Cross-industry co-occurrence association identification method and device based on patent data, computer equipment, storage medium and program product

Abstract

The invention provides a cross-industry co-occurrence association identification method and device based on patent data, computer equipment, a storage medium and a program product, and the method comprises the steps: obtaining target domain patent data, screening the target domain patent data with at least two or more different IPC subclass numbers, constructing a co-occurrence data set; taking each IPC subclass number in the co-occurrence data set as a network node, and calculating a co-occurrence intensity value of any two network nodes to form an IPC co-occurrence matrix; and constructing an industrial co-occurrence matrix based on the IPC subclass number and industrial code mapping table and the IPC co-occurrence matrix.

Description

Technical Field

[0001] The present invention relates to the field of data mining technology, and in particular to a method, device, computer equipment, storage medium and program product for identifying cross-industry co-occurrence associations based on patent data. Background Art

[0002] With the rapid development of technological innovation, patent data has become an important basis for measuring the progress of a technological field and its mutual influence with other fields. In patent analysis, the International Patent Classification (IPC) number, as a standard tool for classifying technological fields, can help us refine the classification of patent technologies and accurately identify the field to which the technology belongs. As a traditional and modern industry, the tobacco industry has attracted worldwide attention for its technological progress.

[0003] With the technological progress and changes in consumption habits in the global tobacco industry, the focus of research and development in the tobacco field has gradually shifted from traditional planting and processing technologies to more diversified and high-tech innovations. Especially in the context of increasing health awareness and increasingly stringent regulations, tobacco technology is merging into multiple emerging fields such as green environmental protection, intelligent manufacturing, and biotechnology. This trend is not only reflected in research and development, but also in the technological innovation activities of patents in the tobacco field. By digging out the connections and integrations between technical fields from patents in the tobacco field, identifying industries with strong technological innovation capabilities, and focusing on the advanced technologies of these industries, it is crucial to promote the tobacco industry to take greater steps in diversification, innovation, and sustainable development.

[0004] In order to solve the above problems, people have been seeking an ideal technical solution. Summary of the invention

[0005] Based on this, it is necessary to provide a cross-industry industry co-occurrence association identification method, device, computer equipment, storage medium and program product based on patent data to address the above technical problems.

[0006] In order to achieve the above-mentioned object, the first aspect of the present invention provides a method for identifying cross-industry industry co-occurrence associations based on patent data, comprising the following steps: Obtain patent data in the target field, screen patent data in the target field with at least two different IPC subclass numbers, and construct a co-occurrence dataset; Take each IPC subclass number in the co-occurrence data set as a network node, calculate the co-occurrence strength value of any two network nodes, form an IPC co-occurrence matrix, and draw an IPC co-occurrence undirected network graph based on the IPC co-occurrence matrix; the co-occurrence strength value is the number of co-occurrences of two different IPC subclass numbers in the same patent data; An industry co-occurrence matrix is ​​constructed based on the IPC subclass number and industry code mapping table and the IPC co-occurrence matrix, and an industry co-occurrence undirected network diagram is drawn based on the industry co-occurrence matrix.

[0007] In order to achieve the above-mentioned object, the second aspect of the present invention provides a method for identifying cross-industry industry associations based on patent data, comprising the following steps: In the patent data of the target field, select patent data with at least two different IPC subclass numbers to construct a co-occurrence dataset; Take each IPC subclass number in the co-occurrence data set as a network node, calculate the co-occurrence strength value of any two network nodes, and form an IPC co-occurrence matrix; the co-occurrence strength value is the number of co-occurrences of two different IPC subclass numbers in the same patent data; The industry co-occurrence matrix is ​​constructed based on the IPC subclass number and industry code mapping table and the IPC co-occurrence matrix.

[0008] Take each IPC subclass number in the co-occurrence data set as a network node, calculate the co-occurrence strength value of any two network nodes, and form an IPC co-occurrence matrix; the co-occurrence strength value is the number of co-occurrences of two different IPC subclass numbers in the same patent data; Construct an industry co-occurrence matrix based on the IPC sub-category number and industry code mapping table and the IPC co-occurrence matrix; At the same time, patent data outside the industry is obtained based on the patent mutual citation relationship in the patent data of the target field; Construct an IPC citation matrix of the first category of non-industry patent data and target field patent data based on IPC main and sub-category numbers; Based on the IPC citation matrix, identify the key technology areas outside the industry that have made great contributions to the target field, and construct a network diagram of the first key technology area based on the IPC main and sub-class numbers. The network diagram of the first key technology area is a directed graph, with the direction from the cited patent to the citing patent; quantitatively analyze the node attributes in the network diagram of the first key technology area to obtain the contribution value of the key technology areas outside the industry to each sub-technical area in the target field; Based on the IPC main and sub-category numbers and industrial sector association mapping table and the IPC citation matrix, an industry association matrix of cited patents based on the technical mutual citation relationship between industries is constructed. Based on the cited patent industry association matrix, an industry citation association network is drawn, and the industry citation association network is quantitatively analyzed to obtain the industry citation relationship.

[0009] In an embodiment of the second aspect of the present invention, the cross-industry industry association identification method based on patent data further includes the following steps: Obtain patent data in the target field, and obtain the second category of non-industry patent data based on patent citation relationships; Based on the IPC main and sub-class numbers, an IPC citation matrix of the second category of patent data outside the industry and patent data in the target field is constructed; Based on the IPC citation matrix, identify the key technology areas outside the industry that have made great contributions to the target field, and construct a network diagram of the second key technology area based on the IPC main and sub-class numbers. The network diagram of the second key technology area is a directed graph, with the direction from cited patents to citing patents; quantitatively analyze the node attributes in the network diagram of the second key technology area to obtain the contribution value of each sub-technology area in the target field to the key technology areas outside the industry; Based on the IPC main and sub-category numbers and industrial sector association mapping table and the IPC citation matrix, an industry association matrix of cited patents based on technology citation relationships between industries is constructed. Based on the cited patent industry association matrix, an industry citation directed network diagram is drawn, and the industry citation directed network diagram is quantitatively analyzed to obtain the industry citation relationship.

[0010] In order to achieve the above-mentioned purpose, the third aspect of the present invention provides a cross-industry industry co-occurrence association identification device based on patent data, comprising: The co-occurrence patent dataset construction module is used to obtain patent data in the target field, screen patent data in the target field with at least two different IPC subclass numbers, and construct a co-occurrence dataset; The technology co-occurrence analysis module is used to treat each IPC sub-class number in the co-occurrence data set as a network node, calculate the co-occurrence intensity value of any two network nodes, and form an IPC co-occurrence matrix; Industry co-occurrence analysis module, used to construct industry co-occurrence matrix based on IPC classification number and industry code mapping table and IPC co-occurrence matrix; The first category non-industry patent acquisition module is used to obtain the first category non-industry patent data based on the acquired target field patent data and patent citation relationships; IPC citation matrix construction module, which constructs the IPC citation matrix of the first category of non-industry patent data and target field patent data based on the IPC main and sub-category numbers; The first non-industry key technology field acquisition module is used to identify non-industry key technology fields that contribute greatly to the target field based on the IPC citation matrix; The first technology contribution analysis module is used to construct a network diagram of the first key technology field based on the IPC main and sub-class numbers, where the network diagram of the first key technology field is a directed graph, with the direction from the cited patent to the citing patent; quantitative analysis of the node attributes in the network diagram of the first key technology field is performed to obtain the contribution value of the key technology field outside the industry to each sub-technology field in the target field; The industry citation relationship analysis module is used to construct an industry association matrix of cited patents based on the technical mutual citation relationship between industries based on the IPC main and sub-category numbers and industrial sector association mapping table and the IPC citation matrix, draw an industry citation association network based on the cited patent industry association matrix, and conduct quantitative analysis on the industry citation association network to obtain industry citation relationships.

