Patent keyword network analysis for improving technology development efficiency

Choi, J.,Hwang, Y.S. American Elsevier 2014 TECHNOLOGICAL FORECASTING AND SOCIAL CHANGE Vol.83 No.-

The methods of patent analysis are largely divided into network-based patent analysis and keyword-based morphological patent analysis. Both methods have their shortcomings: internal patent information composed of natural languages cannot be analyzed in the network-based patent analysis method, and the correlation between patents cannot be analyzed in the keyword-based morphological patent analysis method. In this research, we analyze the patents of Light Emitting Diode (LED) and wireless broadband fields via a method that incorporates both the network-based patent analysis and the keyword-based patent analysis methods. And by using network indices, we identify the characteristics of the patent keyword network, and also perform a trend analysis to discover how keywords play a significant role in network changes over time. The analysis results indicate that the patent keyword network is sporadic but clustered and shows a clear power law distribution. Further, the inflow keywords are highly likely to tie new connections with other keywords in the existing associated communities. Also, we confirm the fact that, as time passes, the top core keywords of a particular technology field continue to play an important role in the network and that also the rate of technological changes in wireless broadband field is faster than that of LED. Through the proposed analysis, researchers can easily grasp what technology keywords are important in the specific technology field and identify the relations between the essential technology elements; furthermore, this information can be utilized for developing new technologies by combining these technology elements extracted from community analysis.

Zero-Inflated Poisson and Negative Binomial Regressions for Technology Analysis

Jong-Min Kim,Sunghae Jun 보안공학연구지원센터 2016 International Journal of Software Engineering and Vol.10 No.12

Technology analysis is to understand target technology by analyzing diverse information of developed technologies. Using the results of technology analysis, we can perform the technology management such as technology forecasting, technological innovation, and technology valuation for research and development (R&D) planning. In addition, the R&D planning is built uponin order to improve technological competitiveness of a company. Patent analysis is a popular approach to technology analysis. Many researches on patent analysis have been done because patent documents contain diverse and complete information on developed technology. However, the documents are not suitable for patent analysis based on statistics. So, in much of the work on patent data analysis, the researchers transformed the patent documents into structured data using text mining techniques. Generally, thestructured data set has a sparsity problem, that is, most elements of the data are zero valued. The existing researches in patent analysis have not considered this zero-inflated problem, but it places serious limits on performance when we analyze the patent data. In this paper, to overcome this problem, we propose a methodology for patent analysis using zero-inflated Poisson and negative binomial regressions. We apply the proposed methodology based on zero-inflated Poisson and negative binomial regression models to Apple’s technology analysis.

통계적 분석방법을 이용한 주제어 기반 특허동향분석 : 장갑기술(Armor technology)을 중심으로

최정섭 한국지식재산연구원 2013 지식재산연구 Vol.8 No.1

The field of technology having a long history of research and development warrants the amount of patent applications, but it is not easy to explore new areas of research. Technology related to the defense sector, whereas the public through a patent application is very limited. For these reasons, the case of national defense science and technology,tend to neglect the patent analysis in the pre-planning of the research and development. In this study, collected and analyzed patents related with armor technology to identify the technological trend and get the clues for new research and development area. The analysis method is divided into the basic analysis and keyword-based analysis. In the basic analysis was performed the quantitative analysis for the patent application trend,applicant, type of technology. In the keyword-based analysis, as the existing classification methods based on the keywords in terms of materials and structures classified and examined the chronological changes in technology. Also want to explore a new technology, how to classify the cluster analysis was attempted. Clusters through cluster analysis based on the frequency of keywords emergence grasp the technical characteristics. Particularly reflected the difference in the number of patents in the process of clustering analysis, explored a new field of research for deriving using the suggested index, technology emphasis. Armor technology has a long history but overall increasing patents application trends. The United States holds overwhelmingly many patents, the applicant usually the government and state-owned enterprises. Keyword-based analysis, armor technology of the existing classification methods, structure-related and material-related were separated. The classification is ambiguous, however, the presence of a number of keywords, there was a need to explore new classification scheme. For this, attempts to classify the advent of cluster analysis based on the frequency of keywords. Specifically to take advantage of the Technology Emphasis Index, suggest the possibility that novel research field of navigation. 연구개발의 역사가 긴 분야일수록 출원된 특허의 양이 대체로 방대하며, 새로운 연구 영역 탐색이 쉽지 않다. 국방과학기술은 특허출원을 통한 기술공개가 매우 제한적일 것으로 여겨지는데, 이는 연구개발의 사전 기획과정에서 필수인 특허동향분석을 소홀히 하는 원인으로 작용한다. 본 연구에서는 특허문서의 제목과 초록에 장갑을 포함하고 있는 특허를 수집하여 분석하였다. 분석방법은 크게 기본분석과 주제어 기반 분석으로 구분하였다. 기본분석에서는 연도별 특허출원 추이분석, 출원인 분석, 기술유형분석 등을 수행하였다. 주제어 기반 분석에서는 장갑기술의 기존 분류방식에 따른 연대별 기술변화를 살펴보았다. 또한 주제어 출현 빈도수를 기반으로 한 군집분석을 통하여 군집별 기술적 특성을 파악해 보았다. 특히 참신한 연구 영역 탐색을 위해 군집별 특허수의 차이를 반영한 분석지표를 제안하였다. 장갑기술은 역사가 오래되었음에도 불구하고 특허출원은 전반적으로 증가추세를 보이고 있다. 미국이 압도적으로 많은 특허를 보유하고 있고, 출원인은 정부와국영기업이 다수를 차지하고 있다. 주제어 분석을 통해 장갑기술은 기존 분류 방식인 구조기술과 재료기술로 구분이 가능함을 확인할 수 있었지만, 분류가 모호한주제어들도 상당수 존재하여 새로운 분류 방식이 필요함을 확인하였다. 주제어 출현 빈도를 기준으로 한 군집분석을 수행하여 그 가능성을 확인해 보았다. 또한 기술집중도 지표를 활용하여 참신한 기술의 탐색 가능성을 확인하였다.

