A Two-Step Agglomerative Hierarchical Clustering Method for Patent Time-Dependent Data

Patent data have time-dependent property and also semantic attributes. Technology clustering based on patent time-dependent data processed by trend analysis has been used to help technology relationship identification. However, the raw patent data carry more features than processed data. This paper aims to develop a new methodology to cluster patent frequency data based on its time-related properties. To handle time-dependent attributes of patent data, this study first compares it with typical time series data to propose preferable similarity measurement approach. It then presents a two-step agglomerative hierarchical technology clustering method to cluster original patent time-dependent data directly. Finally, a case study using communication-related patents is given to illustrate the clustering method.

[1]  Murat Bengisu,et al.  Forecasting emerging technologies with the aid of science and technology databases , 2006 .

[2]  Lior Rokach,et al.  Data Mining and Knowledge Discovery Handbook, 2nd ed , 2010, Data Mining and Knowledge Discovery Handbook, 2nd ed..

[3]  Alan L. Porter,et al.  Emerging technologies: quantitative identification and measurement , 2010, Technol. Anal. Strateg. Manag..

[4]  Donald J. Berndt,et al.  Using Dynamic Time Warping to Find Patterns in Time Series , 1994, KDD Workshop.

[5]  Yen-Liang Chen,et al.  A three-phase method for patent classification , 2012, Inf. Process. Manag..

[6]  R. S. Campbell,et al.  Patent trends as a technological forecasting tool , 1983 .

[7]  Alan L. Porter,et al.  Forecasting Innovation Pathways (FIP) for new and emerging science and technologies , 2013 .

[8]  J. H. Ward.  Hierarchical Grouping to Optimize an Objective Function , 1963 .

[9]  Sungjoo Lee,et al.  Technology clustering based on evolutionary patterns: The case of information and communications technologies☆ , 2011 .

[10]  Eamonn J. Keogh,et al.  On the Need for Time Series Data Mining Benchmarks: A Survey and Empirical Demonstration , 2002, Data Mining and Knowledge Discovery.

[11]  Dina Q. Goldin,et al.  On Similarity Queries for Time-Series Data: Constraint Specification and Implementation , 1995, CP.

[12]  Amy J. C. Trappey,et al.  Using patent data for technology forecasting: China RFID patent analysis , 2011, Adv. Eng. Informatics.

[13]  Sungjoo Lee,et al.  Modeling and analyzing technology innovation in the energy sector: Patent-based HMM approach , 2012, Comput. Ind. Eng..

[14]  T. Warren Liao,et al.  Clustering of time series data - a survey , 2005, Pattern Recognit..

[15]  Kwangsoo Kim,et al.  TrendPerceptor: A property-function based technology intelligence system for identifying technology trends from patents , 2012, Expert Syst. Appl..

[16]  Martin J. Adelman.  The America Invents Act, Its Unique First-to-File System and Its Transfer of Power from Juries to the United States Patent and Trademark Office , 2012 .

[17]  Eamonn J. Keogh,et al.  An Enhanced Representation of Time Series Which Allows Fast and Accurate Classification, Clustering and Relevance Feedback , 1998, KDD.

[18]  Lior Rokach,et al.  Data Mining And Knowledge Discovery Handbook , 2005 .

[19]  Türkay Dereli,et al.  Enhancing technology clustering through heuristics by using patent counts , 2011, Expert Syst. Appl..