[0011] In an embodiment of the second aspect of the present invention, the cross-industry industry association identification device based on patent data further includes: The second category non-industry patent acquisition module is used to obtain the second category non-industry patent data based on the acquired target field patent data and patent citation relationship; IPC citation matrix construction module, which constructs the IPC citation matrix of the second category of non-industry patent data and target field patent data based on IPC main and sub-category numbers; The second non-industry key technology field acquisition module is used to identify non-industry key technology fields that have made great contributions to the target field based on the IPC citation matrix; The second technology contribution analysis module is used to construct a network diagram of the second key technology field based on the IPC main and sub-class numbers, where the network diagram of the second key technology field is a directed graph, with the direction from the cited patent to the citing patent; quantitative analysis is performed on the node attributes in the network diagram of the second key technology field to obtain the contribution value of each sub-technical field in the target field to the key technology field outside the industry; The industry citation relationship analysis module builds an industry association matrix of cited patents based on the technology citation relationship between industries based on the IPC major and sub-category numbers and industrial sector association mapping table and the IPC citation matrix. It draws an industry citation directed network diagram based on the cited patent industry association matrix, and conducts quantitative analysis on the industry citation directed network diagram to obtain the industry citation relationship.

[0012] In order to achieve the above-mentioned purpose, the third aspect of the present invention provides a computer device, including a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory communicate with each other through the communication bus; the memory is used to store computer programs; the processor is used to implement the cross-industry industry co-occurrence association identification method based on patent data as described in the first aspect or the second aspect when executing the program stored in the memory.

[0013] In order to achieve the above-mentioned purpose, the fourth aspect of the present invention provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the cross-industry industry co-occurrence association identification method based on patent data as described in the first aspect or the second aspect is implemented.

[0014] In order to achieve the above-mentioned purpose, the fifth aspect of the present invention provides a computer program product, including a computer program, which, when executed by a processor, implements the cross-industry industry co-occurrence association identification method based on patent data as described in the first aspect or the second aspect.

[0015] The beneficial effects of the present invention are as follows: the present invention can explore the intersection and penetration between different technical fields from the perspective of patents through IPC co-occurrence analysis, especially in modern industries with increasingly complex and interdependent technologies. By analyzing the co-occurrence data between IPC numbers and mapping them to industry codes, we can further explore the degree of technical interaction and integration between industries, making industrial technology integration an important means to promote industrial transformation and upgrading and improve competitiveness, and provide strong support for studying industrial upgrading and innovation paths.

[0016] Furthermore, by analyzing the patents in the target field and their mutual citations with patents in other fields, we can understand the dynamics of technological development in the tobacco field and its interaction with other related technical fields, effectively demonstrate the frequency and degree of correlation of multiple technical fields in patent technology, and reveal the degree of collaborative innovation between different fields. Then, we can expand the perspective to the industrial level and demonstrate how innovations across technical fields give birth to new sub-industries at the industrial level, forming an extension and reconstruction of the upstream and downstream of the industrial chain.

[0017] The IPC subcategories involved in the patents in the target field are defined as related technical fields, and based on this, all IPC subcategories that have a citation or cited relationship with the patents in the target field are counted. On this basis, in order to clarify the technical interaction relationship between the target field and the non-target field, the IPC subcategories in the related technical field are eliminated, thereby screening out the patent IPC number subcategories in the non-target field. This process not only effectively distinguishes the target field from the non-target field, but also provides a more efficient way to analyze the mutual citation relationship between the two. Through the above method, the correlation between the target industry and other industries in the technical field and industrial level can be accurately extracted. This analysis method breaks through the limitations of traditional classification based on major categories or specific industries, and not only shows the impact path of the target industry on the technological development of other industries, but also reveals the role of other industry technologies in promoting the development of the target industry. In addition, it also provides an innovative tool and new perspective for in-depth research on technological integration and interaction between industries. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 It is a flowchart of a method for identifying cross-industry industry co-occurrence associations based on patent data in Embodiment 1 of the present invention; Figure 2 is a schematic diagram of an IPC co-occurrence graph in Example 1 of the present invention; Figure 3 is a schematic diagram of an industry co-occurrence map in Example 1 of the present invention; Figure 4 It is a flowchart of a method for identifying cross-industry industry associations based on patent data in Embodiment 2 of the present invention; Figure 5 It is a schematic diagram of a network diagram in the first key technical field in Example 2 of the present invention; Figure 6 is a schematic diagram of an industry citation association network in Example 2 of the present invention; Figure 7 is a flowchart of a method for identifying cross-industry industry associations based on patent data in Embodiment 3 of the present invention; Figure 8 It is a schematic diagram of a network diagram in the second key technical field in Example 3 of the present invention; Fig. 9 is a schematic diagram of an industry citation association network in Example 3 of the present invention; Fig.10 It is a schematic diagram of the structure of the computer device of the present invention. DETAILED DESCRIPTION

[0019] As a global patent classification standard, the International Patent Classification (IPC) provides a unified framework for studying patents in different technical fields. By analyzing the co-occurrence of IPC classifications of patents, the association and integration between technical fields can be clearly revealed. In particular, applying social network analysis methods to patent data can intuitively display the degree of interdependence and integration between technical fields by calculating the co-occurrence intensity of different technical fields.

[0020] The technical solution of the present invention is further described in detail below through specific implementation methods.

[0021] Example 1 This embodiment provides a cross-industry industry co-occurrence association identification method based on patent data, such as Figure 1 As shown, the following steps are included: Step S1, obtain patent data in the target field, screen patent data in the target field with at least two different IPC subclass numbers, and construct a co-occurrence dataset.

[0022] Specifically, a complete IPC classification number is composed of a combination of symbols representing departments, major categories, minor categories, large groups or small groups. A patent may be composed of multiple IPC classification numbers. By extracting the first 4 digits of different classification numbers, the different technical fields involved in the patent can be obtained, and then a patent-based technology co-occurrence network can be constructed.

[0023] This embodiment takes the tobacco field as an example to provide specific steps of the cross-industry industry co-occurrence association identification method: (1) Obtain patent data in the target field, and screen patent data with at least two IPC classification numbers to construct a primary patent data set; in this example, 30,958 patents with more than one IPC classification number were screened from the original tobacco patents from 2019 to 2023.

[0024] (2) In the primary patent dataset, patent data with at least two different IPC subclass numbers are screened to construct a co-occurrence dataset. In this example, the first 4 digits of the IPC classification number corresponding to each patent are extracted, and patents with more than 1 IPC classification number (4 digits) are retained, totaling 14,281, to form a co-occurrence dataset.

[0025] Step S2: Take each IPC sub-category number in the co-occurrence data set as a network node, calculate the co-occurrence strength value of any two network nodes, and form an IPC co-occurrence matrix.

[0026] Specifically, the number of times that IPC subclass A and IPC subclass B co-appear in the same patent is used as an indicator to measure the co-occurrence intensity, and then a co-occurrence matrix between technical fields is formed based on the calculated co-occurrence intensity between different IPC subclasses, where A and B are different.

[0027] Furthermore, after the processing of step S2, the co-occurrence intensity statistics between the IPC pairs in the co-occurrence dataset of 14,281 patent data are shown in Table 1.

[0028] Table 1 Descriptive statistics of IPC co-occurrence times Number of IPC pairs Mean Standard Deviation Minimum Lower quartile Upper quartile Maximum 3893 6.27 24.93 1 1 4 687 It can be seen that there are 3893 different IPC pairs, and the average number of co-occurrences of IPC pairs is 6.27, indicating that the average number of co-occurrences of each pair of IPC subclasses is small, and the number of co-occurrences between the vast majority of IPC pairs is relatively low, reflecting that the correlation between most technical fields in tobacco patents is weak or scattered. The maximum number of co-occurrences is as high as 687 times, and the minimum is only 1. The upper quartile value of 4 means that after sorting from large to small in terms of co-occurrences, the top 25% of the co-occurrences are greater than 4, and the lower quartile value of 1 means that the co-occurrences of the bottom 25% are less than 1.

[0029] In order to explore the overall co-occurrence of the technical field, an IPC co-occurrence matrix was constructed based on the IPC co-occurrence times. Since the co-occurrence relationship does not have directionality, the matrix is ​​a symmetric matrix. It means node and nodes The number of co-occurrences between .

[0030] Furthermore, in order to intuitively perceive the co-occurrence of IPC sub-category numbers, after obtaining the IPC co-occurrence matrix, an IPC co-occurrence undirected network graph is drawn based on the IPC co-occurrence matrix.

[0031] In the IPC co-occurrence undirected network graph, each IPC subclass is a node. The size of the node is related to the number of nodes connected to it, that is, the degree of the node. The larger the degree, the larger the node. The thickness of the line is related to the number of co-occurrences between nodes. The more co-occurrences between nodes, the thicker the line connecting the two nodes. Considering that there are too many nodes in the technical field involved in the co-occurrence network, the more active technical fields in the network are locked in by calculating the network degree.

