The Technological Impact of Library Science Research: A Patent Analysis Gali Halevi* and Henk F. Moed** *
[email protected] Elsevier, Inc. 360 Park Avenue South, 8th Floor, New York, NY 10010 USA. **
[email protected] Elsevier, Radarweg 29, 1043NX Amsterdam, the Netherlands
Abstract This study examines the characteristics of research articles published in Library Science journals and the manner by which they are cited in patents. A search of 42 top performing LIS journals within the non patent literature citations between 1991 and 2011 revealed 8 distinct journals cited in patents. The articles cited were mainly ones featuring library information and customer management systems as well as indexing and classification methodologies; the patents citing these articles featured mainly online commerce applications. A long time lap of about 20 years between the publication year of the articles and the application year of the patents which they were cited was identified. This finding bares the conclusion that library systems and records management methodologies were answering an advanced set of remote customers needs using computerized solutions long before the concepts of online commerce and internet access were conceived. Introduction There are many factors working in today’s scientific landscape, most prevalent being budgetary constraints, that make the ability to measure Return on Investment (ROI) crucial for funding decisions. Academic research is being scrutinized in search of a metric or evaluative model that will enable decision makers understand the potential of its results to impact the economy and society as a whole. One of the frequently used and most naturally occurring ways to measure science’s impact has been in its patentability which is also evident in numerous studies that explored the phenomenon of basic research patenting and its effects on both academic and industrial progress (Kirschenbaum, 2002; Sampat & Saathoff, 2010). The passage of the Bayh-Dole Act in the USA in 1980 contributed to the increase of university patents applications. This act gave universities the right to own and license the results of their government-funded research and in return share a portion of the revenue derived from such patents with the inventor. It has been noted that this increase is more evident in certain disciplines and areas such as biotechnology, pharmacy, engineering etc. (Thursby & al. 2009; Rosell & Agrawal, 2009). Patents are traditionally viewed as indicators of innovation and of economic impact, but more importantly, being one of the most rigorously peer reviewed forms of publications, their embedded citations which are also referred to as ‘prior art’ can provide an accurate view of related topics and research that pertain to the patent. These citations are divided into two main types; patent citations and non-patent-literature citations. Patent citations usually used to denote the similarities and differences between current and prior granted patents. Non-patent-literature citations, however, can include several forms of sources such as peer reviewed articles, technical
reports, book chapters, news items and more. In addition to these types of citations, one also has to consider who is citing. In patents one must distinguish between citations given by the applicants of the patent or the inventors of the new product or process described in the patent on the one hand, and by the patent examiners on the other. Both types of patents citations have been important sources of data in quantitative science and technology studies. Citations in patents to other patents have been used in various ways to indicate ‘technological significance’, ‘social-economic value’ and ‘private economic value’ (Breitzmann & Mogee, 2002; Grupp & Mogee, 2004.). They are also increasingly interpreted as ‘paper trails' reflecting knowledge flows from the inventor or applicant of the cited patent to the inventor/applicant of the citing one, and used to analyse social networks of researchers or inventors (Breschi & Lissoni, 2004). Citations in patents to scientific literature are used to calculate an index of “Science Linkage.” In the words of Francis Narin and his colleagues at CHI Research, a firm pioneering in the analysis of patent citations: “Science Linkage is a measure of the extent to which a company’s technology builds upon cutting edge scientific research. It is calculated on the basis of the average number of references on a company’s patents to scientific papers, as distinct from references to previous patents. Companies whose patents cite a large number of scientific papers are assumed to be working closely with the latest scientific developments” (Narin, Breitzman & Thomas, 2004). As pointed out by Grupp and Mogee (2004) in their historical account of the development of S&T indicators, in order to meet the pressure to show the usefulness of government funded basic research, one of the key indicators was the number of patent citations to basic research. In the past decade several studies of the science base of modern technology were published. These related