1. Commonly used patent information analysis methods
Patent information analysis methods are based on bibliometrics and are carried out with the help of knowledge and related tools from other disciplines. In the past, patent information analysis mainly involved manually extracting a large amount of patent information from patent documents, using relevant statistical methods, combined with industry experience to analyze and process, and explore the intelligence hidden behind patent documents to serve the decision-making of enterprise technological innovation management. Therefore, the main analysis methods at that time included the original text analysis method, simple statistical analysis method, chart method based on simple statistics, dynamic vector method, etc.
1. Original text analysis method.
Understand the development characteristics of competitors' new products and new technologies by searching the patent specifications of competitors' companies, carefully reading and analyzing them, including gap-finding methods, technology improvement methods, technology comprehensive methods and patents Technical principles method.
2. Simple statistical analysis.
Statistical analysis is conducted according to the number of patent inventors, patent applicants, patent classification numbers and patent documents. Through statistical analysis of relevant situations, we can understand the current status of scientific and technological progress in various countries, the transfer of technological research interests or hot spots, and to a certain extent, we can understand the attention of current technology inventors and the development direction of this technology field. We can see Based on the competition in a certain technical field, you can even determine the most active fields.
3. Combination statistical analysis.
By combining the patent classification number, patentee, patent application date (authorized publication date) and patent application country in patent statistics, various statistical information is obtained, and then these statistical information are Perform analysis.
4. Keyword frequency statistics.
① Delete duplicate patent applications, and then extract a number of keywords with technical experimental concepts from the patent rights, abstracts and titles; ② Count the frequency of keywords; ③ It is the reading meaning, for Keywords with a relatively high probability of occurrence are logically combined to re-understand technical concepts.
5. Statistics after technical breakdown.
Carry out technical subdivision and expansion of a certain technology according to the hierarchical tree principle, and conduct statistics on its subordinate concepts item by item.
6. Indicator change chart and technical dynamics and characteristics comparison table.
The technology dynamics and characteristics table is mainly used to reflect the technology dynamics and characteristics of patent applications in different years and by different companies from the perspective of technical fields, certain functions of products, etc., so as to compare the technology development trends and characteristics of various companies. direction. The main forms include: comparison of technological development of enterprises in different years and different technical fields, comparison of different scientific research topics, comparison of different scientific research topics of different enterprises, and regression analysis between various factors.
7. Vector dynamic model method.
In addition to reflecting the quantitative relationship between science and technology, patent documents also imply the direction of scientific and technological development. Therefore, we use the concept of vector to represent it. The application of vector model method is to implement vector modeling of statistical dynamic data, and then evaluate and predict scientific development trends.
8. Patent citation analysis method.
Analyze and study the phenomenon of reference citations in patent documents, reveal their quantitative characteristics and memory rules, and evaluate technological development trends accordingly.
9. Thematic data analysis method.
The so-called "topic data analysis method" is based on the dispersion of patent documents in the International Classification of Inventions, the geographical distribution, research content, etc. of a certain topic document are arranged, combined and analyzed to predict from it The countries with the most active creative and inventive activities in the world and the fields that focus on research.
2. The latest development of patent information analysis methods
With the popularization of computers, the development of information technology and network technology, patent information analysis has gradually transitioned from manual processing to computer-based processing. era. This provides great convenience for patent information analysis, and also promotes the development of patent information analysis methods in the direction of automation, intelligence, and visualization.
1. Quantitative processing of computers.
To conduct a quantitative analysis of the hundreds or thousands of patent documents included in each field, it is necessary to analyze the filing date, main classification number, priority filing country, applicant, and publication number of these patent documents. Waiting for entries to be retrieved, filtered, counted, and graphs drawn. Manual methods are obviously extremely time-consuming and labor-intensive. To this end, it is necessary to create a patent information analysis database, and conduct quantitative analysis of patent documents using computer analysis and statistical patent document description items in the database.