[0032] The network degree (Degree) is an indicator that measures the number of connections a node, or a technology field, has in the network. The higher the degree, the more connections the node has with other nodes, and the stronger the technology mobility. The standardized network degree (NrmDegree) is to normalize the degree so that it can be compared on the same scale. This indicator helps to evaluate the relative position and technical influence of different technology fields in the network.

[0033] According to statistics, the IPC co-occurrence undirected network diagram in this embodiment is as follows: Figure 2 As shown in the figure, the average network degree is 116.13, that is, each node in each technical field is associated with 116 other technical fields on average, and the standard deviation of the network degree is 323.962, showing that the network degrees of different technical fields vary greatly, indicating that the degree of dependence and integration depth of different technical fields on tobacco technology are different. Among them, the maximum degree value is 3960, and the corresponding IOPC subclass number is A24B (manufacture or preparation of smoking or chewing tobacco; tobacco; snuff), and the minimum degree value is only 1. It should be noted that nodes with small degree values ​​do not mean that they are unimportant, but may be highly specialized or independent in certain specific technical fields.

[0034] Step S3, constructing an industry co-occurrence matrix based on the IPC sub-category number and industry code mapping table and the IPC co-occurrence matrix.

[0035] IPC co-occurrence analysis can explore the intersection and penetration between different technical fields from the perspective of patents, especially in modern industries with increasingly complex and interdependent technologies. By analyzing the co-occurrence data between IPC numbers and mapping them to industry codes, we can further explore the degree of technical interaction and integration between industries, making industrial technology integration an important means to promote industrial transformation and upgrading and improve competitiveness, and provide strong support for the study of industrial upgrading and innovation paths.

[0036] Table 2 shows the descriptive statistics based on the number of co-occurrences of industry pairs. According to statistics, there are 421 industry pairs composed of different industries, showing the diversity of cross-industry technology intersections, reflecting the universality of technology integration in modern industries. The average number of co-occurrences is 43.53, that is, there are an average of 43.53 technology co-occurrences between each industry pair. In general, the technology interactions between different industries are relatively frequent. The standard deviation is 141.26, which is much higher than the mean, indicating that there are significant differences in the number of technology co-occurrences between industry pairs. The minimum number of co-occurrences is only 1, and the maximum number is as high as 1433. Through the distribution of quartiles, it can be seen that the lower quartile is 2, indicating that the number of technology co-occurrences between one-quarter of the industry pairs is small, less than or equal to 2 times, and the upper quartile is 25, indicating that the number of co-occurrences between one-quarter of the industry pairs is higher than 25 times.

[0037] Table 2 Descriptive statistics of co-occurrence times of industry pairs Number of industry pairs Mean Standard Deviation Minimum Lower quartile Upper quartile Maximum 421 43.53 141.26 1 2 25 1433 Similarly, after constructing the industry co-occurrence matrix, an industry co-occurrence undirected network graph is drawn based on the industry co-occurrence matrix. The industry co-occurrence undirected network graph is as follows: Figure 3 As shown in the figure. Since there is no technology flow direction in co-occurrence, the industry co-occurrence undirected network diagram is an undirected graph. Nodes with co-occurrence relationships are connected by straight lines. The thickness of the line is related to the number of co-occurrences. The more co-occurrences between industries, the thicker the connecting straight line. The size of the node is related to the degree of the node, that is, the more lines connected to the node, the larger the node.

[0038] Furthermore, in order to understand the relative position and influence of each industry in the industry co-occurrence undirected network diagram, the degree value of the industry co-occurrence undirected network diagram was quantitatively analyzed. According to statistics, the average degree value of the network is 872.619, indicating that each industry node is connected to 873 other industry nodes on average in the network. This high degree value indicates that the technical correlation between industries is strong, forming a relatively close technical cooperation network. The mean value of the relative network degree value is 1.450, indicating that each industry plays a certain role in technical cooperation, and there is no situation dominated by a single or a few industries. The standard deviation of the degree value is as high as 1285.884, indicating that the degree of technical connection between different industries varies significantly. Some industries have frequent technical interactions with other industries, forming a strong network, while some industries are less connected, showing the unevenness of the network. The standard deviation of the relative network degree value is 2.137, which also reflects the differences in participation and influence of different industries in technical interactions.

[0039] It should be understood that, although the steps in the flowcharts involved in the above embodiments are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps is not strictly limited in order, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above embodiments may include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily carried out in sequence, but can be executed in turn or alternately with other steps or at least a part of the steps or stages in other steps.

[0040] Example 2 This embodiment provides a method for identifying cross-industry industry associations based on patent data. Figure 4 As shown, the following steps are included: In the patent data of the target field, select patent data with at least two different IPC subclass numbers to construct a co-occurrence dataset; Take each IPC subclass number in the co-occurrence data set as a network node, calculate the co-occurrence strength value of any two network nodes, and form an IPC co-occurrence matrix; the co-occurrence strength value is the number of co-occurrences of two different IPC subclass numbers in the same patent data; The industry co-occurrence matrix is ​​constructed based on the IPC subclass number and industry code mapping table and the IPC co-occurrence matrix.

[0041] Take each IPC subclass number in the co-occurrence data set as a network node, calculate the co-occurrence strength value of any two network nodes, and form an IPC co-occurrence matrix; the co-occurrence strength value is the number of co-occurrences of two different IPC subclass numbers in the same patent data; Construct an industry co-occurrence matrix based on the IPC sub-category number and industry code mapping table and the IPC co-occurrence matrix; At the same time, patent data outside the industry is obtained based on the patent mutual citation relationship in the patent data of the target field; Construct an IPC citation matrix of the first category of non-industry patent data and target field patent data based on IPC main and sub-category numbers; Based on the IPC citation matrix, identify the key technology areas outside the industry that have made great contributions to the target field, and construct a network diagram of the first key technology area based on the IPC main and sub-class numbers. The network diagram of the first key technology area is a directed graph, with the direction from the cited patent to the citing patent; quantitatively analyze the node attributes in the network diagram of the first key technology area to obtain the contribution value of the key technology areas outside the industry to each sub-technical area in the target field; Based on the IPC main and sub-category numbers and industrial sector association mapping table and the IPC citation matrix, an industry association matrix of cited patents based on the technical mutual citation relationship between industries is constructed. Based on the cited patent industry association matrix, an industry citation association network is drawn, and the industry citation association network is quantitatively analyzed to obtain the industry citation relationship.

[0042] Specifically, the steps for constructing the IPC co-occurrence matrix and the industry co-occurrence matrix refer to Example 1, which will not be repeated here.

[0043] The steps for constructing the IPC citation matrix and industry citation network are as follows: Step S21, obtaining patent data in the target field, and obtaining the first category of non-industry patent data based on patent citation relationships.

[0044] Step S22, constructing an IPC citation matrix of the first category of non-industry patent data and the target field patent data based on the IPC main classification number.

[0045] Specifically, the specific steps of step S2 include: Define the IPC main classification codes involved in the patent data in the target field as relevant technical fields; Count all IPC main classification numbers in the first category of non-industry patent data that have citation relationships with patents in the target field; Screen all IPC main classification codes that do not belong to the relevant technical field as IPC main classification codes for non-target fields; Based on the IPC main classification number of the non-target field obtained in the previous step, an IPC citation matrix of the first category of non-industry patent data and the target field patent data is constructed.

[0046] Specifically, the number of times a patent is cited reflects the importance of the patent to a certain extent. If a patent is cited multiple times, it means that the patent has great reference significance and the technology involved in the patent can be absorbed and utilized by other patents. By counting the number of occurrences of IPC classification numbers in cited patents and summarizing and calculating the corresponding IPC classification number frequencies, the importance of different technical fields can be reflected.

[0047] Step S23, based on the IPC citation matrix, identify the key technology fields outside the industry that have made great contributions to the target field, and construct a network diagram of the first key technology field based on the IPC main classification number, wherein the network diagram of the first key technology field is a directed graph, and the direction is from the cited patent to the citing patent; quantitatively analyze the node attributes in the network diagram of the first key technology field to obtain the contribution value of the key technology fields outside the industry to each sub-technology field in the tobacco field.

