publications and Patents as Means to Analyze Science and Technology Interaction ................................................................................................................................ 22 2.3.1. Scientific publications as indicator of science.................................................. 24 2.3.1.1.. The Metrics of Bibliometrics ............................................................................ 24. 2.3.1.2.. Strengths and Weakness of the use of Bibliometrics ................................. 25. 2.3.2. Patents as Indicator of Technology ................................................................... 28 2.3.2.1.. Patent Definition................................................................................................ 29. 2.3.2.2.. The Structure of a Patent Document ............................................................. 29. 2.3.2.3.. Main Sources of Patents & Their Differences .............................................. 30. 2.3.2.4.. Patent Analysis ................................................................................................. 31. 2.3.2.5.. The Utilizations of patent analysis ................................................................. 32. 2.3.2.6.. Strengths and Weakness of Patent Analysis ............................................... 33. 2.4. Science and Technology Interaction ........................................................................ 35 2.5. Methods of Evaluating Science and Technology Interactions.............................. 37 2.5.1. Analyzing Industrial Scientific Publications Activity ........................................ 39 2.5.2. Analyzing University and Academic Patents ................................................... 41 2.5.3. Cross-Citation Analysis ....................................................................................... 42 5.

(7) 2.5.4. Lexical Approach by Using Text Mining Techniques...................................... 47 2.5.4.1.. Knowledge Discovery in Text (KDT) & Text Mining .................................... 48. 2.5.4.2.. Vector Space Model ......................................................................................... 50. 2.6. Conclusions .................................................................................................................. 50 Chapter 3 ..................................................................................................................................... 52 Research Methodology ............................................................................................................. 52 3.. Research Methodology ..................................................................................................... 52 3.1. Research Approach and Design Strategy ............................................................... 52 3.2. The overall Research Process .................................................................................. 54 3.2.1. Data Collection and Description ........................................................................ 56 3.2.2. Data Preprocessing ............................................................................................. 58 3.2.3. Data Analysis ........................................................................................................ 59 3.2.3.1.. Analyzing Co-Occurrence of Inventor-Author Names ................................ 59. 3.2.3.2.. Lexical Approach by Using Text mining Techniques .................................. 60. 3.2.3.2.1.. Text mining Framework ............................................................................... 61. 3.2.3.2.1.1. Text preprocessing & Matrix Representation ........................................... 61 3.2.3.2.1.2. Terms Weighting ........................................................................................... 62 3.2.3.2.1.3. Dimensionality Reduction ............................................................................ 65 3.2.3.2.1.4. Clustering ....................................................................................................... 67 3.2.3.2.1.5. Clusters Representation & Labeling .......................................................... 72 3.2.4. Interpretation and Validation of Results............................................................ 73 Chapter 4 ..................................................................................................................................... 74 Results and Analysis ................................................................................................................. 74 4.. Results and Analysis ......................................................................................................... 74 4.1. Results of Analyzing Co-Occurrence of Inventor-Author Names ........................ 74 4.2. Results of Lexical Approach by Using Text mining Techniques .......................... 89. Chapter 5 ................................................................................................................................... 118 Conclusions and Future Research Directions ..................................................................... 118 5.. Conclusions and Future research Directions ............................................................... 118 5.1. An Overview of Study ............................................................................................... 118 5.2. Findings ....................................................................................................................... 119 5.3. Managerial Implications ............................................................................................ 121 5.4. Research Limitations and Problems ....................................................................... 122 5.5. Recommendations for Future Research ................................................................ 123 6.

(8) References ................................................................................................................................ 125 Appendix 1 .................................................................................................................................. 136 Nano Discipline Key Words Used for Searching Data according to Web of Science Format 136 Nano Discipline Key Words Used for Searching Data according to USPTO Patent Database Format ..................................................................................................................................... 136 Appendix 2 ................................................................................................................................ 137 Part of Pivot Table used in Lexical Approach for Cluster Analysis ....................................... 137. 7.

(9) List of Figures Figure 1-1: Outline of the Entire Thesis ....................................................................................... 18 Figure 2-1: Different approaches of Triple Helix model .............................................................. 21 Figure 2-2: Co-authorship map of Swedish pharmaceuticals firm Astra ..................................... 40 Figure 2-3: Distribution of SCI papers citing patents ................................................................... 43 Figure 2-4: Knowledge Discovery in Text (KDT) Process .......................................................... 48 Figure 3-1: Research Design of the Study .................................................................................... 54 Figure 3-2: Number of Nano Articles published during 1896-2008 in Web of Science Database ....................................................................................................................................................... 57 Figure 3-3: Number of Patent Records Filed in United State Patent and Trademark Office from 1970 to 1995 ................................................................................................................................. 58 Figure 3-4: Number of Patent Records Filed in United State Patent and Trademark Office from 1996 to 2007 ................................................................................................................................. 58 Figure 3-5: A term frequency matrix showing the frequency of terms per document ................. 62 Figure 3-6: Linkages in Measuring Distance in Hierarchical Clustering ..................................... 68 Figure 4-1: Part of TF-IDF Matrix................................................................................................ 90 Figure 4-2: Mountain Visualization of Clusters Using gCLUTO Software ................................. 91 Figure 4-3: Results of Cluster Analysis in Each Year .................................................................. 93. 8.

(10) List of Tables Table 2-1: Researches with quantitative approach to science and technology evolutions ........... 22 Table 2-2: Bibliometric as an S&T Indicator. .............................................................................. 27 Table 2-3: Patents as S&T Indicator. ............................................................................................ 34 Table 2-4: Different ways of measuring science-technology interaction. .................................... 37 Table 4-1: The Common Author’s/Inventor’s Names with their International Patent Class and Field Date in High Citation Sample .............................................................................................. 75 Table 4-2: The Distribution of Four Digit International Patent Classes in High Citation Sample 76 Table 4-3: The Common Author’s/Inventor’s Names with their International Patent Class and Field Date in Random Samples..................................................................................................... 80 Table 4-4: The Distribution of Four Digit International Patent Classes in Random Samples ...... 83 Table 4-5: Cluster's Specifications ............................................................................................... 91 Table 4-6: Part of Pivot Table Used for Analyzing Trend of Evolution in Nano Sub Disciplines ....................................................................................................................................................... 94 Table 4-7: Cluster's Labels ......................................................................................................... 115 Table 5-1: The probability of finding common inventor’s author’s names in different chosen samples........................................................................................................................................ 120 Table 5-2: Results of Counting 2 digit IPC Code for sample include authors of articles with highest citation rank .................................................................................................................... 120 Table 5-3: Results of Counting 2 digit IPC Code for random samples ...................................... 121 . 9.