In the patent information analysis database, use the powerful functions of EXCEL spreadsheets to sort each description item, filter out the patent application numbers of each year, country, and company, and then compare the above By conducting quantitative statistics on each item, statistical charts can be made in a targeted manner to understand the patent development in this field from all aspects. Specifically, it mainly includes: ① The chronological distribution of the number of patent applications can measure the future development trend of a certain technical field; ② The classification of technical topics of patent documents, which can classify and count technical topics and make a technical topic distribution map; ③ By counting the number of domestic patent applications in each country based on the country of priority, we can understand the technical strength of each country in this field; ④ By counting the number of patents owned by each company, we can understand the technical and economic strength of each company in this technical field; ⑤ Each country For the company's foreign patent applications, count the number of patent applications from more than two countries, and draw a distribution map of the number of foreign patent applications of each company; ⑥The distribution of the number of patent applications in each country, and find out the competitors faced in different countries , you can understand the degree of market competition of foreign companies in these countries; ⑦ Statistical analysis of patent applications in each country, you can understand the competitive situation of companies in each country.
2. Patent information analysis method based on similarity function.
The similarity functions described here include: similarity functions based on patent citations and similarity functions based on term concurrency. Similar functions based on the concurrency of terms in the literature are also valid. Reliable data sources provide consistent index terms for each document, and these are used to build similarity functions based on term concurrency.
This function is easy to execute using SQL query statements. After calculating the similarity values ??between documents, force direct placement or self-organizing neural network technology is used to map the documents into a two-dimensional space to form document clusters to explore the relationship between documents. The force-directed configuration method is that at the beginning, all documents are placed at the center of the plane. The force between the documents is inversely proportional to their distance and proportional to their similarity value; if the similarity value is greater than 0, it is attraction. force, otherwise it is a repulsive force; the direction of the force obeys Coulomb's law in mechanics. This process is repeated until a stable literature map is formed. Self-organizing neural network technology uses behavioral input variables of similar matrices to train a rectangular self-organizing neural network. Typically a self-organizing neuron matrix is ??used. After training is completed, the density distribution of neuron connection weights in N-dimensional space matches the density distribution of the training vector.
In general, patent analysis is divided into two types: quantitative analysis and qualitative analysis.
1. Quantitative analysis
Quantitative analysis mainly conducts statistical analysis through the appearance characteristics of patent documents, that is, identifying relevant documents through the indexing items inherent in patent documents, and then Make statistics on relevant indicators, and finally use different methods to explain changes in relevant data to obtain information on dynamic development trends.
(1) Statistical objects and angles
①The statistical objects are generally based on the number of patents. ② Statistics can be conducted from different angles according to patent classification, patentee, year, country, etc. When statistics of patent information are made by category, based on the number of patents in each field, we can know which scientific and technological fields have the most active inventions, which technologies will make breakthroughs, and which technologies will soon be eliminated. If patent information is collected by country, we can discover the scientific and technological development strategies of the countries being counted and their status in various fields. This statistical result helps people understand the focus of scientific research and development in various countries in a certain period.
