Matching method, device, electronic device and storage medium of patent applicant

Abstract

The present disclosure relates to a method, device, electronic device and storage medium for matching patent applicants. The matching method includes: based on the feature information corresponding to the applicant to be stored in the patent document to be stored, retrieving the candidate applicant corresponding to the applicant to be stored in a preset database; determining the similarity between the candidate applicant and the applicant to be stored; based on the similarity, determining whether the applicant to be stored successfully matches the candidate applicant. The matching method, device, electronic device and storage medium for patent applicants provided by the present disclosure have a higher correlation between the selected candidate applicants and the applicant to be stored, thereby effectively improving the matching accuracy between the applicant to be stored and the candidate applicant, and reducing the probability of data redundancy or matching errors in related technologies.

Description

Technical Field

[0001] The present disclosure relates to the field of text processing, and in particular to a matching method, device, electronic device and storage medium of a patent applicant. Background Art

[0002] At present, when entering patent documents into the databases of major patent search platforms, it is usually necessary to match the applicants to be included in the database with the applicants already in the database. An accurate matching method can effectively avoid the redundancy of applicant data. For example, if the applicant to be included actually already exists in the database, but if the match is not successful, the database will create a new applicant ID. This not only occupies database resources, but also causes different patent documents of the same applicant to be unable to be correctly retrieved together, and may even cause the reader's patent layout to conflict with the unretrieved patent documents, leading to unnecessary intellectual property disputes. Therefore, how to accurately establish a matching relationship between applicants to be included in the database and applicants already in the database is a technical problem that the major patent search platforms currently need to solve. Summary of the invention

[0003] According to a first aspect of the present disclosure, a method for matching patent applicants is provided, the matching method comprising: based on feature information corresponding to the applicant to be stored in the patent document to be stored, retrieving candidate applicants corresponding to the applicant to be stored in a preset database; determining the similarity between the candidate applicant and the applicant to be stored; based on the similarity, determining whether the applicant to be stored is successfully matched with the candidate applicant; wherein the feature information comprises at least one of: address information corresponding to the applicant to be stored, patent family number corresponding to the applicant to be stored, classification information corresponding to the applicant to be stored, inventor information corresponding to the applicant to be stored, foreign language information corresponding to the applicant to be stored, patent semantic information corresponding to the applicant to be stored, and identification information corresponding to the applicant to be stored.

[0004] In a possible implementation, the matching method includes at least one of the following: dividing the first address information corresponding to the patent document to be stored into second address information according to a hierarchical relationship, and using the region number corresponding to the second address information as the address information corresponding to the applicant to be stored; obtaining the patent family number corresponding to the patent document to be stored as the patent family number corresponding to the applicant to be stored; obtaining the patent classification number corresponding to the patent document to be stored as the classification information corresponding to the applicant to be stored; obtaining the inventor information corresponding to the patent document to be stored as the inventor information corresponding to the applicant to be stored; converting the name of the applicant to be stored into a foreign applicant name, and normalizing the foreign applicant name as the foreign information corresponding to the applicant to be stored; determining the patent semantic information corresponding to the applicant to be stored according to the semantic vector corresponding to the patent document to be stored; dividing the name of the applicant to be stored into the identification information corresponding to the applicant to be stored, wherein the identification information includes: at least one of the applicant subject, field name, and company category.

[0005] In a possible implementation, based on the characteristic information corresponding to the applicant to be included in the patent document to be included, the candidate applicant corresponding to the applicant to be included is retrieved in a preset database, including: dividing the characteristic information corresponding to the applicant to be included into at least one query field, and querying the at least one query field in the existing index table in the preset database; and selecting the first preset number of first applicants with the highest matching degree in the preset database as the candidate applicant.

[0006] In a possible implementation, the matching method further includes: when it is determined that the applicant to be admitted is successfully matched, taking each of the query fields as an index, taking all of the query fields as entities corresponding to the indexes, and updating the index table.