(11) Chapter 1 Introduction. 1. Introduction This chapter tries to provide the readers with an insight to the research area. It begins with brief introduction to the subject of study. Next the background of research area and importance of topic will be described. Problem description will be followed later; we will then guide the readers to the research questions and objectives. Finally the structure of entire thesis will be presented.. 1.1. Introduction and Preface During the course of the twentieth century, particularly after the Second World War, science and technology have become driving forces in society and vehicles of economic growth and development (Moed et al 2004; Guan and He 2007). Science and Technology are major drivers behind economic progress in the emerging knowledge economy. The more important they became, the more the need was recognized to monitor their development, to examine the conditions under which they reach an optimal performance, and to formulate and carry out policies aimed at enhancing its performance and setting its priorities (Moed et al 2004). Information diffusion between science and technology is the adaptation of information and flow of knowledge, between science and technology. Information flows, or interaction in a more general sense, between science and technology takes many different forms, each associated with specific channels and types of indicator. For example, knowledge and information may be transferred by means of personal contacts at conferences and workshops, or by mobility (change of jobs) of researchers, by (graduate) students, by joint research projects, or by publication channels such as scientific articles and patents (Nomaler and Verspagen 2007). The relationship between science and technology becomes one of the crucial issues for science policy guidance, innovation and economic studies. It is generally accepted that science 10.

(12) and technology are closely connected, interacting and interdependent (Guan and He 2007). Academics and policy makers are investigating the relations between science and technology in different disciplines. A number of efforts on theory and model exploration as well as empirical studies have been undertaken to uncover the nature and type of the relation and interaction between science and technology (Meyer 2001; Meyer 2002; Verbeek et al. 2002; Bhattacharya and Meyer 2003; Looy et al. 2003; Guan and He 2007; Nomaler and Verspagen 2007). In literature two approaches for studying the information diffusion between science and technology have been identified. The ‘indirect linkage approach’ and the ‘direct linkage approach’. The former tries to understand the S&T linkage via mobility of scientists or engineers and the latter uses bibliographic information in publication and patent documents (Verbeek et al. 2002). In this study, we focus on direct linkage approach and try to extend it by means of contentbased lexical methods. We investigate the linkage between science and technology by using two approaches. First we match author’s name in Web of Science publication data base with the inventor’s name in patent data base from United State Patent and Trademark office (USPTO). Lexical approach will be considered as our second approach. Articles and patent’s data in nano discipline are used in a quantitative methodology for monitoring the evolution of science and technology in nano discipline. Text-mining techniques will be utilized to extract the sub disciplines in the emerging field of nano science and technology and test whether changes in nano science and nano technology follow the same route or not.. 1.2. Background of the Study The relationship between science and technology becomes one of the crucial issues for science policy guidance, innovation and economic studies. A number of efforts on theory and model exploration as well as empirical studies have been undertaken to uncover the nature and type of the relation and interaction between science and technology (Price 1965; Meyer 2000; Meyer 2001; Deleus and Hulle 2002; Meyer 2002; Verbeek et al. 2002; Bhattacharya and Meyer 2003; Deleus and Van Hulle 2003; Looy et al. 2003; Guan and He 2007; Nomaler and Verspagen 2007; Boyack and Klavans 2008).. 11.

(13) For a long time, the science and technology (S&T) relationship was depicted through the “linear model” which emphasized the technological progress dependence on scientific discovery (Narin and Olivastro 1992). Some retrospective studies were conducted based on this “linear model” for revealing the contribution of basic research to technology development (Sherwin and Isenson, 1967; Mowery and Rosenberg, 1982 cited by (Guan and He 2007)). However, the “linear model” unilaterally stressed the influence of scientific research on technology development and ignored that technology often shaped science in important ways (Bhattacharya and Meyer 2003). Toynbee (1963) compared the S&T interaction to a “pair of dancers” and he felt it difficult to judge whether science pushes technology or the opposite. As Toynbee (1963) also has pointed out, it is difficult to distinguish between science and technology (Toynbee 1963). Consequently, later studies pointed out the multifaceted relationship between science and technology and the “linear model” was replaced by the “network model”. The “network model” is definitely more adequate to characterize the complex relation between actors in S&T system and led to the insight that it is too simplistic to think of the technology dependence on science (Pavitt, 1997; Steinmueller, 1994; David et al., 1997 cited by (Guan and He 2007)). During the second half of the 1990s, the Triple Helix model drew attentions on the interaction between industry, academia and government and was widely used in science and innovation policy (Etzkowitz and Leydesdorff 1998; Leydesdorff and Etzkowitz 1998). Bilateral relations between government and university, academia and industry and government and industry in Triple Helix model have expanded into triadic relationships among the spheres. Where the institutional spheres overlap, collaborate and cooperate with each other (Etzkowitz 2002). Science and technology sources in the form of articles and patents are valuable sources of knowledge. They contain important research results that are valuable to the researchers, scholars, industry experts and policy-making communities. Since the information age and the knowledge economy, the availability of information in digital format has tremendously grown and is continuously increasing (Janssens 2007). During the past decade the World Wide Web has become a most important general source of information. More and more scientific and technological information, are made available and actually retrieved through the web. The comprehensive coverage of research articles published in many thousands of online peerreviewed international scientific and technical journals like Thomson/ISI Science Citation Index 12.

