為了掌握科技與產業趨勢之脈動，面對知識經濟時代接踵而來的挑戰，企業運用科技預測提供之訊息進行決策與佈局。科技預測可用於評估科技是否達到發展瓶頸，預測手法亦推陳出新。其中對於預測結果最顯著影響的關鍵，在於分析資料本身的意義，過去科技預測主要以市場占有率或是功能方面成長等商業資料分析，然而學者Campbell (1983)研究指出，專利數量的文章篇數比科學刊物中的文章篇數，更能表現技術發展的起伏。根據世界智慧財產局 (WIPO, 1996)報告，專利是唯一能揭露核心技術的知識文件，其傳遞科技核心的比率遠高於期刊、報導或是研究報告，因此專利本身也能作為科技預測的分析對象，更延伸出將每年專利數量視為「知識市場」的市場占有率的概念。在本論文中，我們提出一個以非耗盡分群方法運用於科技預測系統，非耗盡分群方法最大的特色在於一個物件可能被分群在多個群集中，能夠表達專利文件本身有多元的內容與意涵。藉由定義好的RFID本體論以及NTF-IDF為基之技術，擷取該領域代表字進行分群，藉以獲得專利文件之群集。分群步驟完成之後，使用者能夠依照專利文件群集進行科技預測，可獲得各群集隨時間消長起伏之趨勢。本論文預期成果是系統能精確的擷取領域關鍵辭彙，以群集彼此可重疊的分群結果作為預測之輸入，而預測之趨勢能以視覺化圖形呈現，幫助使用者或專利工程師能夠快速找到所需的專利隸屬的群集，以及這群專利的未來發展趨勢，以利企業之間專利技術之攻防。

Facing the challenges from a knowledge-based economy, having a comprehensive understanding insights of technology and industry development is the basic and necessary requirement to gain competitive edges for enterprises. Consequently, enterprises apply technology forecasting techniques to assist generating useful information for further R&D strategic decisions. Nonetheless, current technology forecasting analyses base mostly on macro-indicators such as market share and growth rate rather than on specific technology development information, such as invention and patents of certain technology. In Campbell’s (1983) research, he found that patent documents often better expresses the development trend of technology when compare with ordinary scientific journals. Moreover, according to the report of WIPO (1996), patent documents can better reveal the core technology and innovation than other knowledge documents such as journal papers and technical reports. As a result, we try to incorporate patent analysis while proceed technology forecasting.
In this research, a non-exhaustive clustering methodology is proposed as the basis for a novel technology forecasting system. Non-exhaustive clustering methodology allows overlapping of patent documents, which is plausible when any patent can claim multiple key technical inventions. The characteristic of non-exhaustivity emphasizes that one patent contains multiple technology breakthroughs. We use Radio Frequency Identification (RFID) as case example in this research. RFID ontology is constructed. Afterward, refined Normalized Term Frequency/Inverse Document Frequency (NTF-IDF) key-phrase extraction methodology is developed to extract representative key phrases for following clustering procedure. Finally, the non-exhaustive clustering methodology is applied to generate overlapping clusters of patents. The clustering results and analysis of growth trend for each cluster provide users a clear view of patent distribution in a given broad technology area (e.g., RFID). The expected results of this research contain extracting domain key phrases precisely, using the non-exhaustive clustering results as input data of technology forecasting and finally visualize the technology trend. This system enables R&D engineers and managers to find the existing patents related to their interested technical domains (clusters) and enable them to strengthen their R&D efforts offensively and defensively.

1.Armstrong, J. S., 2001, “Principles of Forecasting: A Handbook for Researchers and Practitioners,” Kluwer Academic Publishers, Boston, Massachusetts.
2.Berkhin, P., 2002, “Survey of clustering data mining techniques,” Technical Report, Accrue Software.
3.Berry, M. J. A. and Linoff, G.., 1997, “Data Mining Techniques: For Marketing, Sale, and Customer Support,” John Wiley & Sons Inc., pp. 55.
4.Campbell, R., 1983, “Patent Trends as a Technological Forecasting Tool,” World Patent Information, Vol. 5, No. 3, pp. 137-143.
5.Chen, B., Tai, P.C., Harrison, R., and Yi Pan, 2005, “Novel Hybrid Hierarchical-K-means Clustering Method (H-K-means) for Microarray Analysis,” Computational Systems Bioinformatics Conference, Aug. 8-11, Sandford CA, USA.
6.Chen, C. P. (Advisor: Prof. Trappey, C. V.), 2004, “The Test of Technological Forecasting Models: Comparison between Extended Logistic Model, Fisher-Pry Model, and Gompertz Model,” MA Thesis, Department of Management Science, National Chiao Tung University, Hsinchu, Taiwan.