If you count patents by patentee, you can find important technology owners in a certain field, or which company has an important position in that field. (2) Main indicators of statistics
①Number of patents. The number of patents in a certain technology category can be used to measure the level of technical activities in this technical field; and the number of patents applied by a certain company or a patentee over the years reflects the occurrence, development process and development trends of its technical activities. The number of patents can be used to compare the output of technological activities and the intention to seek industrial property rights protection in different countries in different periods and in different fields. ② Number of patent families. The number of patent families for a certain invention reflects the breadth of the company’s patent applications, as well as the potential value of the invention - applying for a patent abroad is much more expensive than in your own country due to the cost of translation and specialized legal help. , only those inventions recognized by companies as having the most commercial value will apply for patents in multiple countries in order to protect future investments and exclusive rights to product output. ③The number of patent citations. The number of times a certain patent is cited by subsequent patents can reflect the importance of the patent, because after the emergence of an important patent, a large number of improvement patents will appear, and this important patent will be repeatedly cited by the improvement patents. Citations reveal the connections between patents and can be used to track patent networks corresponding to different technologies and discover patents at the intersection of different technologies. It is a pity that my country's patent database cannot provide the number of citations of patents, which is a shortcoming. ④Patent growth rate. The patent growth rate measures the percentage change in the number of patents over time, which can show whether technological innovation is increasing or slowing down over time. For example, the quarterly patent growth rate compares the number of patents obtained by a company in a certain quarter with the number of patents obtained in the previous quarter, and calculates the percentage increase or decrease in patents obtained in that quarter compared with the previous quarter. The annual patent growth rate measures the percentage change in patent growth compared with the previous year, and is used to measure changes in the development of technological activities over the past year. ⑤Scientific relevance. Scientific relevance is measured by the number of scientific academic papers or research reports cited by patents (subject patents). This indicator measures the relationship between patented technology and cutting-edge scientific research. The value of scientific relevance is industry-dependent: the average value of scientific relevance in the machinery industry is close to zero, while the high-tech biochemical industry may be as high as 15. ⑥ Technology life cycle. The technology life cycle is measured by the average age of the patented technology cited in the title page of the company's patent application documents. Therefore, the technology life cycle can be understood like this: it is the period of time between the latest patent and the earliest patent. If the technology life cycle is short, it means that efforts are being made to develop a relatively new technology, and this technology is developing and innovating very quickly. The technology life cycle is industry-dependent. The technology life cycle of relatively popular industries is shorter, such as electronics, which is about 3 to 4 years, pharmaceuticals, which are about 8 to 9 years, and shipbuilding, which may be as long as 15 years. ⑦Patent efficiency. Patent efficiency measures the patent quantity output created by a certain R&D expenditure. This indicator is used to evaluate the scientific research capabilities and cost efficiency of an enterprise's patent quantity output within a predetermined time. The more patents are produced and the higher the patent efficiency, the stronger the company's technology research and development capabilities. ⑧ Patent implementation rate. Whether the patent implementation rate can be beneficially implemented and whether it can bring technological innovation is still unknown for those patented technologies that have not yet been implemented. The implementation of general invention patents must go through a development process, and development is not always successful. There are many invention patent technologies that cannot be solved due to technical difficulties during the development process or cannot achieve the expected results under existing technical conditions. Don’t give up halfway or eventually give up. The implementation of invention patents can be measured from multiple perspectives such as technical performance, economic benefits, social benefits, market factors, industrial development and production capabilities, macro environment, and industrialization risks. The higher the patent implementation rate, the greater the contribution that patents make to technological development and technological innovation, and the closer they are integrated with technological development. my country's patent implementation rate only remains at around 30, which is far lower than the levels of European, American and Japanese countries. ⑨Industrial standardization indicators.
In cross-industry horizontal comparisons, differences between industries bring trouble to the comparison of patent indicator values ??between different industries. For this reason, industry standardized indicators need to be used. The value of the industry standardized indicator is obtained by dividing the indicator value of an enterprise by the average value of the indicator in the industry in which the enterprise is located. For example, there are 30 companies in the chemical industry, and their average scientific relevance value is 3.7. Then the standard indicator value of each chemical company's scientific relevance should be obtained by dividing the scientific relevance index value of each company by 3.7. In this way, the different impacts of different industries can be eliminated and the best-performing companies in each industry can be found.
