Abstract
(Englisch)
|
Measurement and assessment of technological innovation is a very specific scientific subject, which is extremely important for many actors, like for example public and private policy makers, organizations and individuals. Especially nowadays in a globalized economy, the development of new indicators that will provide a clear view of the S&T progress is of high importance for the economic performance of countries and regions.
The main objective of STING project is to develop efficient, innovative methodologies and tools for the analysis of existing information related to the technological innovation in a European level, based on data related to patents.
The objectives of the work-package 'Definition and Assessment of User Requirements' were to capture the real life user needs that should be covered by the system and to study the requirements for a user-friendly working environment. Furthermore, the identification and study of existing software in the market specialized in the analysis of patent data was performed. The following approach was applied: first, a user survey was carried out on the basis of a questionnaire with open and close questions. Then, the analysis of identified software specialized in patent data analysis was conducted, mainly on the Internet. A detailed description of the characteristics of every system, as well as a comparison was performed. Particular emphasis was given to the user requirements as these form the basis for the design of functional specifications.
The objectives of the work-package 'Statistical Analysis and Development of Statistical Methodology' were related to the study and development of a statistical methodology based on textual techniques that will enhance the ability of users to exploit information 'hidden' in patent data. The developed methodology supports multidimensional comparisons between countries, sectors of activity and companies. It also enables to identify competition in given sectors of technology or between societies and countries. Interactions that might exist between domains of technological activity and poles of innovation inside these domains are also captured. Furthermore, indicators for the measurement of innovation in different fields of technology were proposed. Within this work package, the STING project makes use of advanced statistical techniques, as well as of various NLP preprocessing techniques for the exploitation of the textual information stored in patents. Correspondence and cluster analysis are the main multidimensional techniques used. Various patent based indicators were proposed in order to depict in a quantitative way the technological activity in a European level.
As far as robustness is concerned, the STING project relies on the production of confidence ellipsoids/convex hulls resulting from the use of bootstrap techniques in association with correspondence analysis.
The objectives of the work-package 'Information Dissemination and Exploitation' are related to the dissemination of the project and its results to all the interested parties. The dissemination is based on a dissemination plan that aims both at advertising the project results and the gained experience and at commercializing the targeted product. All partners were asked to define potential end users for the STING software. The aim was to establish a core of interested users that would act as mediate or intermediate beneficiaries of project results. A User Group was established and its first meeting has taken place in Lausanne in November 2001. In addition, an information leaflet was designed and distributed and a Web site, serving as a portal with patent related information, statistical analysis, similar projects and publications, was set up (http://sting.cti.gr).
|