to newly emerging scientific fields, especially Nanoscience and Technology, and Biotechnology. There are several studies which looked into the characteristics of non-patent-literature citations in patents and looked into the “science intensity” of these patents from technological and industry convergence perspectives. Karvonen and Kassi (2011) for example studied the interaction between science and technology as they emerge through the comparison of patent and non-patent citations. Their analysis of 464,225 patent application and 506,225 non-patent references (NPR) in the period 1978-2006 revealed an average of 4.2 references to patents and 1.1 references to non-patent literature. Lo (2009) sketched the linkage between science and technology development through non-patent citation analysis in the field of genetic engineering. This study found that 90% of the citations that appear in these patents are non patent literature and the majority of those were journal articles. Similar studies were conducted in the fields of chemistry by Bregonje (2005) and nanotechnology by Hu & al. (2007). Unlike research in natural and life sciences, social sciences, as well as arts and humanities are more difficult to measure on the research-patent-revenue scale. These disciplines, by their very nature, explore personal, social, national and international phenomena over time and their results qualitatively inform policy and economy in ways that are not necessarily patentable. To the best of our knowledge, no systematic studies have been carried out to trace the relationships between technology and research in the social sciences or humanities. The current paper presents an analysis of one of the subject fields in social sciences, Library and Information Science, and its technological counterpart, Information Technology. 372
The field of Library Science has always been considered a hybrid area of research which also evolved over time to include Information Science. Aligning more closely to Social Sciences in its early years, the field expanded to include elements of computer Science and Information Management. Since the overwhelming part of journals included in our study sample contain the term “library” in their titles, this set will be labelled as that of ”Library Science” journals. But the link with Information Science is strong. Nowadays several journals exist containing both the term ”library science” and ”information science” in their titles, and the two subject fields cannot be strictly separated. Examining the field and its development from paper to electronic information solutions, one might assume that technology was the driver of this transformation. This paper will show that in fact, it was library science research that informed and inspired the development of information retrieval solutions, sometimes years before the technology was available to translate it into viable algorithms and computerized modules. Several studies, especially those by Lilien, G.L., Morrison, P.D., Searls, K., Sonnack, M. & von Hippel, E. (2002), Franke, N., von Hippel, E. & Schreier, M. (2006) and Oliveira, P. & von Hippel, E.A. (2009) address the importance of lead-users in the innovation process and new products developments. These studies highlight the processes by which users are leading the development of solutions based on their experiences and the needs that they identify in their own environments. The findings above complement these previously published studies whereas users drive innovation by developing solutions which are later on adapted in the market. The purpose of this study is to demonstrate the technological and economical viability of Library Science and to show the areas of technology where research in this discipline had the largest influence. Influence was measured by analyzing the manners by which articles that were published in library journals are cited in patents. Our analysis addressed the following aspects: (a) How many library journals were cited in the patents covered by TotalPatent™ between 19922011, and how often? (b) Which articles were cited most frequently? (c) How can one characterize the content of the cited articles and the patents citing these, using keywords or subject classification systems? (d) Who were the assignees of the patents citing library journals? (e) What was the time delay between the publication year of the cited work and the year of application of the citing (granted) patent? Data and Methodology We examined leading library journals showing a high SNIP score. SNIP is a journal metric taking into account the citation behaviour and characteristics in the subject covered by a journal. The journals included in this study are covered by Scopus™, a scientific database covering more than 18,000 peer reviewed journals and conferences proceedings. Scopus™ journal analytics includes the SNIP metric which allows a comparison of subject-related journals; in our case, Library & Information Science journals. In this case study, the search was limited to journals containing the phrase “library science” or the word “library” in their titles.