[0007] In a possible implementation, determining whether the applicant to be admitted has been successfully matched with the candidate applicant based on the similarity includes: taking the candidate applicant corresponding to the highest similarity as the result applicant who has been successfully matched with the applicant to be admitted, and determining that the applicant to be admitted has been successfully matched; or taking the candidate applicant corresponding to the highest similarity and the highest similarity being higher than a preset similarity threshold as the result applicant who has been successfully matched with the applicant to be admitted, and determining that the applicant to be admitted has been successfully matched.

[0008] In one possible implementation, determining the similarity between the candidate applicant and the applicant to be admitted includes: generating feature vectors corresponding to the candidate applicant and the applicant to be admitted based on feature information corresponding to the candidate applicant and the applicant to be admitted; and generating the similarity between the candidate applicant and the applicant to be admitted through a machine learning model based on the feature vectors corresponding to the candidate applicant and the applicant to be admitted.

[0009] In a possible implementation, the preset database is an Elasticsearch database.

[0010] In a possible implementation, the matching method further includes: establishing a link between the applicant to be included in the database and the result applicant, and saving the patent document to be included corresponding to the applicant to be included in the database into the first database.

[0011] In a possible implementation, the matching method further includes: obtaining characteristic information corresponding to the first applicant; using at least one characteristic information corresponding to the first applicant as an index, using all characteristic information corresponding to the first applicant as entities corresponding to the index, and establishing an index table in the preset database.

[0012] In a possible implementation, the obtaining of characteristic information corresponding to the first applicant includes at least one of the following: dividing the first address information corresponding to the first applicant into second address information according to a hierarchical relationship, and using the region code corresponding to the second address information as the address information corresponding to the first applicant; obtaining the patent family number corresponding to the first applicant in the first database as the patent family number corresponding to the first applicant; obtaining a second preset number of patent classification numbers corresponding to the first applicant in the first database as the classification information corresponding to the first applicant; obtaining a third preset number of inventor information corresponding to the first applicant in the first database as the inventor information corresponding to the first applicant; normalizing the foreign applicant name corresponding to the first applicant in the first database as the foreign information corresponding to the first applicant; determining the patent semantic information corresponding to the first applicant based on the semantic vector corresponding to the first applicant in the first database; and dividing the name of the first applicant into identification information corresponding to the first applicant, wherein the identification information includes at least one of the applicant entity, field name, and company category.

[0013] According to a second aspect of the present disclosure, a matching device for patent applicants is provided, the matching device comprising: a candidate generation module, for retrieving candidate applicants corresponding to the applicants to be stored in a preset database based on feature information corresponding to the applicants to be stored in the patent documents to be stored; a similarity generation module, for determining the similarity between the candidate applicants and the applicants to be stored; a matching result generation module, for determining whether the applicants to be stored are successfully matched with the candidate applicants based on the similarity; wherein the feature information comprises: at least one of the address information corresponding to the applicants to be stored, the patent family number corresponding to the applicants to be stored, the classification information corresponding to the applicants to be stored, the inventor information corresponding to the applicants to be stored, the foreign language information corresponding to the applicants to be stored, the patent semantic information corresponding to the applicants to be stored, and the identification information corresponding to the applicants to be stored.

[0014] According to a third aspect of the present disclosure, there is provided an electronic device comprising: a processor; and a memory for storing instructions executable by the processor; wherein the processor is configured to execute any one of the matching methods of the patent applicant described above.

[0015] According to a fourth aspect of the present disclosure, a non-volatile computer-readable storage medium is provided, on which computer program instructions are stored, and when the computer program instructions are executed by a processor, the matching method of the patent applicant described in any one of the above items is implemented.

