Patent text mining and informetric-based patent technology morphological analysis: an empirical study

Patent technology morphological analysis utilises the advantages of both patent information analysis and morphological analysis to provide more detailed information on current and future patent technology. Current patent technology morphological analysis methods are largely reliant on manual expertise in the construction of morphological boxes with few approaches to the evaluation of future morphological configurations. We developed a patent text mining and informetric-based patent technology morphological analysis technique. We defined the basic parameters of the morphological box as the factors in factor analysis, with a patent keywords matrix, and the clusters in clustering analysis with factor scores. Patent citation, year of patent registration, keyword frequency, and contributing factors were used to evaluate future morphological configurations. We used an empirical study of liquid crystal display wide viewing angle patents to validate our method. The Thomson Reuters Derwent Innovation Index was used to collect patent text datasets. Our study indicates that this method is feasible for the implementation of patent technology morphological analysis. Our method provides advantages in terms of cost and time reductions during morphological box construction and more flexible methods for evaluating morphological configurations. We discuss problems and possible development of our method in the last section.     

Extracting the commercialization gap between science and technology — Case study of a solar cell

In this paper, we compared structures of the citation network of scientific publications with those of patents, and discussed the differences between them. A case study was performed in a solar cell to develop a method of detecting gaps between science and technology. Scientific research has tended to be more basic, especially in terms of cell design, whereas patents have focused on more applied technology used in solar cell modules. Of the major citation clusters of scientific publications, only two, namely silicon and compound solar cells, corresponded semantically with patent clusters. Conversely, there were no patent clusters corresponding to the other two scientific research fronts, namely dye-sensitized and polymer solar cells. These research areas could be regarded as opportunities for industrial commercialization because scientific activities exist but not technological applications. Our results could offer an intellectual basis for discovering potential opportunities for industrial commercialization.     

Monitoring trends of technological changes based on the dynamic patent lattice: A modified formal concept analysis approach

The strategic importance of monitoring technological changes is highlighted given the ever faster pace and increasing complexity of technological innovation. In this respect, patent citation analysis has been the most frequently adopted tool among others. However, patent citation analysis is subject to certain drawbacks that stem from only consideration of citing-cited information and time lags between citing and cited patents. This study proposes a formal concept analysis (FCA)-based approach to developing a dynamic patent lattice that can analyze complex relations among patents and monitor trends of technological changes. The FCA is a mathematical tool for grouping objects with shared properties based on the lattice theory. The distinct strengths of FCA, vis-á-vis other methods, lie in structuring and displaying the relations among objects from a massive amount of data. For the purpose of technology monitoring, the FCA is modified to take into account time periods and changes of patent keywords. A patent context is first constructed with the aid of domain experts and text mining technique. Two types of dynamic patent lattices are then developed by executing the modified FCA algorithm. A case study of laser technology in lithography for semiconductor manufacturing shows that the suggested dynamic patent lattice has considerable advantages over conventional patent citation maps in terms of visualization and informative power.     

A stochastic patent citation analysis approach to assessing future technological impacts

The anticipation and forecast of technological changes are of vital importance, as technological advances become increasingly fast and complex. What is at the core is identification of the current technologies that will drive technological changes over the coming few years. In this respect, numerous approaches have been devised to assess future technological impacts based on patent citation information, but do not provide a fair reflection of dynamic and idiosyncratic aspects of technological impacts as they are deterministic methods based on simple citation counts. We propose a stochastic patent citation analysis that can assess future technological impacts in a time period of interest by employing the future citation count as a proxy. At the heart of the proposed approach is a Pareto/NBD (Negative Binomial Distribution) model for taking into account the dynamic and idiosyncratic aspects of technological impacts. A patent citation matrix is first constructed for each time unit with citation patterns of the past. The future technological impacts are then derived by Pareto/NBD sub-model and gamma–gamma sub-model. A case study of the display technology patents is presented to illustrate the proposed approach. We believe our method can be employed in various research fields, from narrow patent valuation, to broad technological analysis and planning.     