373
The Scopus™ SNIP journal ranking analysis retrieved 42 journals which were then searched for, by using the Non-Patent-Literature citation field in TotalPatent™. TotalPatent™ is a comprehensive database covering applications and patents granted at/by a large number of patent offices around the world, including the US (USPTO), European (EPO) patent offices and World Patent Office (WPO) from 1992. In the second phase, all patents citing these journals were retrieved and the non-patent-literature cited in them was extracted. These citations were manually analyzed and all the library journals’ articles were collected. The third phase of the study involved a built-up of a small database that included various data field such as Journal Title, Total Number of Citations, Number of Unique Cited Articles, Unique Articles Titles, and Year of publication, Number of Citations, Patents Numbers, Patent Titles, Filing/Issue Dates, Inventor, Assignee, and Classifications. It must be noted that the numbers of citations presented are approximate, due to unexpected variations in the journal titles included in the non-patent citations, and to double counts because of the occurrence of patent families of more or less identical patents submitted to multiple patent offices. Findings Out of an initial list of 42 library journals, 8 were found to be cited in patents covered by TotalPatent™. These are listed on the horizontal axis of Figure 1. In addition to the total number of citations, we also examined the number of unique articles cited (see graph below). The Journal of the American Society for Information Science and Technology was the highest cited with a total of 76 citations overall and 24 unique articles cited. Library Hi-tech and Library Journal followed with 58 and 50 total citations and 17 and 13 unique articles citations respectively.
Figure 1. Citations to Library Science journals.
374
In order to better understand the themes covered in the articles and sketch the domains to which they pertain, we collected the articles’ author given, index and titles keywords from Scopus™ and built a word cloud that features them presented in Figure 2. The word cloud was created using Wordle™ a free web-based application that enables the generation of word clouds from free text. In order to create an accurate word cloud as possible, phrases within the titles and keywords were kept by using Wordle™ advanced functionality.
Figure 2. Word cloud of keywords and titles of articles in Library Science journals cited in patents.
The analysis of the articles keywords as demonstrated by the word cloud shows that the articles were featuring information retrieval and indexing, information and documents management systems which pertain to electronic and digital libraries development. This finding was of particular interest because the years of publications saw peaks in the years when the electronic library and automated information retrieval systems were beginning to take shape in terms of research. The graph below which features the publication years of the cited articles, clearly demonstrates relatively high citations of articles that were published at the end of the 1980s and late 1990s when information retrieval and management research was flourishing.
375
Figure 3. Publication years of Library Science articles cited in patents.
A further analysis showed that the time lapse between the publication year of the articles and the year of application of the patents citing them is significant, ranging from 10 to 20 years. This indicates both technical and conceptual developments within the field before the technology was there to apply its broader concepts such as online commerce. A good example that demonstrates this is NOTIS: The System and Its Features, published by James Meyer (1985) in Library Hi-Tech. This article was cited 11 times in patents published between 1999 and 2006. The article featured an online library management system that integrates the public access catalogue and in addition included acquisitions, serials management, authority control, and circulation. Patents citing this article include information management system as well as online purchasing systems that handle products management, purchasing and exchanges. Another example is “MAGGIE III: The Prototypical Library System” article, published by Kenneth E. Dowlin, (1986) in Library Hi Tech. This article featured an integrated library system that supports a public access catalogue and included a cataloguing interface, bibliographic maintenance, circulation, electronic mail, and community information databases. The article was cited 10 times in patents published between 1999 and 2008. The patents citing this article made use of some of its concepts to develop electronic commercial sites that manage information such as sales transactions and processing of products registration and returns. To be able and visualize the subjects covered by the citing patents, the titles’ keywords were collected and constructed in a word cloud (see image below). As can be seen, the patents focus on electronic information administration, navigation, and products and services management in commercial systems.
376
Figure 4. Word clouds of keywords and titles of patents citing Library Science journals.
The subject areas as they emerge from the titles’ keywords correspond to the major classes to which the patents were assigned. When analyzing the classifications of the citing patents it was evident that a large majority of them fall in the areas of the Data Processing category with subcategories ranging from financial, business, and databases structure to digital processing (See graph below). For example in the patents keywords word cloud the topics of information systems, personalization, and computers are clearly brought forth while the classifications pertain to parallel application in the areas of computer processing which in turn correspond to the heavy emphasis on information management in the journals articles. The thread of information and data management combined with customer management and personalization is carried through the articles keywords and the patents titles and classifications.