[0016] The patent applicant matching method provided in the embodiment of the present disclosure can retrieve the candidate applicant corresponding to the applicant to be stored in the preset database based on the feature information corresponding to the applicant to be stored in the patent document to be stored, and then determine the similarity between the candidate applicant and the applicant to be stored, and determine whether the applicant to be stored is successfully matched with the candidate applicant based on the similarity. Since the candidate applicant is obtained based on the feature information corresponding to the applicant to be stored, the correlation between the candidate applicant and the applicant to be stored is higher, which can effectively improve the matching accuracy of the applicant to be stored and the candidate applicant, and reduce the probability of data redundancy or matching errors in related technologies.

[0017] It should be understood that the above general description and the following detailed description are exemplary and explanatory only and do not limit the present disclosure. Other features and aspects of the present disclosure will become clear from the following detailed description of exemplary embodiments with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] The accompanying drawings, which are incorporated in and constitute a part of the specification, illustrate exemplary embodiments, features, and aspects of the disclosure and, together with the description, serve to explain the principles of the disclosure.

[0019] Figure 1 A flowchart of a method for matching patent applicants provided according to an embodiment of the present disclosure.

[0020] Figure 2 A flowchart of a method for matching patent applicants provided according to an embodiment of the present disclosure.

[0021] Figure 3 A block diagram of a matching device provided by the patent applicant according to an embodiment of the present disclosure

[0022] Figure 4 The present invention is a block diagram of an electronic device provided according to an embodiment of the present disclosure. DETAILED DESCRIPTION

[0023] Various exemplary embodiments, features and aspects of the present disclosure will be described in detail below with reference to the accompanying drawings. The same reference numerals in the accompanying drawings represent elements with the same or similar functions. Although various aspects of the embodiments are shown in the accompanying drawings, the drawings are not necessarily drawn to scale unless otherwise specified.

[0024] The word “exemplary” is used exclusively herein to mean “serving as an example, example, or illustration.” Any embodiment described herein as “exemplary” is not necessarily to be construed as preferred or advantageous over other embodiments.

[0025] In addition, in order to better illustrate the present disclosure, numerous specific details are given in the following specific embodiments. It should be understood by those skilled in the art that the present disclosure can also be implemented without certain specific details. In some examples, methods, means, components and circuits well known to those skilled in the art are not described in detail in order to highlight the subject matter of the present disclosure.

[0026] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the features. In the description of the present disclosure, the meaning of "plurality" is two or more, unless otherwise clearly and specifically defined.

[0027] At present, the matching method for patent applicants usually uses string matching to match and link the applicants to be included in the database with the applicants already in the database. Two common methods are listed here for reference. Example 1: Directly match the string of the applicant to be included in the database with the string already in the database, or clean the string of the applicant to be included in the database before matching. If set in this way, the matching effect will be poor, which may easily cause data redundancy or matching errors. Example 2: Match the string of the applicant to be included in the database with the string of the applicant already in the database by calculating the edit distance between the two, but such a setting still cannot solve the problem of data redundancy or matching errors well.

[0028] The patent applicant matching method provided in the embodiment of the present disclosure can retrieve the candidate applicant corresponding to the applicant to be stored in the preset database based on the feature information corresponding to the applicant to be stored in the patent document to be stored, and then determine the similarity between the candidate applicant and the applicant to be stored, and determine whether the applicant to be stored is successfully matched with the candidate applicant based on the similarity. Since the candidate applicant is obtained based on the feature information corresponding to the applicant to be stored, the correlation between the candidate applicant and the applicant to be stored is higher, which can effectively improve the matching accuracy of the applicant to be stored and the candidate applicant, and reduce the probability of data redundancy or matching errors in related technologies.

[0029] In a possible implementation, the patent document number classification method provided in the embodiment of the present disclosure can be executed by an electronic device such as a terminal device or a server. The terminal device can be a user equipment (UE), a mobile device, a user terminal, a terminal, a cellular phone, a cordless phone, a personal digital assistant (PDA), a handheld device, a computing device, a vehicle-mounted device, a wearable device, etc. The method can be implemented by a processor calling a computer-readable instruction stored in a memory. Alternatively, the method can be executed by a server. Exemplarily, the server can be a cloud server or a local server.