Business planning based on technological capabilities: Patent analysis for technology-driven roadmapping

Lee, S.,Yoon, B.,Lee, C.,Park, J. American Elsevier 2009 TECHNOLOGICAL FORECASTING AND SOCIAL CHANGE Vol.76 No.6

This research responds to the needs of technology-driven business by focusing on how firms can find new business opportunities based on their technological capabilities. It proposes a technology-driven roadmapping processes that starts from capability analysis for technology planning and ends with business opportunity analysis for market planning. We suggest the use of patent data as a proxy measure of technological capability for this purpose and develop four analysis modules - Monitoring, Collaboration, Diversification, and Benchmarking - to support decision-making during the process. Various analysis techniques such as text-mining, network analysis, citation analysis and index analysis are applied to discover meaningful implications from the patent data, which are summarized in four maps - Actor-similarity map, Actor-relations map, Technology-industry map, and Technology-affinity map. For the purpose of illustration, RFID-related patents are collected and the 18 firms with the most patents used, focusing especially on the third biggest. We believe using roadmapping and patent analysis together can play complementary roles for each other. Putting roadmapping techniques together with patent analysis can increase the objectivity and reliability of technology roadmap, while using patent analysis restricted to technological information together with roadmapping techniques can ensure that a more valuable breadth of strategic information is extracted from patents.

장벽특허의 무효화 분석을 위한 선행특허기술 발굴 시스템 개발

정병기 ( Jeong Byeongki ),고남욱 ( Ko Namuk ),경진영 ( Kyung Jinyoung ),최덕용 ( Choi Dugyong ),윤장혁 ( Yoon Janghyeok ) (주)엘지씨엔에스(구 LGCNS 엔트루정보기술연구소) 2017 Entrue Journal of Information Technology Vol.16 No.1

기업에서 기술 아이템을 사업화할 경우 특허침해와 같이 사업화에 장애를 유발하는 장벽특허가 존재할 수 있다. 장벽특허가 존재할 경우, 기업은 무효화 전략, 회피설계, 비침해 논리개발과 같은 대응전략을 수립할 수 있으며, 본 논문은 장벽특허를 무력화시킬 수 있는 선행특허를 발굴하는 시스템의 제안을 목표로 한다. 미국특허청 특허에 대한 서지정보와 전문정보를 포괄하는 특허 데이터 베이스에 기반하여, 본 시스템은 장벽특허에 대한 관심어 그룹 설정 모듈, 임의의 특허가 관심어 그룹의 내용을 담고 있는 정도를 파악하는 정보엔트로피 계산 모듈, 토픽모델링을 통해 장벽특허와 임의의 특허 간의 의미론적 특허 유사도 계산 모듈을 거쳐, 최종적으로 장벽특허를 무효화시킬 수 있는 잠재성을 지닌 선행특허들을 추천한다. 본 시스템의 활용성을 검토하기 위해, 미국의 특허 무효화 소송 판정사례들에 대해 제안된 시스템을 적용한 사례연구가 제시된다. 본 시스템의 자동화 프로세스를 통해 선행특허기술 수집 및 내용분석에 소요되는 시간을 획기적으로 단축할 수 있으며, 이에 따라 기술전문가들이 연구개발 전략수립과 같은 본연의 고부가가치 업무에 많은 시간을 투자할 수 있는 긍정적 효과가 기대된다. Barrier patents refer to the existing patents which impede technology-based firms from commercializing their technology item. When barrier patents exist, a firm usually employs some strategic alternatives to cope with these commercialization-hindering patents, such as invalidation analysis, avoidance design and non-infringement logic. The present study focuses on the invalidation analysis among those alternatives and thus aims to developing a patent prior art search system for invalidation analysis of such barrier patents. Building on the patent database constructed using all patent registrations in the United States Trademark and Patent Office, this system recommends potential prior patents for validation analysis of a barrier patent, using three modules: 1) a user interest term definition module to organize invalidation factors about a given barrier patent, 2) an information entropy calculation module to measure the amount of the user interests contained in each patent, and 3) a text similarity measurement module to calculate the relatedness between the barrier patent and any patent. We show the availability of this system by applying the system to real patent invalidation litigations. It is expected that this system will dramatically reduce the time and costs required for patent invalidation analysis and accordingly help researchers more concentrate on their own knowledge intensive tasks such as R&D planning.