Exploring the effect of structural patent indicators in forward patent citation networks on patent price from firm market value

This paper introduces structural patent indicators in the forward patent citation networks (FPCNs). Using the estimates of their relationship with firm market value, it explores the effect of the structural patent indicators on patent price. When applied to US smartphone industry and US drug and biotechnology industry, the common finding is that a firm's registered patent gets higher price if influenced faster by the previously registered patents of the same firm through self-citations. The empirical results imply that merely counting forward citations, shown related to patent price by the previous works, is insufficient; the structural properties in FPCNs are essential for the more accurate estimation of patent price. They also show that the relationships between structural patent indicators in the FPCNs and patent price are different between industries. It urges the necessity of selecting structural patent indicators, well fitted to each industry, for the better patent valuation using forward citations.     

Schumpeter,Schumacher and the Greening of Technology

We propose a semantic patent claim analysis that can examine patents for possible infringements and identify which needs to be manually perused. So far, numerous approaches have been devised to systemise this burden, but have not been useful in practice because of a lack of consideration of semi-structure of patent claim data and claim element-based procedure of adjudicating patent infringement. At the heart of our method is a hierarchical keyword vector for representing the dependency relationships among claim elements (as well as unstructured textual information) and a tree matching algorithm for comparing claim elements of patents. A case study of the patents about DNA chip technology shows our method has considerable advantages in terms of accuracy and significance. We believe the suggested method could be employed in various research areas and serve as a starting point for developing more general models.     

Monitoring the organic structure of technology based on the patent development paths

As the strategic importance of understanding changes in technology for successful business of most firms increases, the ability to analyze and monitor the current stage and history of technology is reckoned as a critical asset both for gaining competitive advantage and identifying promising niches. Patent citation networks have been widely used for systematic and empirical analysis of technology development. Understanding of technology's detailed changes in large patent citation networks, however, is difficult to achieve because of these networks' large and complex structures. To overcome this problem, we suggest an algorithm that identifies patent development paths from a large patent citation network by evaluating the weight of citations between patents. We then apply this algorithm to flash memory patents in an empirical study. Our algorithm is a new methodology that can be used to analyze the dynamic and complex structure of individual technologies.     

基于专利分析的产业共性技术识别概念框架

在重新定义产业共性技术的基础上,初步构建了经由综合文本挖掘、德温特手工代码共现、专利引用3个环节而确立的产业共性技术识别概念框架。该框架从3个评估视角———技术基础性、技术应用范围及技术社会效益,对产业共性技术的3个特性进行逐步识别,最终筛选出产业共性技术。     

Quantitative mapping of patented technology — The case of electrical conducting polymer nanocomposite

This study aims to obtain global technology evolution by constructing and analyzing patent citation network and patent citation map for the field of electrical conducting polymer nanocomposite. A total of 1421 patents are retrieved from USPTO patent database and patent citation network is established by combing both patent citation and social network analysis. Network properties, e.g. Degree Centrality, Betweenness Centrality, and Closeness Centrality, are calculated for representing several technology evolution mechanisms that first proposed in this study. Also, a distance-based patent citation map is constructed by calculating relative distances and positions of patents in the patent citation network. Quantitative ways of exploring technology evolution are investigated in this study to unveil important or emerging techniques as well as to demonstrate dynamics and visualization of technology evolutions.     

Patterns of technology life cycles: stochastic analysis based on patent citations

Patent analysis has been considered as an effective means of estimating phases of a technology life cycle. However, previous studies have not considered the dynamic and idiosyncratic aspects of a technology’s progression since they were based on deterministic methods, mainly fitting s- or double s-shaped curves to patent application counts. Moreover, previous methods cannot be executed at the individual patent level. We propose a stochastic technology life cycle analysis to trace the phases of a technology’s progression based on patent citations and identify the patterns of technology life cycles at the individual patent level. At the heart of the proposed approach are a hidden Markov model to estimate the probability of a system being at a certain hidden state from observation and cluster analysis to group a set of objects according to their similarities. A case study of patents about laser technology in lithography is presented.     