(14) or the Web of Science, and the patent databases such as the United States Patent and Trademark Office (USPTO), the European Patent Office (EPO) and the Japan Patent Office (JPO) are valuable sources of information used in quantitative studies of S&T systems . Analyzing these data sources properly could help policymakers, administrators, R&D managers, scientists and scholars getting familiar with progress in different science and technology fields and provide them with an insight for future planning.. 1.3. Importance of Topic Scientific progress and technological innovation have become major drivers of economic progress in the knowledge oriented capitalist economies where growth, productivity, and competitiveness are increasingly based on improved technologies, novel products, upgraded processes or customized services (Moed et al 2004). The creation of radically new knowledge, improving existent knowledge, or imitation of others, has become central to economic development. Today, as markets and technologies are rapidly changing and product life cycles become ever shorter, firms continue to develop and introduce new products to market places worldwide that form part of their core competency (Lee et al. 2008). Firms have used a variety of methodologies to derive new ideas and direct them to the best development paths for creating the next generation of products and related technologies. These methodologies range from general methods including brainstorming, user analysis, benchmarking, technology forecasting, Delphi, etc. to specific techniques such as quality function deployment (QFD), which is used to link customer needs and product attributes, TRIZ to generate inventive ideas for problem solving, analytic hierarchy process (AHP) to analyze multiple new product development (NPD) scenarios and technology road mapping (TRM) to connect product planning and technology planning (Lee et al. 2008). Providing meaningful information from science and technology data bases can help experts get acquainted with recent progress in science and technology and will increase the efficiency of pre-mentioned methods. Measuring science and technology in a quantitative manner, is crucial to understand and interpret the relational structure of both science and technology spheres. Several quantitative, particularly bibliometric, indicators have been established and applied to determine the linkage between science and technology (Narin and Olivastro 1992; Narin and Olivastro 1998; Meyer 13.

(15) 2001; Deleus and Hulle 2002; Looy et al. 2003; Chen and Hicks 2004; Guan and He 2007; Nomaler and Verspagen 2007; Van Looy et al. 2007). In most of the studies conducted in this area scientific articles have been considered as indicators of science and patents considered as indicator of technology. Evaluating science-technology relations and tracing the transfer of knowledge from science to technology or technology to science, are crucial issues for understanding the knowledge society and addressing science and technology policy questions. It provides tools for science and technology monitoring and determines the orientation of R&D policies. Creating the framework for monitoring and tracing science and technology developments and knowledge transfer between them has lots of advantages. In macro view it helps policy makers and administrators develop their strategic plan, determine the R&D priorities and specify the knowledge gaps in different fields. In micro level it aids scientists, scholars and technicians getting acquainted with science and technology trends. It enables firms to respond to ever more sophisticated consumer demands and stay ahead of their competitors, both domestically and internationally. It also helps them in monitoring and tracing their competitor’s behavior (Verbeek et al. 2003). Nano discipline, the science of minuscule molecule with its continuous developments, is a field of research with high speed of progress that is becoming a very important aspect for modern life. Its impact is most felt in sectors like biotechnology, electronics, energy and industrial products. Advances in information technology enable efficient investigation of publications, their contents, and relationships for large sets of nano-technology related documents in order to assess the status of the field (Li et al. 2008). Currently there are twenty two SMEs active in nano area in Iran and Iranian government supports this industry. Improving in this area required getting access to efficient tools that help policy makers and scientists, monitoring and tracing the evolutions of science and technology in this field. This research aims to provide such frameworks and help government, policy makers, administrators, R&D managers, scientists and scholars in this matter. Matching author’s names with patent’s inventors and clustering nanotech patents and articles with text mining techniques for investigating the similarity between them has been suggested as an indicator of evaluating science and technology interaction. Monitoring the evolution of science and technology in. 14.

(16) different nano discipline by analyzing patents and papers document is another purpose of this study.. 1.4. Problem Description In an era of rapidly expanding digital content, the number of e-documents and the amount of knowledge available through internet, frequently overwhelm the R&D teams. Organizations are utilizing an ever expanding number of electronic knowledge documents. Engineers and researchers, in particular, are retrieving and processing large numbers of technical reports and patent documents to safeguard existing intellectual properties and to avoid infringing upon the rights of others (Trappey and Trappey 2008). For individuals and organizations alike, this overwhelming amount of digital data leads to major difficulties to find and process relevant information and knowledge (Janssens 2007). Although the internet has greatly facilitated the search and retrieval of knowledge and search engines are essential to find relevant information, but they often return a mass of irrelevant results within very long result lists. As Trappey mentioned in his research, Information retrieval should be complemented with other algorithms to move beyond the mere finding of interesting documents. R&D teams rarely have the time or budget to read and understand all of the documents retrieved and, as a result, are increasingly overloaded with too much information (Trappey and Trappey 2008). Existing classifications of science are inherently outdated because of the pace at which scientific knowledge advances (Janssens 2007). Strategic planning for science and technology development and R&D progress require having deep insight of science and technology evolutions. The high speed of science and technology progress and overwhelming data available on internet makes it difficult for human beings to extract useful knowledge without powerful tools. Text mining techniques can help them automatically extracting useful knowledge out of textual resources (Tseng, Lin et al. 2007). Governments face with uncertainty when they want to divide their budget in different fields. They must decide on the way of spending their R&D resources. This specially is crucial for developing countries. They have to find out their knowledge capabilities and determine making progress in which field could become source of competitive advantage for them in future.. 15.

(17) One of the main reasons that causes new products or new technologies don’t face with good market is not having enough knowledge about trend and speed of progress in different knowledge fields. Organizations invest huge amounts of money for developing new products which after reaching to mass production level couldn't find good market and the investment become useless. This usually happens because R&D managers or new product development experts in organizations don’t pay enough attention to the behavior of their competitors and substitute products or technologies. Monitoring science and technology evolutions is the prerequisite of effective future R&D planning. It helps policy makers get acquainted with progress in different knowledge fields.. 1.5. Research Objectives and Questions Identification The main purpose of this study is developing framework for monitoring science and technology evolutions and evaluating the linkages between science and technology based on information in scientific articles and patents. Nano-science and nanotechnology were chosen for doing the research. We want to give an insight to strategic managers, policy makers and scholars in organizations about how could they use codified S&T knowledge sources more efficiently. Information in patents and scientific articles are good source of codified knowledge. Monitoring their evolution can help managers, policy makers and experts, in different levels, in better decision making and planning for future more efficiently. By using text mining techniques we try to develop some tools for better monitoring and tracing science and technology evolutions. Study the information diffusion between patents and scientific articles are one way that can help organizations find areas that need more interaction with academic places, they could find in which area patenting is more common and find areas that putting effort on them would be a good business opportunity for future. Developing framework for comparing science and technology behavior is an efficient tool for strategic planning in different science and technology disciplines. In order to reach to these purposes, different methods of evaluating science and technology interaction were studied. Analyzing Co-Occurrence of inventor-author names in patents and scientific articles, and lexical analysis by using text mining techniques are two approaches which we applied in this study for measuring the information diffusion between patents and scientific articles in nano discipline. By finding common inventor-author names, we can identify the 16.