국내 의료기기 제조기업의 특허 동향 분석 : 연구개발을 통한 국내등록특허 보유기업 98개 社중심으로

김도성,이정수,조성한,김민석,김남현 경성대학교 산업개발연구소 2018 산업혁신연구 Vol.34 No.3

본 연구에서는 우리나라 의료기기 제조기업 중 실제 연구개발을 통해 특허를 출원·등록한 기 업들을 선별하고, 해당 기업이 보유한 국내등록특허들의 양적/질적 분석을 실시하였다. 양적인 측면 분석을 위하여 해당 기업이 보유한 특허의 출원·등록 현황과 주요 출원인에 대한 분석, 해 당 특허의 대표 IPC(International Patent Classification)별 분석을 실시하였고, 질적인 측면 분석 을 위하여 SMART3(System to Measure, Analyze and Rate patent Technology 3) 분석 결과 를 바탕으로 우수특허 보유 기업 분석 및 대표IPC별 분석을 실시하였다. 분석대상 기업은 KISLINE 서비스를 활용하여 한국표준산업분류(KSIC-9) 중 의료용 기기 제조업(C27) 분야에 해당하는 기업 중 연구개발 정보, 특허정보, 경영성과 정보가 모두 공시된 기업 98개 社를 선별 하였고, 해당 기업이 보유한 국내등록특허 2764건을 분석대상 특허로 설정하였으며, 특허 질적 분석은 권리가 소멸된 등록특허를 제외한 2706건에 대해 실시하였다. 분석결과 국내 의료기기 생산액 중 가장 많은 비중을 차지하는 초음파 의료기기 및 치과 임플란트 분야와 관련된 기업 이 많은 수의 등록특허를 보유하고 있는 것으로 분석되었고, 해당 기업이 보유한 특허도 상대적 으로 우수한 품질을 가지고 있는 것으로 분석되었다. 향후에는 이러한 분야 뿐만 아니라 타 분 야 의료기기에 대한 집중적인 육성도 필요할 것으로 예상된다. In this study, we selected the companies that applied for and registered patents through real R&D among the nation's medical device manufacturers and conducted quantitative and qualitative analyses of the registered patents held by the companies. For quantitative side analysis, the company analyzed the application status and registration status of the patents held by the respective companies, the key applicants, and analyzed the representative IPC of the patents and the results of the SMART analysis for qualitative side analysis. The 98 companies subject to analysis used KISLINE services to R&D information, patent information, and managerial performance information among all companies in the Medical Equipment Manufacturing (C27) field of the Korea Standard Industry Classification (KSIC-9), 2764 domestic registered patents held by the company were analyzed and the qualitative analysis of patents was carried out on 2706 cases, excluding those with expired rights. Analysis shows that companies involved in ultrasound medical devices and implants, which account for the largest share of the nation's medical device production, have a large number of registered patents, it was analyzed that the patents held by the companies were of relatively superior quality. In the future, intensive fostering of these areas as well as other medical devices is expected.

Big Data Smoothing and Outlier Removal for Patent Big Data Analysis

JunHyeog Choi(최준혁),Sunghae Jun(전성해) 한국컴퓨터정보학회 2016 韓國컴퓨터情報學會論文誌 Vol.21 No.8

In general statistical analysis, we need to make a normal assumption. If this assumption is not satisfied, we cannot expect a good result of statistical data analysis. Most of statistical methods processing the outlier and noise also need to the assumption. But the assumption is not satisfied in big data because of its large volume and heterogeneity. So we propose a methodology based on box-plot and data smoothing for controling outlier and noise in big data analysis. The proposed methodology is not dependent upon the normal assumption. In addition, we select patent documents as target domain of big data because patent big data analysis is a important issue in management of technology. We analyze patent documents using big data learning methods for technology analysis. The collected patent data from patent databases on the world are preprocessed and analyzed by text mining and statistics. But the most researches about patent big data analysis did not consider the outlier and noise problem. This problem decreases the accuracy of prediction and increases the variance of parameter estimation. In this paper, we check the existence of the outlier and noise in patent big data. To know whether the outlier is or not in the patent big data, we use box-plot and smoothing visualization. We use the patent documents related to three dimensional printing technology to illustrate how the proposed methodology can be used for finding the existence of noise in the searched patent big data.

