Analysis methods and main indicators of the hottes

  • Detail

Patent analysis method and main indicators

patent analysis method refers to the method of screening, statistics and analysis of relevant patent documents to convert them into usable information. Patent analysis begins with patent retrieval, which generally includes novelty retrieval, patent legal status retrieval, technology tracking retrieval, infringement defensive retrieval, etc. The search entry can be a technical subject (keyword or classification number), patent applicant (patentee) (country, company, individual), patent application (Publication) time, patent region or combination search, forming a study on whether there is a certain amount of water in a technical water supply system, an important country, an important company, a natural person, and a major competitor's patent status within a certain time and geographical range. Focus on the development of cathode materials, cathode materials, high-performance copper foil and aluminum foil for large capacity and long-life energy storage batteries. Quantitative analysis refers to the statistics of the external characteristics of patent documents (various description items of patent documents) according to certain indicators (such as the number of patents), and the interpretation and analysis of relevant data. Qualitative analysis is an analysis process that takes the content of patents as the object, merges patent documents according to technical characteristics, and makes them orderly. Usually, we need to combine the two to achieve better results

first, the main indicators of quantitative analysis and their applications

there are many quantitative indicators of patent analysis, and different indicators reveal patent information from different angles. Many countries and intellectual property consulting institutions have established their own set of analysis index systems, such as the American Morgan Research & Analysis Association and the American intellectual property consulting company Chi. As countries and regions have different habits of applying for patents, the calculation method of indicators may also be different. As early as the early 1970s, Chi worked with the National Science Foundation to develop the world's first scientific achievement index. The National Science Foundation compiled and published the American science and engineering index, which reported the adoption of Chi's patent index system. The OECD science and technology indicators series manual, namely the patent manual in Frascati series manual, also introduces the concept and calculation method of Chi indicators, and points out that Chi indicators were originally designed for companies, but also applicable to countries and regions. Next, some main indicators of Chi and other commonly used indicators are introduced as follows:

precautions for the use of patent indicators:

(1) refinement, synthesis and time and space analysis of patent indicators. The description of the above patent indicators is still relatively rough, and can be further subdivided. For example, the citation rate can be divided into self citation rate and (external) citation rate. The self citation rate describes the situation of subsequent patents citing their own patents. The higher the citation rate, the stronger the autonomy of research and development. The higher the citation rate, the higher the content of this technology. In addition, various patent indicators are not isolated. Only by combining strategic indicators such as patent index of the same family and technical indicators such as scientific relevance items, and making a comprehensive analysis with time and space as the axis, can we get more comprehensive information

(2) many patent pointers are industry dependent. For example, the technology life cycle will be different due to different technical fields. The TCT of relatively popular technologies is relatively short, such as electronic TCT for about 3 to 4 years, while shipbuilding TCT may be as long as 15 years

(3) patent indicators can measure technological development, but cannot be absolute. Important inventions or innovations, especially those that can bring long-term economic benefits, generally need to apply for patents. However, many technological inventions do not adopt patent protection, but adopt the form of trade secrets or industrialization as soon as possible. Therefore, patents do not correspond to technological inventions or technological innovations one by one

(4) data comparison between different fields should be cautious. The tendency of patent application varies greatly between different technical fields. In some fields, technological innovation activities are very active, but it is not suitable to obtain protection by means of patents. For example, in the field of electronic information, economic returns are often achieved by first mover advantage in the market. Another example is that patents do not include inventions in software, organizational methods, etc., thus excluding technological innovation in the service industry. In this way, when comparing between different fields, we should understand the characteristics of the analyzed industry in order to compare more accurately

(5) obstacles to data comparison caused by differences in patent systems in various countries. Due to the differences of patent systems in various countries, there are differences in the specific examination standards and approval procedures of patents in various countries. For example, multiple interrelated technological inventions can be applied for a patent in the United States, but in Japan, several divisional applications must be filed. Comparing different countries in this way sometimes brings great deviation. If there is a huge gap in the number of patents, it can also explain the problem; However, if it is on the same or similar order of magnitude, it is difficult to draw a definite conclusion. Therefore, when making international comparisons of patents, we generally do not use the data of different patent systems, but choose the same large patent system, such as the United States Patent and Trademark Office or the European Patent Office. But this brings another problem, namely "local advantage"

(6) influence of patent document time lag on analysis data. Despite the implementation of the early disclosure and delayed examination system of patent applications, there is usually an 18 month time lag between the application date and the publication date of many patents, which will directly affect the analysis and prediction of results for those enterprises with frequent technological innovation, especially those industrial enterprises with short technological life cycle