(18) linkage channels among academic institutes and industrial organizations active in nano discipline. By using this approach we can find in which nano sub disciplines there is high probability that scientific activity leads to register patents or reach to commercial applications. In lexical approach by using text mining techniques, we try to develop the framework for monitoring nano-science and nano-technology evolutions. We try to determine nano sub disciplines and study the trend of each group in recent time interval. We will also draw the scurve of each cluster for analyzing their future evolution. The main objectives of this research include: 1. Investigating different methods of evaluating science and technology interactions based on information in scientific articles and patents. 2. Developing a framework for evaluating science and technology evolutions by using scientific articles and patents databases. 3. Helping policy makers, administrators and R&D managers in nano discipline, in their future strategic planning by monitoring information diffusion between scientific articles and patents in nano discipline. 4. Finding areas that have strong linkage between science and technology by matching common information between patents and papers. Regarding the pre-mentioned problems, the research questions are as follows: 1. What are the different methods used in evaluating science and technology interactions based on information in scientific articles and patents? 2. How could policy makers, administrators and NPD managers use scientific articles and patent data sources for informing about areas that require strong linkage between science and technology? 3. How could S&T policy makers, scientists and scholars monitor science and technology evolutions by using information in scientific articles and patents?. 1.6. Outline of the Entire Thesis The entire research process will be presented in 5 chapters. In figure1.1 the outline of entire thesis has been shown.. 17.

(19) Chapter 1 included brief introduction of entire research. Background of study, problem description, importance of topic, research questions and research objectives are subjects described in this chapter. In chapter 2, the major groundwork and preliminaries related to the subject of study, is going to be reviewed and Different approaches of measuring S&T interactions will be discussed. Chapter 3 is related to describing the research methodology used for reaching to the research objectives and answering the research questions. After clarifying research approach and design strategy, all steps involved in research process will be explained. Analyzing CoOccurrence of inventor-author names and lexical analysis by using text mining techniques are two approaches which we applied in data analysis for measuring the information diffusion between patents and scientific articles in nano discipline. In chapter 4, the data analysis procedure will be described and the results of implementing two approaches for data analysis will be explained. Finally in chapter 5, we will mention to the main findings of this research. The managerial implications, research limitations and problems will be discussed in this chapter and some recommendations for future researches will be suggested.. Chapter 1: Introduction. Chapter 2: Literature Review. Chapter 3: Research Methodology. Chapter 4: Results and Analysis. Chapter 5: Conclusions. Figure 1-1: Outline of the Entire Thesis. 18.

(20) Chapter 2 Literature Review. 2. Literature Review The main purpose of this research is studying the interaction of science and technology in nano science and nanotechnology field by measuring the information diffusion between patents and scientific articles. We want to develop framework for monitoring science and technology evolutions. In this chapter, the major groundwork and preliminaries related to the subject of study, is going to be reviewed. Scientific articles will be considered as indicators of science and patents considered as indicator of technology. Different approaches of measuring S&T interactions will be discussed with their strengths and weaknesses.. 2.1. Science &Technology as Source of Competitive Advantage The importance of science and technology (S&T) in discovering new knowledge, driving market innovations, and delivering public goods has accelerated in recent decades (Narin and Olivastro 1992; Geisler 2000; Meyer 2002; Chen and Hicks 2004; Kostoff and Geisler 2007; Nomaler and Verspagen 2007) Knowledge is nowadays generally accepted as a fourth production factor (besides labour, land and capital) accounting for a major portion of the variation in economic growth and development between nations and continents (Verbeek et al. 2003). The rate of technological change, the industrial structure of countries and the rate of scientific progress are all means of measuring countries economic growth (Griliches 1990). At macro-economic level, a nation’s innovation capacity is a major driver for (sustained) economic growth. Competitiveness is at least partly the result of an actor’s capacity to create, 19.

(21) trace, absorb and to assimilate new technological and scientific evolutions- the so-called absorptive capacity (Verbeek et al. 2003). Scientific research, and more specifically basic scientific research, is a major dimension of this capacity to absorb and to internalize new knowledge. Spillovers, originating from the scientific literature, are an important dimension of knowledge accumulation and growth (Allen, 1977 and Grupp, 1998 cited by (Verbeek et al. 2003)). In science-based technologies and sectors (i.e. technologies closely related to scientific progress), especially in newly emerging fields, the capacity to create and to absorb state-of-theart research is of critical importance for developing and maintaining a strong technological position (Verbeek et al. 2003). Technology-based economic growth is increasingly dependent on product differentiation and strategic strengths in novel technological fields. The genesis of some of these emerging fields can be traced to the close interaction between universities/public research institutions and industries (Bhattacharya and Meyer 2003). One of the reasons that cause Europe has generally been depicted as lagging behind the US in terms of technology development is because of its inability to successfully translating science and technology into market applications and subsequent revenue generation. (See European Innovation Scoreboards (2007) and (2002)). At micro-economic level, Ability to extract knowledge from science and technology toward innovation enables firms to respond to ever more sophisticated consumer demands and stay ahead of their competitors, both domestically and internationally (Verbeek et al. 2003). Outputs from research and development (R&D) and science and technology (S&T) were linked to productivity of organizations, sales of new products, profitability, and similar economic variables (Griliches 1998; Geisler 2005). In order to have science and technology as source of competitive advantage, it is necessary to have a strong and well-coordinated interaction between government, academia and industry.. 2.2. Actors in Science and Technology System In domain of this research, government, academia and industry are considered as central actors in science and technology system. A strong and well-coordinated interaction between government, academia and industry is a prerequisite of successful progress in science and technology area (Verbeek et al. 2003). 20.