Can patent family size and composition signal patent value?

Recent research has proposed a method of patent valuation based on weighting patent family size by the market size of the countries in the family. The premise is that inventors tend to seek greater international coverage for their more valuable patents. The paper presents a novel way to test the ability of market size-weighted patent families to predict patent value and compares the method against extant measures of patent valuation based on patent citations and renewal behaviour. We use forecasting techniques to show that the weighted patent family size measure outperforms other methods in terms of predicting patent life and the number of citations. An advantage of the weighted patent family size measure is that it is based on ex-ante information and is easy to construct for purposes of evaluating patent value. We demonstrate this advantage using a large, comprehensive database of international patent families.     

地理邻近与知识产权——基于中国1 331项新能源技术发明专利的实证研究

虽然远程交流技术快速发展，但现有研究表明专利发明在很大程度上仍依赖于信息的地理邻近。以中国1992-2009年的1 331个新能源发明专利为研究对象，运用Jung Wonn,Jaffe等的实验设计方法，研究了专利引证的地理邻近是否会随时间递进而增大。结果表明，个人、大学、公司及其它种类发明专利引证的地理邻近会随着时间变化而增大，而政府受到的影响较小。另外，由于区域城市分属不同省份，因此并不能证明区域专利引证的地理邻近也会随着时间变化而增大。     

A semantic analysis approach for identifying patent infringement based on a product–patent map

Identifying patent infringement beforehand is highly important for reducing the risk of damages. However, as the approach to patents is changing from the technological and legal approaches of the past to the business approach, strategic patent infringement litigation and securement of patents can be important tools for companies. From various viewpoints, companies can consider an aggressive patent infringement litigation as well as evasion strategy of patent infringement litigation to keep competitors in check. Previous research on patent infringement identification only considered the possibility of infringement between patents. However, infringement cases between patent and product are prevalent. Thus, this research aims to suggest an automated method for identifying patent infringement between the patent and product in business. To this end, this paper proposes the subject–action–object (SAO) semantic technological similarity-based product–patent method for generating infringement maps. Several indices and subgrouping methods are suggested to interpret the map. As an exemplary case, data on technology and products related to the light emitting diode (LED) lamp are collected and exploited.     

专利特征对加入专利池的影响——基于MPEG LA的实证研究

什么特征的专利更容易加入专利池？以MPEGLA专利池内1073件专利与池外1073件控制专利为样本,实证研究专利技术范围、权利与地域范围、审批与维持时间、前向与后向引用对加入专利池的影响。结果表明,技术和地域覆盖范围越广,维持时间越长,专利引用与被引数次越多,越有可能加入专利池;专利审批时间越长,非专利文献引用越多,越难加入专利池;权利要求范围对专利能否入池无显著影响。研究结论对企业专利文件撰写、专利入池申请及专利池管理机构开展专利评估具有重要意义。     

基于关键节点与关键路径的专利集群网络演进研究

专利作为技术和知识的重要载体，是研究技术演进和产业发展的重要信息源。目前缺乏针对专利集群网络中关键节点与关键路径中节点的比较分析和技术知识挖掘。构建基于关键节点和关键路径的专利集群网络演进模型，从两个层面综合分析技术演进特征。检索德温特数据库(Derwent Innovation Index)得到碳化硅肖特基势垒二极管（SiC-SBD）相关专利作为实证数据，时间跨度为1986-2017年。结果表明，SiC-SBD专利集群网络经历了萌芽期、成长期、成熟期和衰退期，其中，萌芽期专利主要涉及基础技术，成长期为器件结构，以完善器件结构、改进二极管性能为主要研究方向。近年来SiC-SBD专利年增长率下降，技术发展速度放缓，专利价值更多体现在其商业价值上。SiC-SBD专利集群网络关键节点与关键路径中节点重合度较高，核心专利识别可通过多视角分析得到。通过关键节点与关键路径分析专利集群网络演进过程，有助于更全面呈现技术演进过程，为决策者识别核心专利、预测技术发展提供参考。     