[0048] Step S24, based on the IPC main classification number and industry sector association mapping table and the IPC citation matrix, construct a cited patent industry association matrix based on the technical mutual citation relationship between industries, draw an industry citation association network based on the cited patent industry association matrix, and conduct quantitative analysis on the industry citation association network to obtain the industry citation relationship.

[0049] Specifically, taking the tobacco industry as an example, the tobacco field patent data with citation relationships in the tobacco field patent data downloaded in this embodiment involves a total of 329 IPC main classification numbers, and the patents cited involve 527 IPC main classification numbers. In order to capture the citation situation in technical fields outside the industry, the IPC main classification numbers that have appeared in the tobacco field patents in the cited patents are removed and statistical sorting is performed. The descriptive statistical results are shown in the following table.

[0050] Table 3 Descriptive statistics of cited patent IPC Number of IPCs Mean Standard Deviation Minimum Lower quartile Upper quartile Maximum 203 5.98 7.01 1 2 8 49 After removing the IPC numbers that appeared in tobacco patents from the cited patents, we obtained 203 IPC numbers of non-industry patents. Based on the screened IPC numbers of non-industry patents, we extracted 793 groups of citation relationships between non-industry technical fields and tobacco technical fields. The total number of citations of non-tobacco patents was 1,214 times. The most citations of a single IPC by tobacco patents was 49 times. The upper quartile of 8 means that 25% of the IPC numbers were cited by tobacco patents more than or equal to 8 times, and the lower quartile of 2 means that 25% of the IPC numbers were cited by tobacco patents less than or equal to 2 times. The average number of citations was 5.98, and 70 IPC numbers were cited higher than the average, accounting for 34.48%.

[0051] Table 4 The top 10 IPC technology fields in terms of citations IPC Citations Technical Field A47J 49 Kitchen appliances; coffee grinders; spice grinders; beverage preparation devices A23G 38 Cocoa; cocoa products; substitutes for cocoa or cocoa products; confectionery; chewing gum; ice cream; preparation thereof B21B 33 Metal rolling C12G 31 Wine; its preparation; alcoholic beverages A23P 30 Forming or processing of food materials not fully covered by another single subclass B28B 25 Forming of clay or other ceramic compositions; forming of slag; forming of mixtures containing cementitious materials H04B 23 transmission C08K 23 Use inorganic or non-polymer organic substances as ingredients A61F 22 Implantable filters; prostheses; devices for providing openings to tubular structures in the body or for preventing their collapse; warm compresses; treatment or protection of the eyes or ears; bandages, dressings, or absorbent pads; first aid kits B05D 21 General process for applying fluid to a surface D06F 21 Washing, drying, ironing, pressing or folding of textiles E21B 20 Drilling of soil or rock After sorting the top 10 IPC technology fields by the number of citations from large to small, it can be found through analysis that the tobacco industry has drawn important technical support from multiple technical fields. These technical fields cover a wide range of fields, from kitchen utensils and food preparation to metal processing and electronic communications. Among them, H04B and C08K have the same number of citations, both 23 times, and B05D and D06F have both 21 times. Therefore, there are 12 IPC classification numbers corresponding to the top 10 citations, accounting for about 6% of the IPC numbers in the non-industry technical field, and the number of citations is 336 times, accounting for 27.68% of the number of patent citations in the non-tobacco field, reflecting the concentration of citations in the technical field outside the industry. As can be seen from the table, the top rankings are mostly Part A (necessities for human life) and Part B (operation; transportation). Among them, the most referenced technical field is A47J (kitchen utensils; coffee grinders; spice grinders; beverage preparation devices), with 49 citations, accounting for 4% of the total number of citations of non-tobacco patents, followed by A23G, B21B, etc. These non-industry technical fields have made outstanding contributions to tobacco patents.

[0052] In order to better and more clearly show the contribution of non-industry technology fields to the tobacco field and promote the sustainable development of patent research in the tobacco field, the 12 IPC numbers ranked in the top 10 in the cited patents were taken as key technology fields outside the industry, and their technology flows were deeply analyzed, aiming to reveal the most direct objects of these key technology fields in tobacco patents. Through a systematic search of these 12 IPC numbers, a total of 159 groups of citation relationships with patents in the tobacco field were found, covering 94 different IPC classification numbers.

[0053] Figure 5 The network diagram of the first key technology field drawn by Python is shown, presenting the citation relationship between the technology field outside the industry and tobacco patents in the form of a directed graph. In order to analyze the dissemination path of technology, the point degree centrality index is used, including in-degree and out-degree. The direction of the arrow points from the IPC technology field outside the industry to the tobacco field, showing the specific flow direction of the technology. The size of the node reflects the size of the point degree centrality value. The larger the point degree centrality value, the larger the node, reflecting the importance of the node in the citation network. The thickness of the line is related to the number of citations between nodes. The more citations between nodes, the thicker the line.

[0054] After identifying external key technology areas, we analyze how the technologies in these areas directly affect the innovation activities of the tobacco industry. By screening the tobacco patent data for patents that directly apply external key technology areas, we can further clarify the specific impact of external industry technologies on the tobacco industry.

[0055] In order to show the information in the network structure diagram between key technology fields in more detail, Ucinet is used to quantitatively analyze the node attributes in the network diagram of the first key technology field. As shown in the following table, InDegree in the table is the absolute in-degree, that is, the number of citations of the node. The larger the absolute point degree centrality value, the more nodes are directly associated with it, which means that this node is closely connected with other nodes and occupies an important position in the network. NrmInDeg represents the relative in-degree, that is, the ratio of the corresponding absolute centrality value to the maximum possible degree in the network. The absolute point degree centrality is proportional to the relative point degree centrality.

[0056] Table 5 IPC entry TOP1 in key technology fields IPC InDegree NrmInDeg Technical Field A24B 76 4.54 Manufacture or preparation of smoking or chewing tobacco; tobacco; snuff A24D 31 1.85 Cigars; cigarettes; oil filters; cigar or cigarette holders; manufacture of oil filters or cigarette holders G01N 23 1.37 Testing or analyzing materials by determining their chemical or physical properties C11B 16 0.96 Production, e.g. by pressing raw materials or extracting from waste materials, refining or preserving fats and fatty substances B65G 15 0.90 Transport or storage equipment A24F 11 0.66 Smokers' articles; matchboxes; simulated smoking devices B01J 8 0.48 Chemical or physical methods, such as catalysis or colloid chemistry; apparatus therefor G05B 7 0.42 Control or regulation systems in general; functional units of such systems; monitoring or testing equipment for such systems or units A01G 7 0.42 Gardening; cultivation of vegetables, flowers, rice, fruit trees, grapes, hops or seaweed; forestry; watering B65B 6 0.36 Machinery, apparatus or equipment, or method for packaging articles or materials According to the statistical results, the absolute in-degree mean value of the network diagram is 3.574, the relative in-degree mean value is 0.214, and there are 82 nodes in the tobacco-related technical field, that is, the in-degree is greater than 0, of which 19 nodes are higher than the in-degree mean value, accounting for 23.17%. In addition, since the in-degree represents the inflow of knowledge in the technical field, the larger the in-degree, the more important the technical field is, and it represents the direction of popular research fields. According to the IPC in-degree ranking, it can be seen that the IPC classification number A24B, which ranks first, has the largest in-degree and 76 citations, indicating that the manufacturing field of smoking or chewing tobacco is the core research direction of tobacco patents. Secondly, A24D (patents related to cigars and cigarettes) and G01N (material property testing) are ranked at the top with in-degrees of 31 and 23 respectively, showing the tobacco industry's emphasis on these technical fields in production and quality inspection.

[0057] Furthermore, we use the consistency between IPC and industrial sectors, that is, one technology type is matched with only one industrial sector. With reference to the literature, we assign corresponding industries and technology fields. Combined with the citation frequency of technology fields outside the industry, we conduct statistics on technology citations based on the industry level. The descriptive statistical results are shown in the following table.

[0058] Table 6 Descriptive statistics of industry citations Number of industries Mean Standard Deviation Minimum Lower quartile Upper quartile Maximum 35 34.69 41.40 1 6.5 46.5 195 In order to further explore the knowledge flow and mutual influence between industries, a citation matrix based on the technology mutual citation relationship between industries was constructed. This matrix describes the knowledge transfer and citation relationship between industries based on the IPC classification numbers of the selected non-industry technology fields and the citation flow of non-industry technology fields. The behavior of the matrix is ​​the cited party, that is, the output of knowledge and technology, and the column is the citing party, that is, the recipient of knowledge and technology. The value k_(i,j) in the matrix represents the number of times the technical knowledge flows from industry i to industry j. Table 4 lists part of the correlation matrix. As shown in the table, the value of k_(1,2) in the first row and the second column is 64, indicating that the technology of industry 1 has been cited 64 times by industry 2.