[0030] Exemplarily, when a user needs to enter a patent document to be stored into a patent database, the patent document to be stored can be first entered into the terminal device, and then the terminal device determines whether the applicant to be stored corresponding to the patent document to be stored is successfully matched with a candidate applicant through the matching method of patent applicants provided in the embodiment of the present disclosure. The matching result (for example: the result applicant who successfully matches among the candidate applicants) can be output to the display screen of the terminal device for the user to manually enter the relevant information of the patent document to be stored (for example: the number or identifier corresponding to the result applicant) into the patent database (that is, the first database) to establish a link between the applicant to be stored and the result applicant. Exemplarily, the terminal device can also automatically enter the relevant information of the patent document to be stored into the patent database to establish a link between the applicant to be stored and the result applicant (for example: different patent documents belonging to the same patent applicant, the corresponding patent applicant numbers or identifiers can be set to be the same).

[0031] See also Figure 1 As shown, Figure 1 FIG. 1 is a flow chart of a method for matching patent applicants according to an embodiment of the present disclosure. Figure 1 As shown, the matching method includes:

[0032] Step S100, based on the characteristic information corresponding to the applicant to be stored in the patent document to be stored, the candidate applicant corresponding to the applicant to be stored is retrieved in the preset database. Exemplarily, the above characteristic information may include: the address information corresponding to the applicant to be stored, the patent family number corresponding to the applicant to be stored, the classification information corresponding to the applicant to be stored, the inventor information corresponding to the applicant to be stored, the foreign information corresponding to the applicant to be stored, the patent semantic information corresponding to the applicant to be stored, and at least one of the identification information corresponding to the applicant to be stored. The above characteristic information and its combination can accurately represent the various characteristics of the applicant to be stored in the patent document to be stored. For example, in combination with the actual situation, the address information of the same patent applicant is usually the same or similar, and the research field of the same patent applicant is usually similar, that is, the above classification information and patent semantic information are similar or the same. The research team of the same patent applicant may be similar, that is, the above inventor information is similar or the same. The name of the same patent applicant is usually not changed, that is, the above foreign name and identification information are similar or the same. If several patent applicants with similar or identical names correspond to the same group of patents of the same family, then the above patent applicants may be the same patent applicant. Using the above feature information or its combination, the correlation between candidate applicants and applicants to be included in the database can be more accurately judged, which can effectively improve the matching accuracy between applicants to be included in the database and candidate applicants, and reduce the probability of data redundancy or matching errors in related technologies.

[0033] Exemplarily, the above-mentioned address information may be the address recorded in the patent document to be stored, or it may be the hierarchical structure of the address, or it may be the hierarchical id corresponding to the hierarchical structure of the address, for example: the address information may be No. 96, XX Road, XX District, XX City, XX Province, or it may be XX District, XX City, XX Province, or it may be 102030 (here only for exemplary purposes, the hierarchical id of XX Province is taken as 10, the hierarchical id of XX City is taken as 20, and the hierarchical id of XX District is taken as 30). The hierarchical structure and the number of hierarchical structure ids used in the above-mentioned address information can be selected according to the needs of the developer, and the embodiments of the present disclosure are not limited here. Since the geographical locations of the same patent applicant are generally not too far apart, especially when the patent applicant is a company, in the embodiments of the present disclosure, the address information can be used as a kind of characteristic information of the applicant to be stored.

[0034] Exemplarily, the patent family number corresponding to the applicant to be included in the database may be the patent family number corresponding to the patent family document of the patent document to be included in the database. The patent number may be a patent application number, a patent publication number, etc. If a patent document corresponds to multiple patent family documents, the patent family numbers corresponding to different patent family documents may be the same. Under normal circumstances, the patent applicants of the patent family corresponding to a patent document are likely to be the same. Therefore, in the disclosed embodiment, the patent family number can be used as a characteristic information of the applicant to be included in the database.