(22) University-industry-govvernment reelations are well w describbed in the enntire stream of literaturee o Universityy-Industry- Governmennt which became b wid dely used in n on the Triple Heliix model of 1 (Leyddesdorff andd Etzkowitzz science and innovaation policyy during thee second haalf of the 1990s E 20002; Leydesdorff and Frritsch 2006)). 1998; Etzkowitz Three selecction enviroonments aree specified in the Tripple Helix m model (Leyd desdorff and d Meyer 2006): 2 •. dustry) Wealth genneration (ind. •. Novelty production (accademia). •. And publicc control (goovernment).. Acccording to Triple T Helixx model, maaking of sciience and teechnology iis seen as a process off interactiion among tthese three actors whicch have signnificant role in the geneeration, transsfer and usee of scienntific knowleedge in an innteractive syystem. Bilateral reelations beetween gov vernment annd universiity, academ mia and inndustry and d governm ment and inddustry as shhown in figuure 2-1 havee expanded into i triadic rrelationshipss among thee spheres. Where thee institutionnal spheres overlap annd collaboraate and coopperate with h each otherr (Etzkow witz 2002). We can observve an increasing emphaasis on the rrole of partnnerships, onn the dynam mic interplay y and the increasing simultaneityy of knowleedge demannd and supplly, as well aas on the grrowth in thee o the actors involved inn knowledgee generation n processes. trans-disciplinarily and the heteerogeneity of. Figuree 2-1: Different approachees of Triple Helix H model. 211.

(23) 2.3. Publications and Patents as Means to Analyze Science and Technology Interaction Before introducing various approaches to study exchange processes at the interface between science and technology, it seems necessary to clarify the definitions. In most of the S&T interaction studies, papers are viewed as representations of science, while patents are considered as representation of technology (Narin and Olivastro 1998; Meyer 2000; Deleus and Hulle 2002; Verbeek et al. 2002; Chen and Hicks 2004; Meyer 2006; Nomaler and Verspagen 2007; Boyack and Klavans 2008; Larsen 2008). Scientific publications and patent databases are leading sources for R&D policymakers since data are collected in one place and repeatedly updated, the longitudinal analysis could be done easily and the data has standard structure and the definitions and categories change slowly (Moed et al 2004). In table 2-1, some researches in this area, with quantitative approach to science and technology development have been represented.. Table 2-1: Researches with quantitative approach to science and technology evolutions. Authors. Year. Zvi Griliches. 1990. Narin and Olivastro. 1998. Martin Meyer. 2001. Rinia, Leeuwen, Bruins, Van Vuren and Van Raan. 2002. Deleus and Van Hulle. 2002. Martin Meyer. 2002. Van Looy,. 2003. Title Patent Statistics as Economic Indicators. Linkage between Patents and Papers: An Interim EPO/US Comparison Patent citation analysis in a novel field of technology: An exploration of nano-science and nano-technology Measuring knowledge transfer between fields of science Science and technology interactions discovered with a new topographic map-based visualization tool Tracing knowledge flows in innovation systems. Do science technology interactions. Unit of Analysis Patents Non patent references in patent documents Patents and Scientific articless Analyzing Crossdisciplinary citations in journal articles Patents and Scientific articles Patents and Scientific articles Non patent references 22.

(24) Authors. Year. Zimmermann, Veugelers, Verbeek, Mello and Debackere Holger Ernst. 2003. Chen and Hicks Adams, Clemmons, and Stephan Fabry, Ernst, Langholz and Koster. 2004 2006 2006. Nomaler & Verspagen. 2007. Li, Chen, Huang and Roco. 2007. Katrina Larsen. 2008. Title. Unit of Analysis. pay off when developing technology? An exploratory investigation of 10 scienceintensive technology domains. Patent information for strategic technology management. Tracing Knowledge diffusion. in patent documents. How rapidly does science leak out? Patent portfolio analysis as a useful tool for identifying R&D and business opportunities. Knowledge flows, Patent citations and the impact of science on technology. Patent citation network in nanotechnology (1976-2004) Knowledge network hubs and measures of research impact, science structure, and publication output in nano structured solar cell research.. Patent Patent references Patent citations Patents Patent citations Patent citations. Scientific articles. Many scientists and economists believe that public science is a driving force behind high technology and economic growth. They also believe that the transfer of publicly supported knowledge to industry is an important part of the technology transfer process (Perko and Narin 1998). Also the information in patent data can be used for strategic planning purposes. Patent documents contain important research results that are valuable to the industry, business, law, and policy-making communities. If carefully analyzed, they could be a strategic information source that show technological details and relations, reveal business trends and inspire novel industrial solutions (Tseng, Lin et al. 2007). Traditionally, companies patent more than they publish, and university researchers publish usually more than they patent. As Meyer mention to the research done by Pavitt, business firms are granted about 80% of all patents, and many of the remaining 20% are granted to individuals who are owners of small firms (Meyer 2002). Academics publish more than do 23.

(25) their colleagues in industry. In such a situation, with most of the scientific papers published by academics and most patents held by and originated in industrial companies, the tendency to associate the academic sector with science and the industrial sector with technology is only natural (Meyer 2002). In the following section scientific publications and patents as proxies of science and technology will be discussed in more detail. We will also mention their strength and shortcomings, as well as their best use for strategic planning and identifying future business opportunities.. 2.3.1. Scientific publications as indicator of science The specific role of science for technological development, have received ample attention in both the research and the policy communities (Geisler 2000; Meyer 2001; Verbeek et al. 2003; Geisler 2005; Van Looy et al. 2007) Scientific publications from large-scale bibliographic databases are recognized as a central source of information. They are the most authoritative records of research. Databases such as the Science Citation Index (SCI) contain detailed records of these papers and citations to them from other papers, allowing detailed analysis (Perko and Narin 1998). One of the objectives of a scientific publication is to spread scientific findings within and outside the scientific community. Hence, publication in scientific journals the -“serial literature”plays a leading role in the dissemination of the findings of science (Verbeek et al. 2002). They can be used to trace knowledge flows and can describe the linkage structure and intensity of those flows (Meyer 2002). 2.3.1.1.. The Metrics of Bibliometrics. Bibliometrics is the field of science that deals with development and application of quantitative measures and indicators for sciences and technology, based on bibliographic information. This bibliographic information is the representation of codified knowledge as can be found in a diversity of scientific output types, such as serial literature, books, and book chapters, conference proceedings, patents, etc (Moed et al 2004). It measures the outputs that are proximal to the scientific activity, that is, the client outcomes that science generates (Geisler 2005).. 24.