[0059] Since the technical fields involved in tobacco patents have been excluded when considering the knowledge flow in technical fields outside the industry, and the self-citation relationship in the tobacco field is not considered, the technical fields cited by patents outside the industry do not include any technical fields in industry 2 (tobacco products). In this case, the technical fields in industry 2 are not cited by any industry, so the values ​​of the rows in which industry 2 is located in the matrix are all 0.

[0060] After obtaining the industry correlation matrix, Ucinet and Netdraw are used to quantitatively analyze the industry correlation matrix and draw a social network diagram, such as Figure 6 In network analysis, by connecting points, points and lines, and lines and lines, and using the basic parameters of social network analysis, we can analyze network attributes such as "position-structure-relationship", measure the inherent laws of the actual network from a statistical perspective, and thus summarize the network's associations.

[0061] As can be seen from the network diagram, the flow of technology between industries is also directed. Information in industrial nodes may flow to other nodes, or absorb technology from other nodes. The number of arrows from a node to other nodes is recorded as the absolute out-degree (OutDegree), which indicates the degree of external support provided by this industry. The number of arrows pointed to by other nodes is recorded as the absolute in-degree (InDegree), which represents how much industrial knowledge flows into the industry. The larger the absolute point degree centrality value, the more nodes are directly associated with it, which means that this node is closely connected with other nodes and occupies an important position in the network. NrmOutDe and NrmInDeg are the relative out-degree and relative in-degree, respectively, that is, the ratio of the corresponding absolute centrality value to the maximum possible degree in the network. The absolute point degree centrality is proportional to the relative point degree centrality. The following table shows the top 10 industries ranked by point in-degree and their corresponding values.

[0062] Table 7 Top 10 industries by penetration Serial number Industry Code InDegree OutDegree NrmInDeg NrmOutDeg 1 2 300 0 11.433 0 2 25 215 195 8.194 7.431 3 38 127 6 4.84 0.229 4 22 115 21 4.383 0.8 5 10 86 72 3.277 2.744 6 24 42 29 1.601 1.105 7 13 32 7 1.22 0.267 8 23 28 2 1.067 0.076 9 35 27 55 1.029 2.096 10 15 26 43 0.991 1.639 From the calculation results, the average absolute degree centrality of the entire network is 29.61, and there are 7 nodes with absolute in-degree higher than the average, corresponding to industries 2 (tobacco products), 25 (special-purpose machinery), 38 (measuring instruments), 22 (non-specific-purpose machinery), 10 (basic chemicals), 24 (machine tools), and 13 (pharmaceuticals). Figure 3 The distribution of patents in the tobacco field at the industry level can be found in the figure. The higher the in-degree, the more active the research in the technical field of the related industry. Since the research object itself is taken from patents in the tobacco field, industry 2 (tobacco products) has the highest in-degree and is significantly different from the corresponding values ​​of other industries, accounting for about a quarter of the total number of citations. As the research focus of this article, the tobacco industry relies on a wide range of technological inputs, especially machinery manufacturing, chemical and food-related industries, which are closely related to the production equipment, process improvement, product innovation and other aspects required in the tobacco manufacturing process. The second is industry 25 (special-purpose machinery), which shows its importance in multiple industries, especially in the fields of food, beverages, tobacco, etc., and there are tobacco-related technical fields A24C (machinery for making cigars or cigarettes) in this industry, so it has a higher in-degree. Through complex mechanical equipment, tobacco products can be produced faster and on a larger scale. There are also many technical fields involved in many links from tobacco raw material planting and production to tobacco product manufacturing in other industries with in-degrees higher than the average.

[0063] The thickness of the line reflects the intensity of technology flow between industries. The thicker the line, the closer the connection between industries and the more frequent the flow of knowledge. The following table is the statistical results of the top 10 inter-industry citations calculated based on the inter-industry citation relationship matrix. Among them, the cited industry is the industry code corresponding to the IPC classification number of the cited patent technology field from the excluded industry, and the citing industry is the industry code corresponding to the IPC classification number of the tobacco patent technology field with a citation relationship. The technology flow between industries is from the cited industry to the citing industry.

[0064] Table 8 Top 10 industries by citation flow Serial number Cited Industries Cited Industry frequency 1 1 2 64 2 27 2 49 3 25 2 48 4 19 25 21 5 25 22 19 6 10 2 16 7 28 38 16 8 42 22 16 9 1 15 15 10 19 2 14 Overall, there is a close technical interaction between the tobacco industry and many fields such as food, household appliances, machinery, and chemistry. On the one hand, the tobacco industry actively absorbs the technological achievements in the fields of food, household appliances, machinery, and chemicals. By citing these technologies, the tobacco industry can further expand to emerging fields such as e-cigarettes, cigarette accessories, and tobacco flavorings in addition to traditional cigarette manufacturing, enabling the tobacco industry to continuously improve its production processes, improve product characteristics, and explore more diversified product forms; on the other hand, the application needs of the tobacco industry in these fields are also driving the technological progress of the corresponding industries. This mutually reinforcing relationship will help enhance the overall competitiveness of my country's tobacco and promote the sustainable development of the entire tobacco industry in a changing market environment to meet the challenges of the market and policy environment.

[0065] Example 3 The difference between this embodiment and embodiment 2 is that: the cross-industry industry cross-correlation identification method, such as Figure 7 As shown, the following steps are also included: Obtain patent data in the target field and obtain the second category of non-industry patent data based on patent citation relationships.

[0066] Based on the IPC main classification number, an IPC citation matrix of the second category of non-industry patent data and target field patent data is constructed.

[0067] Specifically, an IPC citation matrix of the second category of non-industry patent data and target field patent data is constructed based on the IPC main classification number, including: Define the IPC main classification codes involved in the patent data in the target field as relevant technical fields; Collect all IPC main classification numbers in the second category of non-industry patent data that have citation relationships with patents in the target field; Screen all IPC main classification codes that do not belong to the relevant technical field as IPC main classification codes for non-target fields; Based on the IPC main classification number of the non-target field obtained in the previous step, an IPC citation matrix of the second category of non-industry patent data and target field patent data is constructed.

[0068] Based on the IPC citation matrix, the key technology fields outside the industry that have made great contributions to the target field are identified, and a network diagram of the second key technology field is constructed based on the IPC main classification number. The network diagram of the second key technology field is a directed graph, with the direction from cited patents to citing patents. The node attributes in the network diagram of the second key technology field are quantitatively analyzed to obtain the contribution value of each sub-technology field in the target field to the key technology fields outside the industry.

[0069] Based on the IPC main classification number and industry sector association mapping table and the IPC citation matrix, an industry association matrix of cited patents based on technology citation relationships between industries is constructed. Based on the cited patent industry association matrix, an industry citation directed network diagram is drawn, and the industry citation directed network diagram is quantitatively analyzed to obtain the industry citation relationship.

[0070] This example still takes the tobacco industry as an example, and focuses on analyzing patents outside the industry that cite tobacco patents, namely the so-called "cited patents". By conducting in-depth research on the technical fields of these cited patents, the application scope and value of tobacco patents in other industries can be clarified, which not only helps to reveal the technology output path of the tobacco industry, but also can evaluate the role of tobacco patents in promoting technological innovation outside the industry from a broader perspective.

[0071] After eliminating patent data with no citation relationship, a total of 10,746 patent data in the tobacco field were obtained, and these patents were cited by 18,132 patents outside the industry. By analyzing these data citing patents in the tobacco field, we can more intuitively reveal the contribution and influence of the tobacco industry on other technological fields. This process not only shows the diffusion path of tobacco technology in other industries, but also helps us clarify its role in cross-border technological innovation.