[0035] Exemplarily, the classification information corresponding to the applicant to be included in the database may include at least one of the International Patent Classification Number (IPC Classification Number) and the Joint Patent Classification Number (CPC Classification Number). Under normal circumstances, the research fields of the same patent applicant are usually similar, and the International Patent Classification Number and the Joint Patent Classification Number can represent the technical fields involved in the patent document. Therefore, in the embodiment of the present disclosure, the classification information can be used as a kind of characteristic information of the applicant to be included in the database.

[0036] Exemplarily, the inventor information corresponding to the applicant to be stored may be the inventor information recorded in the patent document to be stored, for example, if the patent document to be stored records 'Applicant (Patent Right) Person: A' and 'Inventor: B, C', then it is considered that the inventor information corresponding to the applicant to be stored A may be the character string corresponding to B and / or C. In the case where the patent applicant is a small enterprise, the same inventor usually corresponds to multiple patent documents, so in the disclosed embodiment, the inventor information may be used as a characteristic information of the applicant to be stored.

[0037] For example, the foreign language information corresponding to the applicant to be included in the database may be the foreign name translated from the Chinese name of the applicant to be included in the database, for example, the foreign name may be English, Japanese, Korean, etc. If the same patent applicant applies for patents in different countries, the name of the patent applicant will usually use the language corresponding to the country, for example, if the name of the applicant in the patent document applied for by the patent applicant in China is David Robert, then in the patent document applied for in the UK, the name of the applicant may be David Robert, in fact, the patent applicant of the two patent documents is the same. Therefore, in the disclosed embodiment, the foreign language information can be used as a kind of characteristic information of the applicant to be included in the database.

[0038] Exemplarily, the patent semantic information corresponding to the applicant to be included in the database may be the text features corresponding to the patent document to be included in the database corresponding to the applicant to be included in the database, for example: the text features corresponding to the patent document to be included may include: the text features of the specification abstract and / or patent name in the patent document to be included in the database (for example, can be represented by a text vector). Since the specification abstract and the patent name can summarize the content of the patent document to a certain extent, the text of the specification abstract and the patent name part can be selected as the patent semantic information corresponding to the above-mentioned patent document to be included. Under normal circumstances, since the research fields of the same patent applicant are usually similar, the signal patent voice information corresponding to the patent documents belonging to the same patent applicant may be similar, that is, the above-mentioned text features may be similar, so in the embodiment of the present disclosure, the patent semantic information can be used as a kind of characteristic information of the applicant to be included in the database.

[0039] Exemplarily, the identification information corresponding to the applicant to be included in the database may include: applicant subject, field name, and company category. For example, if the applicant to be included in the database is "Pineapple Chemical Co., Ltd.", then the corresponding applicant subject is "Pineapple" (i.e., the company brand name), the field name is "Chemicals", and the company category is "Limited Company." Since the name of the patent applicant usually does not change, in the embodiment of the present disclosure, the identification information can be used as a characteristic information of the applicant to be included in the database.

[0040] Continue to refer Figure 1As shown, step S200 determines the similarity between the candidate applicant and the applicant to be admitted. Exemplarily, a similarity calculation function can be used to calculate the difference value of each feature information, and then the similarity between the candidate applicant and the applicant to be admitted is calculated by adding (or weighted summing) the difference values ​​corresponding to each feature information. Exemplarily, the difference value corresponding to each feature information is used to represent the degree of difference between the feature information of the candidate applicant and the feature information of the applicant to be admitted, and the degree of difference between the two can be positively correlated with the difference value, that is, the greater the degree of difference, the greater the difference value. Exemplarily, the above-mentioned difference value can be negatively correlated with the similarity, that is, the higher the difference value, the lower the similarity, that is, the greater the difference between the candidate applicant and the applicant to be admitted. The similarity calculation function can be implemented based on relevant technologies, and this application does not limit this. Exemplarily, the similarity between the candidate applicant and the applicant to be admitted can also be calculated through a machine learning model to avoid the similarity calculation process being too linear, thereby improving the accuracy of the similarity.