Figure 5. Number of patents citing Library Science journals by class.
377
An examination of assignees revealed 55 unique corporate entities with only one exception of a university. Looking at the top assignees, one can notice the domination of information management companies as well as online purchasing and commercial corporations.
Figure 6. Top assignees of patents citing library science journals.
Discussion In studies where the relationships between research and patents links were examined, the scholars looked for a direct linkage between a researcher and his/her patent application. The study presented in this paper was focused on finding citations of Library & Information Science journals in patents filed or granted between 1992 and 2011. The overall analysis of the citation of Library & Information Science journals in patents revealed some interesting observations. First, the most cited journals in this field are those which cover research studies that pertain to software development especially in the domains of information and/or data management. In addition, the articles’ keyword analysis as it is shown in the word cloud strongly indicates the themes of information and documents retrieval which include indexing, mining browsing etc. Other themes indicating the diversity within the field were those pertaining to multimedia management, graphics retrieval and the web. This is of particular interest considering the fact that these articles were mostly written when the internet was in its infancy which are indicators of forward looking and innovative approaches within the field. Examining the citing patents and analyzing their titles keywords showed a strong focus on information systems but also on products which correlates to the above articles’ content and to the overall classifications being in the areas of data processing. The modules featured in these articles were originally developed for library transactions management systems and inspired commercial uses in online commerce. The library system serving the public and exchanging different types of commodities such as books, audio and video
378
items etc. has unique properties that allow for this relationship between commercial and public purchasing. The library systems support exchanges, client information management and public interfaces which are similar in essence to those needed for online purchasing. Overall, the analysis showed that library systems were developed before online commerce was conceived and in a way inspired their development. The time lapse between the articles’ publication year and the year of their citations in patents featuring systems and modules is of importance: These library systems were developed in a time when the internet as we know it today wasn’t in existence and demonstrated forward thinking and innovative breakthroughs that were turned to far reaching applications. Conclusions Studies in the area of library science are cited in patents mainly when they examine or report on computerized systems that manage data. Indexing, classification, records management and user interfaces are some of the topics reported on in these articles. These topics were found to be linked to patents that pertain to online commerce systems that feature much of the same requirements as a library system would. One of the main findings of this study was the long time lap of about 20 years between the publication of the articles and the time they were cited in patents. This finding bares the conclusion that library systems and records management methodologies were answering an advanced set of remote customers needs using computerized solutions long before the concepts of online commerce and internet access were conceived. The articles cited feature innovative and forward thinking library systems that embedded both records management (i.e. books and other media items) and customer management (i.e. returns, exchanges). Based on this study we believe that the linkage between research and patentability can and should be expanded to include examination of the patentability of additional social sciences and humanities research as ways to establish their impact on the economy and product development. Our study compliments the research done by Lilien, G.L., Morrison, P.D., Searls, K., Sonnack, M. & von Hippel, E. (2002), Franke, N., von Hippel, E. & Schreier, M. (2006) and Oliveira, P. & von Hippel, E.A. (2009). It clearly shows how the library