[0072] Specifically, according to statistics, the downloaded tobacco-related patent data with cited relationships involve a total of 272 IPC main classification numbers, and the cited patents involve 382 IPC main classification numbers. In order to obtain the contribution to the technical fields outside the industry, the IPC main classification numbers that have appeared in the tobacco-related patents in the cited patents are removed and statistical sorting is performed. The descriptive statistical results are shown in the following table. After removing the IPC numbers that appeared in tobacco patents from the cited patents, a total of 132 IPC numbers of non-industry patents were obtained. Based on the screened IPC numbers of non-industry patents, 264 groups of citation relationships between non-industry technical fields and tobacco technical fields were extracted. The total number of citations of non-tobacco patents was 320 times, and the maximum number of times non-industry patents cited tobacco patents was 10 times. The upper quartile of 3 means that 25% of the IPC numbers cited tobacco patents more than or equal to 3 times, and the lower quartile of 1 means that 25% of the IPC numbers cited tobacco patents less than or equal to 1 time. The average number of overall citations was 2.42, and 46 IPC numbers had citations higher than the average, accounting for 34.85%.

[0073] Table 9 Descriptive statistics of cited patent IPC Number of IPCs Mean Standard Deviation Minimum Lower quartile Upper quartile Maximum 132 2.42 1.90 1 1 3 10 It can be seen from the descriptive statistics that the contribution of the tobacco industry's technical field to the outside world is relatively scattered and concentrated in a few technical fields. Therefore, similar to the analysis process of cited patent technology fields, the top 10 patents in the tobacco field cited by the outside world are selected for analysis. According to statistics, a total of 11 IPC numbers (including parallel items) were selected as key technical fields flowing outside the industry. Table 10 lists the citations of patents in the tobacco field by patents outside the industry. Compared with the citations of patents outside the industry, it is obvious that there are still certain limitations in the contribution to patents outside the industry. The number of citations listed in the table is 78 times, accounting for 24.38% of the total number of citations, about one-quarter, while the IPC classification number only accounts for about 8.33% of the cited patent IPC. The IPC number B06B ranked first; the number of times it cited tobacco industry patents was only 10 times.

[0074] Table 10 Display of key technology fields flowing out of the industry IPC Inflow times Technical Field B06B 10 The occurrence or transmission of general mechanical vibration E21B 8 Drilling of soil or rock; extraction of oil, gas, water, soluble or fusible substances, or mineral slurries from a well A23F 8 Coffee; tea; substitutes thereof; their manufacture, preparation or brewing B05D 8 General process for applying fluid to a surface H04W 7 Wireless communication network F24D 7 Residential heating systems or district heating systems; residential hot water supply systems; parts or components therefor A47J 6 Kitchen appliances; coffee grinders; spice grinders; beverage preparation devices A23N 6 Machinery or apparatus for processing large quantities of harvested fruit, vegetables or flower bulbs, not otherwise included; peeling of large quantities of vegetables or fruit; apparatus for preparing feed for livestock B67C 6 Bottles, cans, jars, barrels or similar containers filled or semi-liquid or empty, not included in other categories F21V 6 Functional features or components of a lighting device or system In order to understand the specific contribution of tobacco industry technology fields to key technology fields outside the industry, after locking in key technology fields that flow outside the industry, we screened the patents in these technology fields that cited tobacco patents and drew a graph. Figure 8 Network diagram of key technology areas outside the industry shown.

[0075] In the network diagram, the lines point from the cited patents to the citing patents, so the peripheral nodes are all IPC classification numbers involved in the patents in the tobacco field, and the internal nodes are the IPC classification numbers corresponding to the technical fields outside the industry to which they flow. The color of the node is related to the industry to which the technical field belongs. The IPC classification numbers belonging to the same industry have the same color, and the node colors are different for different industries. The size of the node is related to the in-degree of the node. The larger the in-degree, the larger the node. It can be seen that A24B (manufacture or preparation of smoking or chewing tobacco; tobacco; snuff), A24F (smokers' supplies; matchboxes; simulated smoking devices), B65B (machinery, devices or equipment, or methods for packaging objects or materials) in the tobacco industry technology field have made significant contributions to technologies outside the industry.

[0076] When analyzing the impact of the tobacco industry on the outside of the industry, we can identify which tobacco technology fields have a greater impact on other technology fields through the mutual citation relationship between technical fields, especially through statistical out-degree ranking. These tobacco technology fields with more citations represent their contribution and expansion to external industries. As shown in the following table, OutDegree in the table is the absolute out-degree, that is, the number of citations of the node. The larger the absolute point degree centrality value, the more nodes are directly associated with it, which means that this node is closely connected with other nodes and occupies an important position in the network. NrmOutDeg represents the relative out-degree, that is, the ratio of the corresponding absolute centrality value to the maximum possible degree in the network. The absolute point degree centrality is proportional to the relative point degree centrality. The following table shows the top 10 tobacco-related technology fields ranked by node out-degree in the network diagram of key technology fields.

[0077] Table 11 Top 10 IPC outflows in key technology fields IPC OutDegree NrmOutDeg Technical Field A24F 10 2.717 Smokers' articles; matchboxes; simulated smoking devices A24B 7 1.902 Manufacture or preparation of smoking or chewing tobacco; tobacco; snuff C11B 5 1.359 Production, e.g. by pressing raw materials or extracting from waste materials, refining or preserving fats and fatty substances B65B 5 1.359 Machinery, apparatus or equipment, or method for packaging articles or materials B65G 4 1.087 Transport or storage equipment F24F 4 1.087 Air conditioning; Air humidification; Ventilation; Use of air flow as a shield B07B 3 0.815 Separate solids from solids by fine sieves, coarse sieves, sieving or by air flow B08B 3 0.815 General cleaning; general dirt removal F21S 3 0.815 Non-portable lighting devices or systems thereof; vehicle lighting equipment specially adapted for use on the exterior of a vehicle A24C 3 0.815 Machinery for making cigars or cigarettes According to the IPC out-degree ranking, it can be seen that the first-ranked IPC classification number A24F has the largest out-degree and has been cited 10 times, followed by the IPC classification number A24D node, with a corresponding out-degree value of 7, which means it has been cited 7 times. Since the out-degree represents the outflow of knowledge in the technical field, the larger the out-degree, the more contributions the relevant technical field has made to other technical fields, and can provide more value and reference significance. Therefore, the top-ranked technical fields have higher value and innovation than the bottom-ranked technical fields, and can provide rich technical basic theories for patent technologies in other fields.

[0078] Furthermore, based on the citations of tobacco patents by non-industry patents after screening, combined with the IPC and industry correspondence table, statistics were obtained based on industry citations. As shown in Table 12, non-industry patents based on cited data involved 33 industries in total, with an average of 9.7 citations at the industry level. A total of 8 industrial sectors had citations higher than the average, accounting for 24.24%. The maximum number of citations in the industrial sector was 90, and the minimum was only 1. The upper quartile value of 9 means that after sorting the citations from large to small, the top 25% of industries have citations greater than 9, and the lower quartile value of 2 means that the bottom 25% of industries have citations less than 2.

[0079] Table 12 Descriptive statistics of industry citation times Number of industries Mean Standard Deviation Minimum Lower quartile Upper quartile Maximum 33 9.70 15.77 1 2 9 90 Table 13 shows the top 10 industries in terms of citations. The higher the citations at the industry level, the greater the contribution of tobacco industry-related patents to the industry. It can be seen that, similar to the industry analysis results based on cited patents, the industry with the highest citations is also special-purpose machinery, and the gap between its citations and those of other industries is obvious, followed by the food / beverage industry, signal transmission / telecommunications industry, etc. From the perspective of the industrial sectors involved in the citations, patents in the tobacco field have a wider range of effects and are more closely related to research fields within the tobacco industry.

[0080] Table 13 Top 10 industries with the most citations industry Number of citations Industry Description 25 90 Special Purpose Machinery 1 23 Food, Beverages 35 22 Signal transmission, telecommunications 24 21 Machine Tools 37 16 Medical equipment 27 16 Home appliances 20 15 Metal products 18 14 Non-metallic mineral products 34 9 Electrical components 10 8 Basic Chemicals In order to further explore the citation situation between industries, 264 groups of citation relationships between the technical fields outside the industry and the tobacco technology field were obtained based on the IPC numbers of the patents outside the industry after screening, and the correlation matrix of the citation relationship between industries was established. After obtaining the industry correlation matrix, Ucinet and Netdraw were combined to quantitatively analyze the industry correlation matrix and draw a social network diagram, such as Fig. 9 As shown. In the patent industry association network diagram, the node size is related to the node in-degree. The larger the in-degree value, the larger the node. The thickness of the line between nodes is related to the amount of information flow, that is, the number of times the information flows between industries. The more times the flow, the thicker the line. The line from node A to node B means that the technical knowledge flows from industry A to industry B. Node A is the cited party and node B is the citing party.