[0041] Step S300: Based on the similarity, determine whether the applicant to be admitted is successfully matched with the candidate applicant.

[0042] In a possible implementation, the matching method further includes: establishing a link between the applicant to be included and the result applicant, and saving the patent document to be included corresponding to the applicant to be included in the first database. Exemplarily, according to the matching result, the patent document to be included of the applicant to be included can be manually or automatically entered into the first database later, and the embodiment of the present disclosure is not limited here. The above-mentioned first database can be set up independently from the preset database, that is, the preset database only stores the characteristic information of the first applicant, which is mainly used for the selection of candidate applicants, and the storage of the patent document to be included is the responsibility of the first database (that is, the first database is responsible for connecting to the search tasks posted by users in the patent search platform).

[0043] The patent applicant matching method provided in the embodiment of the present disclosure can retrieve the candidate applicant corresponding to the applicant to be stored in the preset database based on the feature information corresponding to the applicant to be stored in the patent document to be stored, and then determine the similarity between the candidate applicant and the applicant to be stored, and determine whether the applicant to be stored is successfully matched with the candidate applicant based on the similarity. Since the candidate applicant is obtained based on the feature information corresponding to the applicant to be stored, the correlation between the candidate applicant and the applicant to be stored is higher, which can effectively improve the matching accuracy of the applicant to be stored and the candidate applicant, and reduce the probability of data redundancy or matching errors in related technologies.

[0044] In a possible implementation manner, before step S100, the following steps may also be included:

[0045] The first address information corresponding to the applicant to be stored in the patent document to be stored is divided into second address information according to the hierarchical relationship, and the regional number corresponding to the second address information is used as the address information corresponding to the applicant to be stored. Which levels are included in the hierarchical relationship can be determined as needed. Exemplarily, the above-mentioned first address information may be the address recorded in the patent document, and the above-mentioned hierarchical relationship may include provinces, cities, and districts, and the above-mentioned regional numbers are the numbers of each province, city, and district in the same database. For example: when the first address information is No. 96, XX Road, XX District, XX City, XX Province, the second address information is XX District, XX City, XX Province, and the regional number is 102030 (this is only for exemplary purposes, if 10 is the number of XX Province, 20 is the number of XX City, and 30 is the number of XX District).

[0046] In a possible implementation manner, before step S100, the following steps may also be included:

[0047] The patent family number corresponding to the patent document to be stored is obtained as the patent family number corresponding to the applicant to be stored. For example, different patent families corresponding to the same patent document may have the same patent family number.

[0048] In a possible implementation manner, before step S100, the following steps may also be included:

[0049] Obtain the patent classification number corresponding to the patent document to be stored as the classification information corresponding to the applicant to be stored. Exemplarily, the classification information may include the International Patent Classification Number (i.e., IPC classification number) and the Combined Patent Classification Number (i.e., CPC classification number). The number of the above patent classification numbers may be selected as any integer according to the needs of the user, and the embodiments of the present disclosure are not limited here. For example: the number of the above patent classification numbers is 3, and the IPC classification numbers corresponding to the patent documents to be stored include: G06N3 / 063, G06N3 / 04, G06N3 / 08, and G06N20 / 00. Then, 3 of the above 4 IPC classification numbers are selected as the classification information corresponding to the applicant to be stored. Exemplarily, the IPC main classification number may also be selected as the classification information corresponding to the applicant to be stored, that is, the number of patent classification numbers may be 1.