7.Chen, E., and Wu, G., 2005, “An ontology learning method enhanced by frame semantics” Proceedings of the Seventh IEEE International Symposium on multimedia, ism, pp. 374-382.
8.Chen, T. S., Tsai, T. H., Chen, Y. T., Lin, C. C., Chen, R. C., Li, S. Y., And Chen, H. Y., 2005, “A Combined K-means and Hierarchical Clustering Method for Improving the Clustering Efficiency of Microarray,” Proceedings of 2005 International Symposium on Intelligent Signal Processing and Communication Systems, pp. 405-408.
9.Clarke, D. W., 2000, “Strategically Evolving the Future: Directed Evolution and Technological Systems Development,” Technological Forecasting & Social Change, Vol. 64, pp. 133-153.
10.Faust, K. and Schell, H., 1983, “International Patent Data: Their Utilization for the Analysis of Technological Developments,” World Patent Information, Vol. 5, No. 3, pp. 144-157.
11.Gupta V. K., 1999, “Technological trends in the area of fullerenes using bibliometric analysis of patents,” Scientometrics, Vol. 44, No. 1, pp. 17-31.
12.Han, J. and Kamber, M., 2000, “Data Mining: Concepts and Techniques,” Morgan Kaufman.
13.Hong S., 2006, “The Magic of Patent Information,” The World Intellectual Property Organization, http://www.wipo.int/sme/en/documents/patent_information.htm, 2006/10/25
14.Hu, H. L. (Advisor: Prof. Chen, Y. L.), 2006, “Optimization in Data Mining An Overlapping Cluster Algorithm to Provide Non- exhaustive Clustering,” Ph.D. Thesis, Department of Information Management, National Central University, Chung-Li, Taiwan.
15.Hsu, F. C. (Advisor: Prof. Trappey, A. J. C.), 2006, “Intelligent Patent Document Analysis Based on Clustering and Categorization Methods,” Ph.D. Thesis, Department of Industrial Engineering and Engineering Management, National Tsing Hua University, Hsinchu, Taiwan.
16.Huebner, J., 2005, “A possible declining trend for worldwide innovation,” Technological Forecasting & Social Change, Vol. 72, pp. 980-986.
17.Jones, K. S., 1972, “A Statistical Interpretation of Term Specificity and Its Application in Retrieval,” Journal of Documentation, Vol. 28, No. 1, pp. 11-20.
18.Kowalski, G. J., and Maybury, M. T., 2000, “Information Storage and Retrieval Systems: Theory and Implementation,” Kluwer International Series on Information Retrieval, Inre 8. Kluwer Academic.
19.Liu, B., Xia, Y., and Yu, P. S., 2000, “Clustering Through Decision Tree Construction,” Proceedings of the ninth international conference on Information and knowledge management, pp. 20-29.
20.Luhn, H. P., 1957, “A Statistical Approach to Mechanized Encoding and Searching of Literary Information,” IBM Journal of Research and Development, Vol. 1, No. 4, pp. 309-317.
21.MacQueen, J. B., 1967, “Some methods for classification and analysis of multivariate observations,” Proceedings of the Fifth Berkeley Symposium on Mathematical Statistics and Probability, Vol. 1, pp. 281-297.
22.Martino, J. P., 1993, “Technological Forecasting for Decision Making,” McGraw-Hill, pp. 1-5.
23.Meade, P. T. and Rabelo, L., 2004, “The technology adoption life cycle attractor: Understanding the dynamics of high-tech markets,” Technological Forecasting & Social Change, Vol. 71, pp. 667-684.
24.Mogee, M.E., 2000, “Foreign patenting behavior of small and large firms,” International Journal of Technology Management, Vol. 19, pp. 149-164.
25.Porter, A. L., Roper, A. T., Mason, T. W., Rossini, F. A., Banks, J. and Wiederholt B. J., 1991, “Forecasting and Management of Technology,” John Wiley & Sons, Inc.
26.Salton, G., and Buckley, C., 1988, “Term-Weighting Approaches in Automatic Text Retrieval,” Journal of Information Processing and Management, Vol. 24, No. 5, pp. 513-523.
27.Sedding, J., and Kazakov, D., 2004, “WordNet-based Text Document Clustering.” Proceedings of the Third Workshop on Robust Methods in Analysis of Natural Language Data (ROMAND), pp.104-113, Geneva.
28.Sharma, S. C., 1996, “Applied Multivariate Techniques,” John Wiley & Sons, Inc.
29.WIPO, World Intellectual Property Organization Website, http://www.wipo.int/.
30.Yoon, B.C. and Park, Y.T., 2004, “Morphology analysis approach for technology forecasting,” Proceedings, The International Engineering Management Conference, Oct. 18-21, Singapore.
31.張立典，2005，「以知識表徵為基之文件分群法」(指導教授：張瑞芬)，國立清華大學工業工程與工程管理研究所碩士論文。