CN117112735A

CN117112735A - Patent database construction method and electronic equipment

Info

Publication number: CN117112735A
Application number: CN202311351864.3A
Authority: CN
Inventors: 王军雷; 王亮亮; 龙悦; 季南; 冀然; 郭少杰; 叶晓雪; 刘兰; 丁强
Original assignee: Zhongqi Intellectual Property Guangzhou Co ltd; China Automobile Information Technology Tianjin Co ltd
Current assignee: Zhongqi Intellectual Property Guangzhou Co ltd; China Automobile Information Technology Tianjin Co ltd
Priority date: 2023-10-19
Filing date: 2023-10-19
Publication date: 2023-11-24
Anticipated expiration: 2043-10-19
Also published as: CN117112735B

Abstract

The application relates to a construction method of a patent database and electronic equipment, and relates to the technical field of electric digital processing, wherein the method comprises the steps of retrieving a plurality of one-level patents of a target enterprise from a comprehensive patent database; based on a semantic similarity model, screening patents with similarity larger than a threshold value with each previous-level patent from the comprehensive patent database to serve as the patent of the level; the threshold value of the level is determined by the patent value of the previous level and the threshold value of the previous level, and the threshold value of the previous level is larger than or equal to the threshold value of the level; and the like, obtaining a plurality of hierarchical patents, and constructing a patent analysis database of a target enterprise. The application realizes progressive patent number expansion by virtue of the semantic similarity model and the variable threshold value, thereby constructing a comprehensive database for enterprise patent analysis, which is closely related to enterprise technology.

Description

Patent database construction method and electronic equipment

Technical Field

The application relates to the technical field of electric digital processing, in particular to a construction method of a patent database and electronic equipment.

Background

The enterprise patent analysis refers to collecting a large amount of patent information from patent documents such as enterprise patent specifications and patent publications, processing, sorting, combining and analyzing the patent information by a scientific method, further converting the information into competition information with global and predictive actions by a statistical method, and providing a reference set for decision making in the technical, product and service development of enterprises.

When providing patent analysis service for enterprises, it is necessary to select patents with high correlation with enterprise technology from mass patents as analysis sources, which is beneficial to accurate analysis and noise reduction. In the prior art, the technical field to which the patent of the enterprise belongs (for example, the technical field indicated by the classification number) is generally determined, and then the patent in the technical field is determined as an analysis source, but in this way, the patent is omitted and noise is not small.

The present application has been made in view of the above-described drawbacks.

Disclosure of Invention

In order to solve the technical problems, the application provides a construction method of a patent database and electronic equipment, which realize progressive patent number expansion by virtue of a semantic similarity model and a variable threshold value, thereby constructing a comprehensive database for enterprise patent analysis, which is closely related to enterprise technology.

The embodiment of the application provides a construction method of a patent database, which comprises the following steps:

s1, retrieving a plurality of one-level patents of a target enterprise from a comprehensive patent database;

s2, based on a semantic similarity model, screening patents with similarity larger than a threshold value with each previous-level patent from the comprehensive patent database, and taking the patents as the patents of the level; the threshold value of the level is determined by the patent value of the previous level and the threshold value of the previous level, and the threshold value of the previous level is larger than or equal to the threshold value of the level;

s3, a plurality of hierarchical patents are obtained by analogy, and a patent analysis database of the target enterprise is constructed.

Optionally, the S1 includes:

s11, constructing a search type based on the target enterprise, and searching in the comprehensive patent database to obtain a level of patent;

s12, taking the cited patent and the family patent of the hierarchy of patents as the hierarchy of patents.

Optionally, the S2 includes:

s21, obtaining the similarity between the patents in the comprehensive patent database and the patents of the previous level, which are output by each module, based on a background similarity module, a summary similarity module, a class number similarity module, an application content similarity module and a main right item similarity module in the semantic similarity model;

s22, weighting and summing the similarity corresponding to each module to obtain final similarity;

s24, screening patents with final similarity larger than a threshold value as the patent of the hierarchy.

Optionally, before S21, the method further includes:

s20, counting the classification numbers of the patents of each previous level and determining the classification number frequency;

the classification number similarity module outputs the weight of the similarity and determines the weight according to the frequency of the patent classification number of the previous level.

Optionally, after S3, the method further includes:

s4, adding the same family patents and the cited patents of the multiple layers of patents to the patent analysis database of the target enterprise.

Optionally, the threshold of the present level is determined by the patent value of the previous level and the threshold of the previous level, including:

if the patent value of the previous level is larger than the set value, the threshold value of the current level is equal to the threshold value of the previous level;

if the patent value of the previous level is less than the set value, the threshold of the present level is determined based on the patent value of the previous level and the threshold of the previous level.