[0050] In a possible implementation manner, before step S100, the following steps may also be included:

[0051] Obtain the inventor information corresponding to the patent document to be stored as the inventor information corresponding to the applicant to be stored. The number of the above-mentioned inventor information can be selected as any integer according to the needs of the user, and the embodiment of the present disclosure is not limited here. For example: the number of the above-mentioned inventor information is 2, and the patent document to be stored includes the inventors: Wang Da, Bai Er, and Zhang San. Then, any two of the above-mentioned three inventors can be selected as the inventor information corresponding to the applicant to be stored. Exemplarily, the fewer the number of the above-mentioned inventor information, the higher the retrieval efficiency.

[0052] In a possible implementation manner, before step S100, the following steps may also be included:

[0053] The name of the applicant to be admitted is converted into a foreign applicant name, and the foreign applicant name is normalized as the foreign language information corresponding to the applicant to be admitted. For example, the foreign applicant name can be the English, Japanese, Korean, etc. name corresponding to the applicant to be admitted. The specific method of the above normalization is not limited in this disclosure, and can be the same as the normalization rule of the foreign applicant name when the preset database is established.

[0054] In a possible implementation manner, before step S100, the following steps may also be included:

[0055] According to the semantic vector corresponding to the patent document to be stored, the patent semantic information corresponding to the applicant to be stored is determined. The above semantic vector can be realized by the text feature extraction method in the relevant technology, and the embodiment of the present disclosure is not limited here. For example: the text corresponding to the patent name and / or abstract text in the patent document to be stored can be converted into a semantic vector in the vector space through VSM (i.e., Vector Space Model). Exemplarily, the average value of multiple patent semantic vectors corresponding to the first applicant in the preset database can be compared with the semantic vector corresponding to the above applicant to be stored, so as to screen out candidate applicants with higher relevance from the first applicant. The method of screening candidate applicants in the preset database is not limited here in the embodiment of the present disclosure.

[0056] In a possible implementation manner, before step S100, the following steps may also be included:

[0057] The names of the applicants to be admitted are classified as identification information corresponding to the applicants to be admitted, wherein the identification information includes: at least one of the applicant subject, field name, and company category. Exemplarily, there may be multiple ways to generate identification information, which are not limited in the embodiments of the present disclosure. For example: the names of the applicants to be admitted may be classified as identification information by manual methods, direct string matching, machine learning models, etc. For example: if the applicant is 'Pineapple Chemical Co., Ltd.', the identification information corresponding to the applicants to be admitted includes: the applicant subject is 'Pineapple' (i.e., the company brand name), the field name is 'Chemical', and the company category is 'Limited Liability Company'.

[0058] See also Figure 2 As shown, Figure 2 FIG. 1 is a flow chart of a method for matching patent applicants according to an embodiment of the present disclosure. Figure 2 As shown, in a possible implementation manner, step S100 includes:

[0059] Step S110, divide the characteristic information corresponding to the applicant to be included in the database into at least one query field, and query the at least one query field in the existing index table in the preset database. Exemplarily, the above-mentioned preset database may be an Elasticsearch database. Elasticsearch database is a distributed database with high scalability and more efficient search capabilities. The above-mentioned preset database may include characteristic information corresponding to multiple first applicants for the applicant to be included in the database to compare with it. Exemplarily: in the case where the above-mentioned characteristic information only includes the patent classification number (for example: G06K9 / 32) and the address information corresponding to the applicant to be included in the database (for example, 102030), 'G06K9 / 32' and '102030' are used as two query fields, and the existing index table is queried respectively. Exemplarily, the above-mentioned index table records multiple indexes (i.e., index), and each index corresponds to an entity (i.e., document). For example, the index table includes indexes: 'G06K9 / 30', 'G06K9 / 50' (i.e., classification information), '231120', '154422' (i.e., address information), 'Peter', 'Alex' (i.e., foreign language information). If the indexes 'G06K9 / 30', '231120', and 'Peter' are characteristic information of the same first applicant, and the indexes 'G06K9 / 50', '154422', and 'Alex' are characteristic information of the same first applicant, then the entity corresponding to the index 'G06K9 / 30' is 'G06K9 / 30, 231120, Peter', the entity corresponding to the index '231120' is also 'G06K9 / 30, 231120, Peter', and the entity corresponding to the index 'Alex' is 'G06K9 / 50, 154422, Alex'. In other words, each index in the index table corresponds to an entity that records the characteristic information of the first applicant. The entity can be recorded in the index table in the form of entity number. For example, the entity number of the entity 'G06K9 / 30, 231120, Peter' is 1, then the entity number 1 is recorded in the entity number corresponding to the indexes 'G06K9 / 30', '231120', and 'Peter' respectively. The entity number can be automatically generated by the preset library when the entity is established. The characteristic information recorded by the same entity (or entity number) belongs to the same first applicant. Since each index may be occupied by multiple first applicants, each index can also correspond to multiple entities. For example: the address of the first applicant A is '102030', and the address of the first applicant B can also be '102030'. In other words, if '102030' represents XX District, XX City, XX Province, then the first applicants A and B are both from XX District, XX City, XX Province. This is a common phenomenon, so each index in this database can also correspond to multiple entities.