(3) Main contents of statistics
① Study on the distribution of patented technologies over time. That is, taking time as the horizontal axis and the number of patent applications (or approvals) as the vertical axis, statistics on the changes in the number of patents over time are generally used for trend prediction. Any technology has a process of emergence, development, maturity and aging. Changes in the number of patents applied for and the number of patent citations over the years can determine the development trend and active period of the technology, providing a basis for major decisions such as scientific research projects and technology development. A comparative study of the time distribution of patents in different technical fields can determine which technical fields are more active and which are stagnant during a certain period. ② Research on the spatial distribution of patented technologies. That is, by comparing the number of patents between different companies and enterprises to reflect their technical level and strength. Spatial distribution is generally used to identify competitors, analyze their technical strategies, etc. By counting the patent applications of a certain technology category by patentee, we can get the distribution of a certain technology among different companies or enterprises, and understand which companies or enterprises have invested more in this field, have more active patent activities, and have higher technological levels. ; By counting the patent frequencies of companies in different technology categories, we can understand the most active areas of each company, that is, the key areas of development. In addition, by searching for patent family members of a certain patent, we can obtain the geographical distribution of these patent applications, thereby judging their commercial value and understanding the key areas of a company's technology output; it can also provide a basis for technology introduction and avoid other parties' product exports. Protected areas provide intelligence.
(4) Statistical parameters that measure different stages of technological development.
Technological growth rate V:. Among them, a: the number of invention patent applications (or the number of approvals) in the current year; A: the cumulative number of invention patent applications (or the cumulative number of approvals) dating back 5 years. Calculated continuously for several years, the V value increases, indicating that the technology is in the budding or growth stage. Technology maturity coefficient α:. Among them, a is the same as above, and b is the number of utility model patent applications (or approvals) in that year. Calculated continuously for several years, the α value decreases, reflecting the increasing maturity of the technology. Technology aging coefficient β:. Among them, a and b are the same as above, and c is the number of design patent or trademark applications (or approval number) in that year. Calculated continuously for several years, the β value increases, indicating that the technology is becoming obsolete. New technology characteristic coefficient N: N = υ2 α2. Among them, υ is the technological growth rate and α is the technological maturity coefficient. It is a comprehensive indicator that reflects the emergence or aging of a certain technology. The larger the N value, the stronger the characteristics of the new technology, which indicates that it has more development potential. 2. Qualitative analysis
Qualitative analysis, also called technical analysis, identifies patents based on their technical content or the “quality” of the patent, and organizes the relevant patents according to their technical characteristics to make them orderly. This is very different from statistical analysis that only relies on the appearance characteristics of patent documents. Qualitative analysis is generally used to obtain information on technology trends, corporate trends, specific rights status, etc. The content of important patents can be considered from the five aspects of invention use, principle, material, structure and method, and important patents can be classified according to similarities and differences in content. If the patent content is mainly about principles, it means that the technology is not yet mature; if the patent content is mainly about the diversity of uses, it means that the technology is already practical.
In addition, by analyzing the patents of major companies in a certain technical field according to the patent content list, we can see the technical characteristics and development focus of each company; dividing the relevant patents into various patent groups according to the similarities and differences in technical content, and analyzing the different patent groups owned by a certain company. Or by analyzing the changes in patent groups in different periods, we can analyze and predict key issues in the development process of a certain technology or product, future development trends and application trends, and relationships with other technologies.
Due to the specific content of technology involved, the work of qualitative analysis is relatively heavy and complicated. Whether to use quantitative analysis or qualitative analysis should depend on the problem you hope to solve and the patent data you have. In fact, it is often necessary to combine qualitative analysis with quantitative analysis to achieve good results. For example, you can first use quantitative analysis to determine which companies have technological advantages in a certain technical field (the number of patent applications or approvals can reflect the level of technical activity), and identify important patents in this technical field (the number of citations of a certain patent by subsequent patents). reflect the importance of patents), and then conduct a qualitative analysis of the important patents of these companies.
Quantitative analysis and qualitative analysis of patent information reflect the development status and development trend of technology through quantitative changes and internal qualitative changes. There are differences between the two, but there is also an inevitable connection. The classification of quantity needs to be based on quality, and quality must be reflected through quantity. Therefore, in actual work, using the two together will achieve better results.