II. Qualitative analysis

qualitative analysis generally classifies or clusters according to the content characteristics of patents, and carries out comparative analysis, combination analysis, correlation analysis, sequence analysis, prediction analysis, etc. in combination with time and space. For example, from the main focus of patent documents, if the content of a technology patent document is mainly based on principles, it indicates that the technology is in an emerging period; If it is mainly applied, it indicates that it is in a mature stage

in recent years, due to the development of information technology and the large number of patent databases, the application of data mining technology in the content of patent documents has become feasible. The research group of technology opportunity analysis (TOA) led by Professor Jay E. PAAP of California Institute of technology has been committed to using text mining methods and special software to carry out time sequence analysis and cluster analysis on the subject words of large databases since the early 1990s, so as to find hot and frontier fields of scientific research and identify research experts and institutions in these fields, This information can be interpreted as an indicator of innovation status and prospects. Professor Porter's toa analysis software includes four steps:

(1) search. That is to find relevant data sources and retrieve electronic data in relevant fields

(2) count. Use software, calculate and arrange results. For example, calculating the number of all high-frequency words, abstracts, authors and their subordinate units can indicate whether a topic is heating up or cooling down, and what kind of distribution of activities related to it in time

(3) contact. The matrix operation provided by toa software is used to help identify connections, cluster, track changes in a period of time, look at emerging new things, and identify who is the most active person in a field

(4) indicates. Convert the results into Excel or other types of spreadsheets and graphic packages, use graphics to show what is happening, and display the search results according to time fragments how the event map evolves over time

III. patent map and its application

patent map refers to the statistical results of primary, secondary, tertiary and other patent documents sorted into various charts, that is, the "mapping" of patent information. As a tool for collecting, sorting and displaying patent information, patent map has strong visibility and is easy to understand. It is generally divided into patent management map, patent technology map and general plastic 500kg sensor

the patent management chart mainly includes: the patent trend chart over the years, the patent proportion chart of various countries, the patent ranking list, the analysis chart of major competitive companies, the average age chart of the company's patents, the comparison chart of the company's invention lineup, the comprehensive analysis table of the company's positioning, the distribution chart of the number of patents of the company, the growth and decline chart of the number of patents of the company, the distribution chart of the number of patents of inventors, etc., which mainly shows the development trend of technology, Analyze the strength and predict the trend of competitive enterprises

patent technology diagrams include IPC (International Classification) analysis diagram, technology life cycle diagram, patent technology efficacy matrix diagram, patent technology distribution diagram, patent technology field accumulation diagram, patent multi view analysis diagram, tunneling technology micrograph, etc. The technology map locks a technology or company for carpet search, which mainly shows the feasibility of technology evolution, diffusion, R & D strategy, avoiding design and digging technology

the patent right map mainly includes the constituent elements of the patent scope, the key points of the patent scope, the patent family map, and the genealogy of important patent citations, which mainly analyzes the R & D space and market space

the production process of patent map is: determine the subject of patent analysis - establish patent retrieval strategies and classification methods - patent document screening and classification - patent document reading and analysis - establish patent management map, technical map, right map - patent map database information analysis - important patent full text analysis - technical innovation, infringement comparison, design avoidance - patent application verification - patent map update


1, OECD, patent manual (1994)

2, Zhang Yanwu, LAN Xiaoyun, one of the methods of enterprise strategy and competition analysis - patent analysis method, information science, August 2003

3, Wei Yanliang, China's science and technology industry: preliminary exploration of enterprise patent information reporting strategy,

4, Cao Lei, patent information analysis and research for patent strategy, research on science and technology management, 2005, issue 3

5 Peng Aidong, a major competitive intelligence - Analysis and research of patent intelligence, "theory and practice of intelligence", March 2000

6, Jin Gang, strategy for intellectual property acquisition,

7, definition of patent indicators,

8, Chi patent pointer - use patents to evaluate the value of companies,

9, Chen Guanhua, patent measurement

10, Zongquan, Bi Yanhong, don't ignore patent indicators in competitive intelligence,

11, Fang Jianguo, Application of text mining in patent analysis, patent literature research, March 2004

12, concept of patent map,

13, Liu Ping, Wu Xinyin, Qi Changwen, application of patent map in enterprise R & D management, research and development management, April 2005 (end)

Copyright © 2011 JIN SHI