Exploring technology diffusion and classification of business methods: Using the patent citation network

Among the many technology forecasting indicators, patents and patent citations are useful and important indicators. The more frequently a certain patent is cited by subsequent patents, the more the related technology can be said to be diffused, implying that the technology is more widely applied and thus more valuable. This paper analyzes the business methods technology which retrieves patents from the USPTO database. There are two purposes of this paper: 1. establish the indicators for finding basic patents and measure the relationship of these basic patents; 2. classify the basic patents and explain the groups of technology diffusion. Finally, this study identifies the two mainstreams in business method technology: one is focused on marketing technology, and the other one stresses on data security. Both are important for Internet data processes or e-commerce activities.     

Measuring knowledge persistence: a genetic approach to patent citation networks

The aim of this paper is to propose a new empirical method for identifying technologically important patents within a patent citation network and to apply it to the telecommunication switching industry. The method proposed is labelled the genetic approach, as it is inspired by population genetics: as geneticists are interested in studying patterns of migration and therefore the common origins of people, in innovation studies we are interested in tracing the origin and the evolution of today knowledge. In the context of patent and citation networks, this is done by calculating the patent’s persistence index, i.e., decomposing patent’s knowledge applying the Mendelian law of gene inheritance. This draws on the idea that the more a patent is related (through citations) to “descendent” patents, the more it affects future technological development and therefore its contribution persists in the technology. Results show that the method proposed is successful in reducing the number of both nodes and links considered. Furthermore, our method is indeed successful in identifying technological discontinuities where previous knowledge is not relevant for current technological development.     

专利联盟中专利许可费的计算方法

专利联盟中专利的组成有三种情况:完全排斥型、完全替代型和部分排斥部分替代型。由于专利联盟是开放型的体系,专利加入时间和起作用时间不同,专利之间具有异质性,于是专利联盟中价值的评估和专利许可费的计算产生了变数。为此,本文从单项专利价值的计算入手,考虑专利联盟中专利的不同类型、不同的加入时间和起作用时间等情况,提出专利联盟中专利组合价值的评估方法,在此基础上借助财务分析中内部收益率来计算专利联盟中的许可费。     

Tracing the Evolution of Biomedical 3D Printing Technology Using Ontology-Based Patent Concept Analysis

Three-dimensional (3D) printing enables products to be custom-designed and produced using additive manufacturing processes. This research develops a patent analysis approach to explore biomedical 3D printing technology trends. First, the method searches for related patents. Second, frequently appearing key terms in patents are extracted as a means to identify key concepts of inventions. Third, in order to structure the domain knowledge, the ontology is created by referring to the relevant key terms and literature. The key terms create indexes used to measure similarities for clustering the patents and sub-technologies. Based on the patent context and dynamic patent concept analysis maps, the evolutionary trends of technology development are depicted. Biomedical 3D printing R&D projects are used as case examples to compare against co-occurring patent evolutions. The proposed patent concept analysis is generalisable for critical decision support of R&D planning and evaluation in any market sector.     

Patent portfolios and knowledge flow(s) of photovoltaic companies

The current trend towards green energy is encouraging manufacturers to invest in photovoltaic technologies. In order to guarantee R&D and optimal operational performance, operators should be able to identify the technological advantages of their competitors for the conduct R&D and to ensure adequate technological knowledge intake. Therefore, from the perspective of patent portfolios and knowledge flow(s) of photovoltaic companies, this study adopted two approaches namely, patent counts and patent citations, to discuss the technical capability of R&D portfolios and the technological knowledge flow. Three patent indices were utilised for the integration of R&D portfolios which include: technology attractiveness, relative patent position, and revealed patent advantage. Technological knowledge flow allows construction of a patent citation network through backward citation of patents. Sources and movement directions of technological knowledge are measured by calculating the relative citation propensity. R&D portfolios and knowledge flow are complementary perspectives of each photovoltaic company.