(26) Each year about 1,000,000 publications are added to the scientific archive of this planet. This number and also numbers for sub-sets of science (fields, institutes) are in many cases sufficiently high to allow quantitative analyses yielding statistically significant findings. Publications offer usable elements for ‘measuring’ important aspects of science: author names, institutional addresses, journal (which indicates not only the field of research but also status!), references (citations), concepts (keywords, keyword combinations). Although not perfect, we adopt a publication as a ‘building block’ of science and as a source of data. This approach clearly defines the basic assumptions of bibliometrics. When the results and findings from the R&D activity are published, overtime a database of knowledge is thus created, and this becomes the state-of-the-art (SOA) of the discipline. Within this database, scientists also compute the citation analysis, which is a process by which citations of articles in the literature are counted and analyzed to show and to study emerging patterns (Moed et al 2004). Industrial companies use bibliometric measures in two complementary forms. The first is the evolution of the performance of their scientific personnel within the R&D unit and activity. The second is the evaluation of the R&D functions within the firm. In general, companies use bibliometrics as an integral part of their system that evaluates scientific and technical employees. Widely-published corporate researchers indicate an outstanding R&D group and generate muchdesired reputation and prestige for the company. Companies hold the belief that the reputation of their scientific cadre will permeate onto their customers, other stakeholders, and the general public. In summary, industrial companies utilize bibliometrics because they believe that by supporting a climate of research and publishing, the firm fosters the intellectual enrichment of its personnel. This, in turn, will ultimately translate into contributions to the bottom-line through innovations, added prestige, and sustained competitiveness (Geisler 2002). 2.3.1.2.. Strengths and Weakness of the use of Bibliometrics. Bibliometrics has a variety of strengths. Geisler (2000) divided these strengths in to three categories: structure, measurement, and representation (Geisler 2000). In table1, the strengths of each category have been shown. The structure category includes those strengths that allow for multiple levels of analysis, and a relatively adequate cost for analyses. In the measurement category, the strengths suggest that the data are available and the analysis is straightforward, making it a relatively simple procedure to plan and to undertake. Finally, bibliometrics is. 25.

(27) considered by the S&T community to be a valid representation of the phenomenon of outcomes from inventive activity. Although bibliometrics is a relatively simple; inexpensive and valid metric of the intellectual outcomes from science, but it is not free of problems. The use of serial literature is not evenly distributed over fields of science (for instance, the dominant use of conference literature within some fields of the technical sciences), bibliometric studies start from the assumption that the most important findings of scientific research finally end up in the international serial literature. This, however, means that, in general, bibliometrics is less applicable in those fields of science in which the internationally oriented scientific journal is not the main medium for communicating research findings to the (international) community in those fields. journal articles are not in all fields the main carrier of scientific knowledge; they are not ‘equivalent’ elements in the scientific process, they differ widely in importance; and they are challenged as the ‘gold standard’ by new types of publication behavior, particularly electronic publishing. However, the daily practice of scientific research shows that inspired scientists in most cases, and particularly in the natural sciences and medical research fields, go for publication in the better and, if possible, the best journals. A similar situation is developing in the social and behavioral sciences, engineering and, to a lesser extent, in the humanities (Moed et al 2004). Four groups of weaknesses and problems are shown in table 2-2. Bibliometrics are best employed in the evaluation of the scientific component of knowledge creation. But, when applied downstream the innovation process, this metric describes only one, perhaps very small, attribute of the phenomenon of R&D and technological development, utilization, and commercialization by the firm. Hence, bibliometrics must be regarded as a limited measure of the upstream activity and outputs, whereas other metrics should be used to assess downstream activities of the innovation process (Geisler 2002).. 26.

(28) Table 2-2: Bibliometric as an S&T Indicator. Source: Geisler, 2000. Strengths and Benefits A. Structure •. • •. Bibliometrics can be applied to various levels of generators of intellectual outputs, such as individuals, groups, instructions, and countries. The cost of collecting the data and conducting meaningful analysis is relatively adequate. The measures are already built into the metric, thus there is no need to establish them and to test them for validity.. B. Measurement •. •. Bibliometrics allows for quantitative assessment of S&T outputs by counts of papers and citations, and for qualitative assessment by analysis of core journals and their relative impacts. Bibliometrics and its analysis is a relatively straightforward approach, relying on a few assumptions.. Weaknesses and Problems A. Coverage •. •. B. Measurement •. •. •. C. Representation • •. •. Bibliometrics can be applied to the entire spectrum of S&T where outcomes take the form of reports, papers, and citations. Bibliometrics, through citation analysis, help to determine the role that individuals and institutions have in the evolution of a scientific discipline. Bibliometrics analysis allows for the identification of trends and developments in science and technology and in scientific. Published articles are only one measure of outputs from scientific activity; hence the metric does not cover reports, other written communications such as electronic mail, letters, and personal communiques. Articles published in peer-reviewed journals and their citations analyses disregard the outputs and intellectual contributions in articles published in technical outlets, as well as work-inprogress.. The “Pied-Piper Effect”: bibliometrics, particularly citation analysis, measures influence, not quality. Citations are selective and refer to those papers that “toe the line” and do not “rock the boat.” Published articles measure output in a given sub discipline or discipline, hence cross-disciplinary analysis may be difficult to validate, because of the different structure and procedures of the scientific investigation in each discipline; particularly ease and rate of publishing, and the nature of the peer selection processes (the “Apples and Oranges Effect”). Counts of publications and citations analysis tend to disregard the influence of the stage in the life of the discipline or area, such as “mature” area versus an “evolving” new area (the “Gulliver Effect”).. C. Generalizability •. •. Counts of publications and citations lack a standard for their validation as a measure of quality. When compared with inputs (investments in R&D), the resulting analysis relies on co variation of two distinct phenomena and disregards the complexities of the R&D process. The only standard for validation of bibliometrics is convention of a small and elite group of influential scientists.. 27.