Optionally, the patent value is determined according to the operation condition and legal status of the patent.

Optionally, the patent value is determined according to the operation condition and legal status of the patent, including:

extracting an operation field and a legal status field of the patent from the comprehensive patent database, wherein the operation field comprises an assignment field, a permission field and a mortgage field;

and obtaining the patent value according to the operation field and the legal status field.

Optionally, after S2, the method further includes:

after the patent of the hierarchy is obtained, randomly selecting a set number of patents from the patent of the hierarchy, and comparing the similarity with each patent of the hierarchy;

if any selected patent is dissimilar to each patent of a hierarchy, discarding the patent of the hierarchy, and stopping patent screening;

if all patents selected are similar to any patent at a level, patent screening continues.

The embodiment of the application provides electronic equipment, which comprises:

one or more processors;

a memory for storing one or more programs,

when the one or more programs are executed by the one or more processors, the one or more processors are caused to implement a method of building any patent database.

The method and the device provided by the application have the following technical effects:

1) According to the application, a level of patent applied by a target enterprise is searched first, and then progressive patent screening is carried out level by level based on a semantic similarity model, so that quick indexing and expansion of the number of similar patents are realized, and the search type and manual screening are not required to be constructed manually.

2) The higher the level of patents the larger the corresponding threshold, the patent expands indefinitely, adding excessive noise.

3) The similarity threshold is determined according to the patent value, and more similarity expansion can be performed based on the high-value patent, so that the method is beneficial to screening out patents which are more similar to a hierarchy of patents and are more valuable.

4) After the patent of the hierarchy is obtained, similarity is calculated by selecting a certain number of patents and a hierarchy of patents, a cut-off condition is set, and expansion is stopped in time.

5) According to the application, considering that a target enterprise has technical research and development under the technical field of high-frequency class number characterization, the class number similarity module outputs the weight of the similarity to be determined according to the frequency of the patent class number of the previous level, and more similar patents can be accurately found by utilizing the characteristics.

Drawings

In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings that are needed in the description of the embodiments or the prior art will be briefly described, and it is obvious that the drawings in the description below are some embodiments of the present application, and other drawings can be obtained according to the drawings without inventive effort for a person skilled in the art.

FIG. 1 is a schematic flow chart of a method for constructing a patent database according to an embodiment of the present application;

FIG. 2 is a schematic illustration of a progressive patent extension provided by an embodiment of the present application;

fig. 3 is a schematic structural diagram of an electronic device according to an embodiment of the present application.

Detailed Description

In order to make the objects, technical solutions and advantages of the present application more apparent, the technical solutions of the present application will be clearly and completely described below. It will be apparent that the described embodiments are only some, but not all, embodiments of the application. All other embodiments, which can be made by one of ordinary skill in the art without undue burden from the application, are within the scope of the application.

Example 1

Fig. 1 is a flowchart of a method for constructing a patent database, which is provided by the embodiment of the present application, and is suitable for screening technology related patents for a target enterprise to be used as a patent analysis data source, where the method provided by the present application includes the following operations:

s1, searching a plurality of one-level patents of a target enterprise from a comprehensive patent database.

The comprehensive patent database comprises the recently published patents at home and abroad, and can be any commercial or free database. The search type of the constructed target enterprise is input into a search box of a comprehensive patent database, and a plurality of patents applied by the target enterprise are searched and obtained, which are called a hierarchy of patents.

Preferably, S11, constructing a search type based on the target enterprise, and searching in the comprehensive patent database to obtain a level of patent; s12, the cited patents and the family patents of the one-level patents are taken as the one-level patents, so that the number of the one-level patents is enlarged.

S2, based on a semantic similarity model, screening patents with similarity larger than a threshold value with each previous-level patent from the comprehensive patent database, and taking the patents as the patents of the level.

To serve the technology of the target enterprise based on patent analysis, in addition to a hierarchy of patents, competitor patents are required to understand the development of the prior art. Based on this, similar patents are screened from the comprehensive patent database, starting from a hierarchy of patents, level by level to communicate with the patent analysis database. The similarity or not is obtained by comparing the output similarity of the semantic similarity model with a threshold value.

Optionally, S21, obtaining the similarity between the patents in the comprehensive patent database and the patents of the previous level, which are output by each module, based on a background similarity module, a summary similarity module, a class number similarity module, an application content similarity module and a main weight similarity module in the semantic similarity model; s22, weighting and summing the similarity corresponding to each module to obtain final similarity; s24, screening patents with final similarity larger than a threshold value as the patent of the hierarchy.