한방차 관련 특허 동향 분석

서연호,송주한,권대훈,이상재,Seo, Yeonho,Song, Juhan,Kwon, Daehoon,Lee, Sangjae 대한예방한의학회 2018 대한예방한의학회지 Vol.22 No.1

Objectives : The aim of this study is to analyze the patent trend in Traditional Korean Tea. In this paper, we investigate the diversity of Korean Tea registered as patents focusing on the kinds of medicinal herbs used, processing methods, and types of tea. Methods : We collected patents data using KIPRIS, and WIPS database. We conducted Quantitative Analysis by year and main patentee, and Qualitative Analysis by patent's contents. Results : 313 patents are selected from 2001 to May 2017. The number of patents have been on the rise and reached 44 in 2014. Individuals have registered the most patents, followed by companies, Industry-University Cooperation Foundation, and local autonomous entities. Ginger, Omija and Cactus were used at high frequency among 109 tea containing one medicinal herb. Licorice and Jujube were frequently used among 204 tea containing more than two kinds of medicinal herbs. In classification of patent by efficacy, 'health promotion and function recovery' and 'treatment and prevention of diseases' are most frequently specified. 128 patents are presenting processing methods for medicinal herbs in the order of roasting, fermentation, and fumigation. Among 164 patents presenting tea types, liquid type accounts for more than half and there are 67 leached tea, and 32 powder tea. Conclusion : The analyses results showed that lots of new recipes, efficacy, processing methods and tea types are being studied and registered as patents. In order to help Traditional Korean Tea industry to develop, active patent registration and further study about up-to-date patent trend would be required.

특허정보의 키워드 네트워크 분석

김미애(Mi-Ae Kim),서창교(Chang-Kyo Suh) 한국인터넷전자상거래학회 2017 인터넷전자상거래연구 Vol.17 No.3

The patent has the detailed information about technology development activities. Analyzing the patent information using the social network analysis contains both patent citation relationship analysis and patent keyword network analysis. Although keyword network analysis usually select keywords based on the frequency of words in the patent, frequently used words do not necessarily mean meaningful words in the patent information. In this paper, we demonstrated alternative approach to improve the selection of keyword for the keyword network analysis based on topic model. First, we retrieved 2,724 abstracts of patent on the supply chain management between 1997 and 2016 from USPTO(www.uspto.gov) and EPO(www.epo.org). Second, we used LDA (latent Dirichlet allocation) algorithm to identify the major topics of the supply chain management patents, and extracted 15 keywords to describe these topics. Then, we built the structured data set through the preprocessing to perform the network analysis and visualization operation. Finally, we conducted component analysis and centrality analysis by using LDA-based DocumentTermMatrix(DTM). Component analysis provides three components in the supply chain management patents. For centrality analysis, degree centrality, betweenness centrality, and closeness centrality are identified. Contributions and further research were also discussed as a part of the conclusions.

발명 특허 - 디자인 특허 융합 분석을 통한기술기회 발굴 지원 방법론

송기식,안재형,박상영,이성주,김재훈 대한산업공학회 2023 대한산업공학회지 Vol.49 No.5

Design patents contain as much unique information as utility patent data. The text dat+a of design patents is simpler compared to that of utility patents. The importance of design patent analysis is emerging to consider legal rights for designs. Despite the importance of design patents as a source of technological information, previous studies have rarely researched utility patents and design patents simultaneously. In addition, research using design patents is limited to macroscopic analysis, which results in a lack of systematic research to identify specific product development opportunities. To fill this research gap, this study proposes an approach for identifying new opportunities in product development through the convergence analysis of utility patents and design patents. This approach was applied to a case study involving a future mobility cockpit. In this approach, we initially gather data using the same search formula for both utility patents and design patents. Then, Utility patent-tree and Design patent-tree are derived based on text-mining and clustering analysis. Finally, by analyzing the similarity between the Utility patent-tree and the Design patent-tree, we present technology strategies for areas with high similarities and strategies for areas with low similarities. The results of this study will contribute to the discovery of opportunities in product technology, considering both technical and design concepts.