[0081] from Fig. 9 It can be seen intuitively that in the patent industry association network diagram, the number of inflows into industry 25 (special-purpose machinery) is the largest, and its corresponding node is the largest. Since industry 2 (tobacco products) is a cited industry in this embodiment and there is no citing situation, its node is the smallest. In addition, it can be seen that most nodes except node 2 have both citing and being cited situations, representing different directions of technology flow between industries. For this reason, based on the inter-industry association matrix, Ucinet is used to quantitatively analyze the flow direction between industries. Table 14 shows the in-and-out degree of the top 10 industry nodes ranked by point out-degree.

[0082] Table 14 Top 10 industries by click-through rate Industry Code OutDegree InDegree NrmOutDeg NrmInDeg 2 55 0 7.639 0 22 51 3 7.083 0.417 25 32 53 4.444 7.361 38 17 2 2.361 0.278 20 13 13 1.806 1.806 15 11 6 1.528 0.833 10 9 5 1.25 0.694 17 8 2 1.111 0.278 21 7 2 0.972 0.278 23 7 1 0.972 0.139 According to statistics, the average values ​​of OutDegree absolute out-degree and InDegree absolute in-degree in the patent industry association network are both 6.919, and the average values ​​of NrmOutDeg relative out-degree and NrmInDeg relative in-degree are 0.961. There are 10 nodes with higher out-degree than the average, accounting for about 27.03% (10 / 37), and 11 nodes with higher in-degree than the average, accounting for about 29.73% (11 / 37). The larger the out-degree corresponding to the industry, the more likely it is that the technology of the industry has a certain diffusion ability, which can spread its innovative achievements through related industries and drive the upgrading and development of related industries. From the results in Table 12, industry 2 has the largest out-degree, which is to a certain extent due to the fact that the cited patents are all from the tobacco field. The second largest industry is industry 22 (non-specific purpose machinery). Most of the patents in this industry do not set specific uses and have a certain degree of versatility. The products or services it produces are widely used, supporting the production activities of many other industries. In addition, since self-citations are not taken into account when calculating the absolute in-degree and out-degree of directed graphs, we can see that the absolute in-degree of industry 25 in Table 12, i.e. the number of citations to other industries, is 53 times, and the number of citations of industry 25 in Table 11 is 90, which means that there are 37 cases of internal technology flows within the industry.

[0083] Example 4 Based on the same inventive concept, the embodiment of the present application also provides a cross-industry industry co-occurrence association identification device based on patent data for implementing the cross-industry industry co-occurrence association identification method based on patent data involved above. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the above method, so the specific limitations in one or more embodiments of the cross-industry industry co-occurrence association identification device based on patent data provided below can be referred to the above limitations on the cross-industry industry co-occurrence association identification method based on patent data, and will not be repeated here.

[0084] Specifically, the cross-industry industry association identification device based on patent data includes: The co-occurrence patent dataset construction module is used to obtain patent data in the target field, screen patent data in the target field with at least two different IPC subclass numbers, and construct a co-occurrence dataset; The technology co-occurrence analysis module is used to treat each IPC sub-class number in the co-occurrence data set as a network node, calculate the co-occurrence intensity value of any two network nodes, and form an IPC co-occurrence matrix; Industry co-occurrence analysis module, used to construct industry co-occurrence matrix based on IPC classification number and industry code mapping table and IPC co-occurrence matrix; The first category non-industry patent acquisition module is used to obtain the first category non-industry patent data based on the acquired target field patent data and patent citation relationships; IPC citation matrix construction module, which constructs the IPC citation matrix of the first category of non-industry patent data and target field patent data based on the IPC main and sub-category numbers; The first non-industry key technology field acquisition module is used to identify non-industry key technology fields that contribute greatly to the target field based on the IPC citation matrix; The first technology contribution analysis module is used to construct a network diagram of the first key technology field based on the IPC main and sub-class numbers, where the network diagram of the first key technology field is a directed graph, with the direction from the cited patent to the citing patent; quantitative analysis of the node attributes in the network diagram of the first key technology field is performed to obtain the contribution value of the key technology field outside the industry to each sub-technology field in the target field; The industry citation relationship analysis module is used to construct an industry association matrix of cited patents based on the technical mutual citation relationship between industries based on the IPC main and sub-category numbers and industrial sector association mapping table and the IPC citation matrix, draw an industry citation association network based on the cited patent industry association matrix, and conduct quantitative analysis on the industry citation association network to obtain industry citation relationships.

[0085] Furthermore, the cross-industry cross-correlation identification device based on patent data also includes: The second category non-industry patent acquisition module is used to obtain the second category non-industry patent data based on the acquired target field patent data and patent citation relationship; IPC citation matrix construction module, which constructs the IPC citation matrix of the second category of non-industry patent data and target field patent data based on the IPC main classification number; The second non-industry key technology field acquisition module is used to identify non-industry key technology fields that have made great contributions to the target field based on the IPC citation matrix; The second technology contribution analysis module is used to construct a network diagram of the second key technology field based on the IPC main classification number, where the network diagram of the second key technology field is a directed graph, with the direction from the cited patent to the citing patent; quantitative analysis of the node attributes in the network diagram of the second key technology field is performed to obtain the contribution value of each sub-technical field in the target field to the key technology field outside the industry; The industry citation relationship analysis module builds an industry association matrix of cited patents based on the technology citation relationship between industries based on the IPC main classification number and industry sector association mapping table and the IPC citation matrix. It draws an industry citation directed network diagram based on the cited patent industry association matrix, and conducts quantitative analysis on the industry citation directed network diagram to obtain the industry citation relationship.

[0086] Example 5 Based on the above embodiment, this embodiment provides a computer device, which may be a terminal, and its internal structure diagram may be as follows: Fig.10As shown. The computer device includes a processor, a memory, an input / output interface, a communication interface, a display unit and an input device. Among them, the processor, the memory and the input / output interface are connected through a system bus, and the communication interface, the display unit and the input device are connected to the system bus through the input / output interface. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The input / output interface of the computer device is used to exchange information between the processor and the external device. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless method can be implemented through WIFI, a mobile cellular network, NFC (near field communication) or other technologies. When the computer program is executed by the processor, it implements the cross-industry industry co-occurrence association identification method based on patent data described in Example 1, or the cross-industry industry association identification method based on patent data as described in Example 2 or 3.

[0087] Those skilled in the art will understand that Fig.10 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0088] Based on the above embodiments, this embodiment provides a computer device including a memory and a processor, wherein a computer program is stored in the memory, and when the processor executes the computer program, the cross-industry industry co-occurrence association identification method based on patent data described in Embodiment 1 is implemented.

[0089] Example 6 Based on the above embodiments, this embodiment provides a readable storage medium on which a computer program is stored. When the computer program is executed by a processor, it implements the cross-industry industry co-occurrence association identification method based on patent data described in Example 1, or the cross-industry industry association identification method based on patent data as described in Example 2 or 3.

[0090] Example 7 Based on the above embodiments, this embodiment provides a computer program product, including a computer program, which, when executed by a processor, implements the cross-industry industry co-occurrence association identification method based on patent data described in Example 1, or the cross-industry industry association identification method based on patent data as described in Example 2 or 3.

[0091] A person of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be implemented by instructing related hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods.

[0092] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0093] The above-described embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present application. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

[0094] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, ordinary technicians in the field should understand that the specific implementation methods of the present invention can still be modified or some technical features can be replaced by equivalents without departing from the spirit of the technical solution of the present invention, which should be included in the scope of the technical solution for protection of the present invention.

Claims

1. A cross-industry industry co-occurrence association identification method based on patent data, characterized in that: The following steps are involved: Obtain patent data in the target field, screen patent data in the target field with at least two different IPC subclass numbers, and construct a co-occurrence dataset; Take each IPC subclass number in the co-occurrence data set as a network node, calculate the co-occurrence strength value of any two network nodes, form an IPC co-occurrence matrix, and draw an IPC co-occurrence undirected network graph based on the IPC co-occurrence matrix; the co-occurrence strength value is the number of co-occurrences of two different IPC subclass numbers in the same patent data; An industry co-occurrence matrix is ​​constructed based on the IPC subclass number and industry code mapping table and the IPC co-occurrence matrix, and an industry co-occurrence undirected network diagram is drawn based on the industry co-occurrence matrix.