The background similarity module, the abstract similarity module, the class number similarity module, the application content similarity module and the main right item similarity module respectively conduct sentence extraction, identification and similarity calculation aiming at the background, the abstract, the class number, the application content and the main right item of the two patents. The output of each module is weighted and summed to obtain the final similarity of the two patents, and the weight of the output similarity of each module can be the same or different.

Preferably, the classification number is used for accurately classifying the technical field of the patent, and the target enterprise has the characteristic of focusing on a certain technical field, if a certain classification number appears much more, the target enterprise is proved to have technical research and development under the technical field characterized by the classification number. More similar patents can be accurately found by utilizing the characteristics, and based on the characteristics, the method also comprises the step of counting the classification numbers of the patents at the previous level to determine the classification number frequency before S21; the classification number similarity module outputs the weight of the similarity and determines the weight according to the frequency of the patent classification number of the previous level.

For example, when calculating the similarity to patent C, patent C is B60F, the frequency of occurrence is 80% in all patents of the previous hierarchy, and the weight of the class number similarity module in the model for calculating the similarity to patent C is 80%.

Specifically, screening patents with similarity larger than a threshold value with a first-level patent from the comprehensive patent database to serve as second-level patents; then screening patents with similarity larger than a threshold value with each two-level patent from a comprehensive patent database to be used as three-level patents; and then, screening patents with similarity larger than a threshold value with each three-level patent from a comprehensive patent database, and as four-level patents, so that the patents of the levels are obtained on the basis of the previous level, and the progressive expansion of the patents is presented, so that the patent analysis database is constructed by the patents of all levels. It should be noted that, the threshold value corresponds to each previous-level patent, and the threshold value of a certain-level patent may be different.

In the present application, each level after a level needs to calculate similarity with the previous level patent and compare with the threshold corresponding to the previous level patent. The threshold value of each layer is larger than or equal to the threshold value of the previous layer, the threshold value is higher and higher, and the requirement of similarity is more and more severe, so that the similarity between each layer and a layer of patent can be maintained, unlimited expansion of the patent is avoided, and excessive noise is added.

Preferably, the threshold of the present hierarchy is determined by the patent value of the previous hierarchy and the threshold of the previous hierarchy. If the patent value of the previous level is larger than the set value, the threshold value of the current level is equal to the threshold value of the previous level; if the patent value of the previous level is less than the set value, the threshold of the present level is determined based on the patent value of the previous level and the threshold of the previous level.

Fig. 2 is a schematic diagram of progressive patent expansion provided by an embodiment of the present application, where the percentage is a similarity threshold. A and B are two-piece one-level patents applied by a target enterprise, and the similarity between the patents in the comprehensive patent database and A or B is calculated one by one. When the output of the semantic similarity model is greater than or equal to a threshold of 60% (the threshold of the second hierarchy may be preset), it is determined as a second hierarchy patent: C. d, E. For each two-level patent, the patents in the comprehensive patent database are computed for similarity, one by one, with C, D or E, and then compared with the threshold corresponding to C, D or E patent. Assuming that the similarity with the patent C is calculated, the patent value of the C needs to be calculated first, and if the value is larger than a set value, the threshold value corresponding to the C is equal to the threshold value in the screening of the patent C and is 60%; if the patent value of C is smaller than the set value, the threshold value corresponding to C is equal to 60% divided by the patent value (the patent value is 0-1), for example 80%. The patent C, D, E may correspond to different similarity thresholds due to the different patent values.

The set value is used for evaluating the value of the patent and can be set according to the service requirement.

Further, the patent value is determined according to the operation condition and legal status of the patent. The more the operation business is, the more the number of times is, the higher the patent value is, the more the technology behind the operation business has analysis value, so the requirement on similar patents is relaxed, and the same threshold value as the previous level is adopted. In a specific implementation, an operation field and a legal status field of a patent are extracted from a comprehensive patent database, wherein the operation field comprises an assignment field, a permission field and a mortgage field; and obtaining the patent value according to the operation field and the legal status field. Illustratively, patent C includes 1 transfer field (value a) and 1 mortgage field (value b), and legal status is authorized (value C), then a+b+c is the patent value of patent C.

Optionally, after S3, the method further includes: the patent of the same family and the cited patent of a plurality of hierarchy patents are added to the target enterprise patent analysis database.