(29) Strengths and Benefits •. disciplines. By convention, bibliometrics has been accepted by the S&T community as valid representation of the outputs from intellectual and inventive activities.. Weaknesses and Problems •. •. As in the “Hindsight” project, the problem is: “How far in time should citations go?” Should articles in quantum mechanics cite Einstein’s work and that of the Greek philosophers and mathematicians? Citations are thus highly selective and refer to a relatively short timeframe, preferably within a few years of the focal paper in which they are cited.. D. Biases •. •. •. •. Because of the almost incestuous nature of the small group of prolific publishers, there is an inordinate amount of self-citations and citations of “friends” and other members of this elite group. Thus, authors who publish in areas that are near the boundaries of their discipline or in cross disciplinary topics are much less likely to be published in top journals or to be cited in them. Criteria for the selection of articles for publication in juried journals are built into the process and will bias the resultant counts of papers and citations in favor of those authors preferred by the reviewers. Selection and analyses of key journals and the interpretation of the counts of papers and citations are based on assumptions of validity of these metrics as measures of quality and internal dynamics of the discipline. Such assumptions are the product of opinions and judgment, hence biased. The selection and analysis process of key journals and papers to be included in this process is biased in that analysts who are generally outside the discipline or the area impose their own views and criteria in making determinations and drawing conclusions that transcend the data. This list is not exhaustive nor in any specific order.. 2.3.2. Patents as Indicator of Technology As a measure of S&T, patents are considered by many economists to be indicators not only of inventive activity but also of technological progress and change at the industrial and 28.

(30) national levels. Patents are considered to be “tangible evidence of technological innovation, “therefore they are considered a reliable measure of technological capability and achievement (Geisler 2002). In this section we briefly explain patents, their structure and main sources. We also discuss about patent analysis and its utilization and illustrate its strength and weak points.. 2.3.2.1. Patent Definition A patent is a document, issued by an authorized governmental agency, granting the right to exclude anyone else from the production or use of a specific new device, apparatus, or process for a stated number of years (17 in the U.S. currently). The grant is issued to the inventor of this device or process after an examination that focuses on both the novelty of the claimed item and its potential utility. The right embedded in the patent can be assigned by the inventor to somebody else, usually to his employer, a corporation, and/or sold to or licensed for use by somebody else. This right can be enforced only by the potential threat of or an actual suit in the courts for infringement damages. The stated purpose of the patent system is to encourage invention and technical progress both by providing a temporary monopoly for the inventor and by forcing the early disclosure of the information necessary for the production of this item or the operation of the new process (Griliches 1990). To be issued, a patent must satisfy three general criteria (Perko and Narin 1998). •. It must be useful;. •. It must be novel;. •. And it must not be obvious.. 2.3.2.2. The Structure of a Patent Document Publicly available patent documents contain a wealth of information such as (Jaffe and Trajtenberg 2002): •. Name and addresses of inventor(s);. •. Invention characteristics;. •. Name and addresses of the patent assignee(s);. •. Date of filing;. •. Technical class of the patent;. •. References to other patents and scientific literature.. 29.

(31) This information can be used in tracking advances in technology. They have been used in wide variety of studies to explore the nature, sources, and economic effects of technology (Griliches 1990; Meyer 2001; Jaffe and Trajtenberg 2002; Hu and Jaffe 2003; Moed et al 2004; Li et al. 2007). Ganguli and Blackman (1995) review the structure of a patent document, and divide it in three parts (Meyer 2000): 1. The title page with bibliographic information; 2. The text, which includes a description of the invention, preferred examples in details as well as drawings, diagrams, and flow charts; 3. The claims.. 2.3.2.3. Main Sources of Patents & Their Differences Statistics shows that the United States Patent and Trademark Office (USPTO), the European Patent Office (EPO) and the Japan Patent Office (JPO) issue nearly 90 percent of the world’s patents (Kowalski et al. 2003). There are, of course, many fundamental differences between main sources of patents. For example some of the main differences between the U.S. patent system and the EPO system, include (Narin and Olivastro 1998; Nomaler and Verspagen 2007): •. The European system is a "first to file" system, where priority rights are based on the first person to file a patent, whereas the U.S. is a "first to invent" system, where priority rights to an invention are given to the person who can prove that he first invented it.. •. In the European system patent applications are published 18 months after application, usually with search reports, whereas in the U.S. system patents are confidential until granted, and are only published at that time.. •. In the USPTO system the applicant, when filing a patent application, is requested to supply a complete list of references to patents and non-patent documents that describe the state-of-the-art of knowledge in the field. In the EPO system, the applicant may optionally supply such a list. In other words, while in the US there is a legal requirement and non-compliance by the patent applicant can lead to subsequent revocation of the patent, in Europe it is not obligatory. As a result, applicants to the 30.

(32) USPTO “rather than running the risk of filing an incomplete list of references, tend to quote each and every reference even if it is only remotely related to what is to be patented. •. Finally, it should be mentioned that in the U.S. system there are many other science references contained in the body of a patent which are very similar in general nature to the references on the front page.. 2.3.2.4. Patent Analysis The detailed information contained in the patent documents is valuable source for analysis. One can study the geographic distribution of particular inventions based on the country information, one can investigate technology trend in different patent classes, and one can actually read the detailed text of a series of patents in a particular field as raw material for an economic technological history of it (Griliches 1990). Patent analysis can be done based on the following information in patent document: •. Statistical information such as the names of inventors and their addresses and the name of the organization to which the patent right may have been assigned. •. patent classes to which it has been assigned by the examiners. •. Information about cited and citing references. •. Detailed description of patents through text mining techniques. Patent analysis consists of following seven steps which can be compressed or expanded as appropriate (Moed et al 2004): 1. Issue specification and data source(s) selection which the research topic and its objective will be clarified. Also the suitable data source for analysis will be selected in this step. 2. Patent search and retrieval; Patent search could be done by choosing appropriate keywords, patent number, the name of inventor or assignee based on the research topic. 3. Data cleaning and preprocessing: In this step unsuitable data will be omitted from collected data base and desired information will be parsed out from data set. This information could be the name of countries, assignees, inventors, citing and cited. 31.