2. According to claim 1, a method for identifying cross-industry industry co-occurrence associations based on patent data is characterized in that: Obtain patent data in the target field, screen patent data in the target field with at least two different IPC subclass numbers, and construct a co-occurrence dataset, including: Obtain patent data in the target field, screen patent data with at least two IPC classification numbers to build a primary patent data set; Patent data with at least two different IPC subclass numbers are screened in the primary patent dataset to construct a co-occurrence dataset.

3. A cross-industry industry association identification method based on patent data, characterized in that: The following steps are involved: In the patent data of the target field, select patent data with at least two different IPC subclass numbers to construct a co-occurrence dataset; Take each IPC subclass number in the co-occurrence data set as a network node, calculate the co-occurrence strength value of any two network nodes, and form an IPC co-occurrence matrix; the co-occurrence strength value is the number of co-occurrences of two different IPC subclass numbers in the same patent data; Construct an industry co-occurrence matrix based on the IPC sub-category number and industry code mapping table and the IPC co-occurrence matrix; Take each IPC subclass number in the co-occurrence data set as a network node, calculate the co-occurrence strength value of any two network nodes, and form an IPC co-occurrence matrix; the co-occurrence strength value is the number of co-occurrences of two different IPC subclass numbers in the same patent data; Construct an industry co-occurrence matrix based on the IPC sub-category number and industry code mapping table and the IPC co-occurrence matrix; At the same time, patent data outside the industry is obtained based on the patent mutual citation relationship in the patent data of the target field; Construct an IPC citation matrix of the first category of non-industry patent data and target field patent data based on IPC main and sub-category numbers; Based on the IPC citation matrix, identify the key technology areas outside the industry that have made great contributions to the target field, and construct a network diagram of the first key technology area based on the IPC main and sub-class numbers. The network diagram of the first key technology area is a directed graph, with the direction from the cited patent to the citing patent; quantitatively analyze the node attributes in the network diagram of the first key technology area to obtain the contribution value of the key technology areas outside the industry to each sub-technical area in the target field; Based on the IPC main and sub-category numbers and industrial sector association mapping table and the IPC citation matrix, an industry association matrix of cited patents based on the technical mutual citation relationship between industries is constructed. Based on the cited patent industry association matrix, an industry citation association network is drawn, and the industry citation association network is quantitatively analyzed to obtain the industry citation relationship.

4. The cross-industry industry association identification method based on patent data according to claim 3 is characterized in that: The following steps are also included: Obtain patent data in the target field, and obtain the second category of non-industry patent data based on patent citation relationships; Based on the IPC main and sub-class numbers, an IPC citation matrix of the second category of patent data outside the industry and patent data in the target field is constructed; Based on the IPC citation matrix, we identify the key technology areas outside the industry that have made great contributions to the target field, and build a network diagram of the second key technology area based on the IPC major and minor categories. The network diagram of the second key technology area is a directed graph, with the direction from the cited patents to the citing patents. Quantitatively analyze the node attributes in the network diagram of the second key technology field to obtain the contribution value of each sub-technology field in the target field to the key technology field outside the industry; Based on the IPC main and sub-category numbers and industrial sector association mapping table and the IPC citation matrix, an industry association matrix of cited patents based on technology citation relationships between industries is constructed. Based on the cited patent industry association matrix, an industry citation directed network diagram is drawn, and the industry citation directed network diagram is quantitatively analyzed to obtain the industry citation relationship.

5. The method for identifying cross-industry industry associations based on patent data according to claim 4 is characterized in that: Based on the IPC main and sub-class numbers, an IPC citation matrix of patent data outside the industry and patent data in the target field is constructed, including: Define the IPC main and sub-class numbers involved in the patent data in the target field as relevant technical fields; Count all IPC main and sub-class numbers in the first category of non-industry patent data that have citation relationships with patents in the target field; Screen all IPC main and sub-class numbers that do not belong to the relevant technical field as the IPC main and sub-class numbers of non-target fields; Based on the IPC major and minor categories of the non-target fields obtained in the previous step, an IPC citation matrix of the first category of non-industry patent data and the target field patent data is constructed; Based on the IPC main and sub-class numbers, the IPC citation matrix of the second category of non-industry patent data and target field patent data is constructed, including: Define the IPC main and sub-class numbers involved in the patent data in the target field as relevant technical fields; Collect all IPC main and sub-class numbers in the second category of non-industry patent data that have citation relationships with patents in the target field; Screen all IPC main and sub-class numbers that do not belong to the relevant technical field as the IPC main and sub-class numbers of non-target fields; Based on the IPC main and sub-category numbers of the non-target fields obtained in the previous step, an IPC citation matrix of the second category of non-industry patent data and target field patent data is constructed.

6. A device for identifying cross-industry industry associations based on patent data, characterized in that: include: The co-occurrence patent dataset construction module is used to obtain patent data in the target field, screen patent data in the target field with at least two different IPC subclass numbers, and construct a co-occurrence dataset; The technology co-occurrence analysis module is used to treat each IPC sub-class number in the co-occurrence data set as a network node, calculate the co-occurrence intensity value of any two network nodes, and form an IPC co-occurrence matrix; Industry co-occurrence analysis module, used to construct industry co-occurrence matrix based on IPC classification number and industry code mapping table and IPC co-occurrence matrix; The first category non-industry patent acquisition module is used to obtain the first category non-industry patent data based on the acquired target field patent data and patent citation relationships; IPC citation matrix construction module, which constructs the IPC citation matrix of the first category of non-industry patent data and target field patent data based on the IPC main and sub-category numbers; The first non-industry key technology field acquisition module is used to identify non-industry key technology fields that contribute greatly to the target field based on the IPC citation matrix; The first technology contribution analysis module is used to construct a network diagram of the first key technology field based on the IPC main and sub-class numbers, where the network diagram of the first key technology field is a directed graph, with the direction from the cited patent to the citing patent; quantitative analysis of the node attributes in the network diagram of the first key technology field is performed to obtain the contribution value of the key technology field outside the industry to each sub-technology field in the target field; The industry citation relationship analysis module is used to construct an industry association matrix of cited patents based on the technical mutual citation relationship between industries based on the IPC main and sub-category numbers and industrial sector association mapping table and the IPC citation matrix, draw an industry citation association network based on the cited patent industry association matrix, and conduct quantitative analysis on the industry citation association network to obtain industry citation relationships.

7. The cross-industry industry association identification device based on patent data according to claim 6 is characterized in that: Also includes: The second category non-industry patent acquisition module is used to obtain the second category non-industry patent data based on the acquired target field patent data and patent citation relationship; IPC citation matrix construction module, which constructs the IPC citation matrix of the second category of non-industry patent data and target field patent data based on IPC main and sub-category numbers; The second non-industry key technology field acquisition module is used to identify non-industry key technology fields that have made great contributions to the target field based on the IPC citation matrix; The second technology contribution analysis module is used to construct a network diagram of the second key technology field based on the IPC main and sub-class numbers, where the network diagram of the second key technology field is a directed graph, with the direction from the cited patent to the citing patent; Quantitatively analyze the node attributes in the network diagram of the second key technology field to obtain the contribution value of each sub-technology field in the target field to the key technology field outside the industry; The industry citation relationship analysis module builds an industry association matrix of cited patents based on the technology citation relationship between industries based on the IPC major and sub-category numbers and industrial sector association mapping table and the IPC citation matrix. It draws an industry citation directed network diagram based on the cited patent industry association matrix, and conducts quantitative analysis on the industry citation directed network diagram to obtain the industry citation relationship.

8. A computer device, characterized in that: It includes a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory communicate with each other via the communication bus; Memory, used to store computer programs; A processor, when used to execute a program stored in a memory, implements a cross-industry industry co-occurrence association identification method based on patent data as described in any one of claims 1 to 2, or implements a cross-industry industry association identification method based on patent data as described in any one of claims 3 to 5.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, it implements the cross-industry industry co-occurrence association identification method based on patent data as described in any one of claims 1 to 2, or implements the cross-industry industry association identification method based on patent data as described in any one of claims 3 to 5.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, it implements the cross-industry industry co-occurrence association identification method based on patent data as described in any one of claims 1 to 2, or implements the cross-industry industry association identification method based on patent data as described in any one of claims 3 to 5.

Images