With progressive expansion from layer to layer, the similarity between the patents at the layer level and the patents at the layer level is lower and lower, and in order to avoid introducing excessive noise, the expansion is stopped in time, and a cut-off condition is set. After S2, further comprising: after the patent of the hierarchy is obtained, randomly selecting a set number of patents from the patent of the hierarchy, and comparing the similarity with each patent of the hierarchy; if any selected patent is dissimilar to each patent of a hierarchy, discarding the patent of the hierarchy, and stopping patent screening; if all patents selected are similar to any patent at a level, patent screening continues. The set number may be 30% of the present level patent.

Assuming that the patent P is the only patent of three levels, calculating the similarity between the patent P and the patents A and B of one level through a semantic similarity model, if the similarity between the patent P and the patents A and B of one level is smaller than a set value, the patent P is not similar, the patents of three levels are abandoned, and the patent screening is stopped to enter S3. If the similarity between the patent P and the patent B is larger than or equal to the set value, continuing the patent screening of the next level.

When more than two patents are randomly selected, each patent needs to be compared with a hierarchy of similarity. Whenever there is a patent that is not similar to each patent of a hierarchy, the patent of the hierarchy is discarded; otherwise, continuing the patent screening of the next level.

Fig. 3 is a schematic structural diagram of an electronic device according to an embodiment of the present application, and as shown in fig. 3, the device includes a processor 30, a memory 31, an input device 32 and an output device 33; the number of processors 30 in the device may be one or more, one processor 30 being taken as an example in fig. 3; the processor 30, memory 31, input means 32 and output means 33 in the device may be connected by a bus or other means, in fig. 3 by way of example.

The memory 31 is a computer readable storage medium, and may be used to store a software program, a computer executable program, and modules, such as program instructions/modules corresponding to the method for constructing a patent database in the embodiment of the present application. The processor 30 executes various functional applications of the apparatus and data processing by running software programs, instructions and modules stored in the memory 31, i.e., implements the above-described construction method of the patent database.

The memory 31 may mainly include a storage program area and a storage data area, wherein the storage program area may store an operating system, at least one application program required for functions; the storage data area may store data created according to the use of the terminal, etc. In addition, the memory 31 may include high-speed random access memory, and may also include non-volatile memory, such as at least one magnetic disk storage device, flash memory device, or other non-volatile solid-state storage device. In some examples, memory 31 may further include memory located remotely from processor 30, which may be connected to the device via a network. Examples of such networks include, but are not limited to, the internet, intranets, local area networks, mobile communication networks, and combinations thereof.

The input device 32 may be used to receive entered numeric or character information and to generate key signal inputs related to user settings and function control of the apparatus. The output means 33 may comprise a display device such as a display screen.

It is noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the scope of the present application. As used in this specification, the terms "a," "an," "the," and/or "the" are not intended to be limiting, but rather are to be construed as covering the singular and the plural, unless the context clearly dictates otherwise. The terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, or apparatus. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method or apparatus comprising such elements.

It should also be noted that the positional or positional relationship indicated by the terms "center", "upper", "lower", "left", "right", "vertical", "horizontal", "inner", "outer", etc. are based on the positional or positional relationship shown in the drawings, are merely for convenience of describing the present application and simplifying the description, and do not indicate or imply that the apparatus or element in question must have a specific orientation, be constructed and operated in a specific orientation, and thus should not be construed as limiting the present application. Unless specifically stated or limited otherwise, the terms "mounted," "connected," and the like are to be construed broadly and may be, for example, fixedly connected, detachably connected, or integrally connected; can be mechanically or electrically connected; can be directly connected or indirectly connected through an intermediate medium, and can be communication between two elements. The specific meaning of the above terms in the present application will be understood in specific cases by those of ordinary skill in the art.

Finally, it should be noted that: the above embodiments are only for illustrating the technical solution of the present application, and not for limiting the same; although the application has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some or all of the technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the essence of the corresponding technical solutions from the technical solutions of the embodiments of the present application.

Claims

1. A method for constructing a patent database, comprising:

2. The method according to claim 1, wherein S1 comprises:

3. The method according to claim 1, wherein S2 comprises:

4. A method according to claim 3, further comprising, prior to S21:

5. The method of claim 1, further comprising, after S3:

6. The method of claim 1, wherein the threshold value for the present hierarchy is determined by the patent value of the previous hierarchy and the threshold value of the previous hierarchy, comprising:

7. The method of claim 1, wherein the patent value is determined based on the operating condition and legal status of the patent.

8. The method of claim 7, wherein the patent value is determined based on the operating condition and legal status of the patent, comprising:

9. The method according to any one of claims 1-8, further comprising, after S2:

10. An electronic device, comprising:

one or more processors;

a memory for storing one or more programs,

the one or more programs, when executed by the one or more processors, cause the one or more processors to implement the method of constructing a patent database as recited in any one of claims 1-9.