science “community” breeds innovations long before they are adapted to and adopted in the broader markets. Further research In order to assess the impact and the influential of the cited references in library science journals were for the patents citing them, an interesting analysis would be to study the percentage of these library science references in the full reference list of the patent. Also, studying the other references would help in understanding in which aspects the library science references more important. One would compare the percentage of cited references in the library science field to tat calculated for references in other disciplines, and analyze whether this percentage is higher or lower compared to the volume of articles published in this discipline. Acknowledgement The authors would like to thank Jon Klein, Eric Van Stegeren and Oliver Curtis from the TotalPatent™ team at Lexis-Nexis for their generous assistance with collecting and accessing the patent data used in this article. 379
References Bregonje, M. (2005). Patents: A Unique Source for Scientific Technical Information in Chemistry Related Industry? World Patent Information, 27 (4), 309-315. Breitzman, A.F. & Mogee, M.E. (2002). The Many Applications of Patent Analysis. Journal of Information Science, 28 (3), 187-205. Breschi, S. & Lissoni, F. (2004). Knowledge Networks from Patent Data: Methodological issues and research targets. In W. Glänzel, H.F. Moed & U. Schmoch (Eds.), Handbook of Quantitative Science and Technology Research. The Use of Publication and Patent Statistics in Studies of S&T Systems (pp. 613-644). Dordrecht: Kluwer Academic Publishers. Franke, N., von Hippel, E. & Schreier, M. (2006). Finding Commercially Attractive User Innovations: A Test of Lead-User Theory. Journal of Product Innovation Management, 23 (4), 301-315. Grupp. H. & Mogee, E. (2004). Indicators for National Science and Technology Policy. In W. Glänzel, H.F. Moed & U. Schmoch (Eds.), Handbook of Quantitative Science and Technology Research. The Use of Publication and Patent Statistics in Studies of S&T Systems (pp. 75-94). Dordrecht: Kluwer Academic Publishers. Hu, D., Chen, H., Huang, Z. & Roco, M.C. (2007). Longitudinal Study on Patent Citations to Academic Research Articles in Nanotechnology (1976-2004). Journal of Nanoparticle Research, 9 (4), 529-542. Karvonen, M., & Kässi, T. (2011). Patent Analysis for Analysing Technological Convergence. Foresight, 13 (5), 34-50. Karvonen, M. & Kässi, T. (2011). Impact of Science on Technology Applications in Converging Technological Environments. Paper presented at the PICMET: Portland International Center for Management of Engineering and Technology. Kirschenbaum, S.R. (2002). Patenting Basic Research: Myths and Realities. Nature Neuroscience, 5 (suppl.), 1025-1027. Lilien, G.L., Morrison, P.D., Searls, K., Sonnack, M. & von Hippel, E. (2002). Performance Assessment of the Lead User Idea-Generation Process for New Product Development. Management Science, 48 (8), 1042-1059 Lo, S.-S. (2010). Scientific Linkage of Science Research and Technology Development: A Case of Genetic Engineering Research. Scientometrics, 82 (1), 109-120. Narin, F., Breitzman, A. & Thomas, P. (2004). Using Patent Citation Indicators to Manage a Stock Portfolio. In W. Glänzel, H.F. Moed & U. Schmoch (Eds.), Handbook of Quantitative Science and Technology Research. The Use of Publication and Patent Statistics in Studies of S&T Systems (pp. 553-568). Dordrecht: Kluwer Academic Publishers. Oliveira, P. & von Hippel, E.A. (2009). Users as Service Innovators: The Case of Banking Services. Research Paper, MIT Sloan School of Management. Retrieved October 20, 2009, from http://papers.ssrn.com/sol3/papers.cfm?abstract_id=1460751 Rosell, C. & Agrawal, A. (2009). Have University Knowledge Flows Narrowed? Evidence from Patent Data. Research Policy, 38 (1), 1-13. Saathoff, J. (2010). Technology Transfer at the Technical University of Braunschweig: Cooperation Projects, Patents and Start-Ups. [Technologietransfer an der Technischen Universität Braunschweig: Kooperationsprojekte, Patente und Existenzgründungen] PTB Mitteilungen Forschen Und Prufen, 120 (4), 308-311. Sampat, B.N. (2006). Patenting and US Academic Research in the 20th Century: The World Before and After Bayh-Dole. Research Policy, 35 (6), 772-789. Thursby, J.G., & Thursby, M.C. (2011). Has the Bayh-Dole Act Compromised Basic Research? Research Policy, 40 (8), 1077-1083. 380