(33) references or patent description. Some preprocessing action also will be done in this step. 4. Data analysis: Extracting relationships and pattern between processed data could be done in this step. According to research objectives analysis could be done in various ways. For example one could do trend analysis in different technology field by monitoring the number of patents issued in subcategories of that technology field. Or we could trace the organization collaborations based on the common assignee name. 5. Information Representation: Extracted knowledge will be visualized in this step by using different visualization and mapping tools. 6. Interpretation: Results will be evaluated according to users need in this step. 7. Utilization which the result will be presented for decision maker and after approving will be spread for execution.. 2.3.2.5. The Utilizations of patent analysis Information obtained from patent analysis can help senior management who uses this information for decision-making and external stakeholders of the firm, such as shareholders and analysts, who have an increasing interest in assessing a firm's technological competence because of its strong impact on the firm's future competitiveness. In particular, one can distinguish three major applications of patent citation analysis: First, following the general science orientation of fields over time by revealing a web of science and technology linkages. This application allows for observing potential governance shifts in certain S&T areas. Second, measuring the intensity of science and technology interaction, Third, tracking potential knowledge flows between scientific and technological fields (Meyer 2002). The role of patent analysis in strategic planning can be summarized in following items (Ernst 1998; Ernst 2003; Ernst, Fabry et al. 2004; Fabry, et al. 2006): •. R&D program management. •. Intellectual asset management. •. Assessing firm's technology portfolio in comparison with the competitors. •. Assessing the attractiveness of technologies, especially new technologies posing a threat or a new opportunity for the existing business. •. Recognizing strategic changes in the firm's competitive environment 32.

(34) •. Evaluate important market partners, especially customers and suppliers, to determine if the firm's R& D strategy is in alignment with the R&D strategy of its major customers and if R&D alliances exist with the most competent suppliers. •. Assessing the patent situation in new business areas which may be explored. •. Improve human resource management regarding leading inventors in specific technological fields.. 2.3.2.6. Strengths and Weakness of Patent Analysis A patent usually is the consequence of a successful R&D activity, thereby offering detailed information on the activity itself. Patents offer an interesting monitoring device to identify main trends in technology development, and, under specific conditions, the possibility to analyse specific R&D processes in more detail (Verbeek et al. 2002). Patents are the most widely available indicator of output of technological activities (Geisler 2000). A country’s technological performance is measured in terms of its technological productivity which can be measure through the number of patents per capita (Looy et al. 2003). It should be considered that although patents have extensive uses and contain valuable information, patent counts and analyses provide only a limited measure of the R&D activity and are not sufficient to be used as a sole metric of R&D. But when used in combination with other metrics, patents offer a manageable piece of data on the level of R&D effort by individuals and firms. One should not confuse the relative ease in obtaining such data and their manipulation with the amount of knowledge that such data provide on the R&D phenomenon. Such knowledge is limited, bounded by the weaknesses and problems displayed in table2. In particular, patents as a metric are of limited value because of different patenting behaviors by companies, and the fact that they are designed to be a legal instrument offering some measure of protection of intellectual property. They were not specifically designed as a metric of the R&D activity nor its outcomes (Geisler 2002). The utilization of data on patents to measure technical and scientific output raises a number of problems (Meyer 2002). •. The requisites for an invention to be patented and the type of examination it is subjected vary from country to country;. 33.

(35) •. The propensity to patent varies according to the industrial sector, size of firm and type of inventor (individual or employed in an organization);. •. It is not known what proportion of inventions is patented and thus one cannot say to what extent patenting reflects the entire area of inventive production;. •. the “quality” and “value” of patents varies greatly;. •. Insufficient data are available on the extent to which patents issued are in fact utilized;. •. A significant proportion of patents are of the strategic type, i.e. applied for in order to forestall potential competitors.. The strengths and weaknesses of patents as indicator of technology evaluation are summarized in table 2-3. Table 2-3: Patents as S&T Indicator. Source: Geisler, 2002. Strengths and Benefits B. Methodology • • • • •. Patent data are quantitative measures. Patent databases have been in existence for many years. Patent data are relatively easy to manipulate. Patent data can be related to other economic/financial measures. Patent citations allow for co-citation analyses.. D. Structure • • • •. Patent data have a similar structure as a legal document. Contain revealing information. Indicate levels of S&T effort. A similar item of information facilitates cross-industries and even cross-national comparison.. E. Impacts and Contributions • •. Considered as a link between S&T and firm performance, patents offer an elegant way of establishing such a link. Patents are viewed as indication of. Weaknesses and Problems C. Firm-Industry Strategy • • • • •. There are marked differences among firms and industries in the propensity to patent inventions. Industrial firms only spend less than 1/3 of their R&D effort to develop new products that would result in patents. Firm-specific information in patent data impacts decisions to invest in S&T. Past performance of patenting activity impacts the propensity to patent.. D. Market Considerations • • •. Distortions in market due to monopolistic behavior. Market factors may impact patenting behavior. Patents do not always lead to commercial applications.. E. Structure and Methodology • •. Patents represent only a small portion of the actual R&D and S&T effort. Patents reveal only selected information about S&T.. 34.

(36) Strengths and Benefits • •. technological achievements. Patents are viewed as measures of the knowledge-base. Patents are viewed as measures of the. Weaknesses and Problems • •. quality of S&T. •. Patentable inventions have become increasingly harder to discover. The link S&T-Patents-Performance is based on co variation methodology and lacks a description of the process and factors that impact this presumed link. There is a lack of a theory to explain how patents contribute to performance and to strategic advantages (except for the link to possibly monopolistic manifestations of the power from patent protection).. 2.4. Science and Technology Interaction Over the last three decades, different ways of approaching the science and technology interaction have been emerged. At first, the knowledge transfer from science to technology was considered to be linear and Science and technology were viewed autonomously. This approach is highly criticized in literature (Meyer 2000; Verbeek et al. 2002; Looy et al. 2003; Meyer 2006). Verbeek and his colleagues (2002) in their study of linking science to technology mention that the use of the linear approach largely ignored due to the following reasons: 1. The empirical evidence that technological change often resulted from experience and ingenuity rather than from scientific theory or method 2. The instrumental role of technological developments in inducing scientific explanation 3. The importance of technology-based instrumentation for scientific investigation De Solla Price developed a two-stream model based on citation analysis of science and technology journals. This model stresses the autonomy of science and technology as cognitive systems and the reciprocal nature of their interplay. Tracing citations in science and technology journals, De Solla Price found separate cumulative structures with scientific knowledge building on old science and technology on old technology. He also detected a weak and reciprocal interaction between the two. De Solla Price saw in his 1965 study the closest interaction between science and technology taking place in the period of education when ‘budding scientists read the archival literature in their fields’ (Price 1965).. 35.

No results found