[0060] Step S120, the first preset number of first applicants with the highest matching degree in the preset database queried are used as the candidate applicants. The above-mentioned first preset number can be selected as any integer according to the needs of the user, and the embodiment of the present disclosure is not limited here. Exemplarily, the matching degree between the above-mentioned first applicant and the applicants to be admitted to the database can be obtained by adding the weights corresponding to the query fields (that is, the matching degree can be determined according to the importance of each query field in the actual situation). Exemplarily, the matching degree between the above-mentioned first applicant and the applicants to be admitted to the database can be positively correlated with the first index number, and the above-mentioned first index number is the same number of indexes in the indexes corresponding to all query fields and the first applicant. For example, the query fields corresponding to the applicants to be added to the database are 'G06K9 / 30', '231120', and 'Peter', the indexes corresponding to the first applicant A are 'G06K9 / 30', '231120', and 'Evan', the indexes corresponding to the first applicant B are 'G06K9 / 30', '111020', and 'Lina', and the indexes corresponding to the first applicant C are 'G50J8 / 11', '111020', and 'Toya'. Then the number of first indexes corresponding to the first applicant A is 2, the number of first indexes corresponding to the first applicant B is 1, and the number of first indexes corresponding to the first applicant C is 0, that is, the matching degrees of the three from high to low are: the first applicant A, the first applicant B, and the first applicant C. Exemplarily, a preset threshold can be set to screen the first applicant as an optional candidate applicant. For example, the preset threshold can be a screening value for the number of first indexes, that is, when the number of first indexes corresponding to the first applicant is greater than or equal to the preset threshold, the first applicant is selected as an optional candidate applicant. In combination with the above example, if the preset threshold is 1, the first applicant A and the first applicant B can be selected as candidate applicants, and then the first preset number of candidate applicants with the highest matching degree are screened out from the selectable candidate applicants to further increase the matching success rate between the candidate applicants and the applicants to be added to the database. This is only an exemplary statement, and the specific screening method of the candidate applicants can be determined according to the search characteristics in the preset database, and the embodiment of the present disclosure is not limited here.

[0061] In one example, when it is determined that the applicant to be admitted is matched successfully, each of the query fields is used as an index, and all the query fields are used as entities corresponding to their indexes, and the index table is updated. For example, if the query fields this time are 'G06K9 / 30', '231120', and 'Peter', and the existing indexes in the index table are the same as the three query fields, then the entity numbers corresponding to the existing indexes in the index table are updated, for example, the entity numbers of the entities 'G06K9 / 30, 231120, and Peter' are added to the entity numbers corresponding to the three indexes 'G06K9 / 30', '231120', and 'Peter'. If the match fails, an index is created, and the entity numbers of the entities 'G06K9 / 30, 231120, and Peter' are recorded in the corresponding positions of the indexes in the index table. For example, if the indexes of 'G06K9 / 30', '231120', and 'Peter' are created for the first time, 'G06K9 / 30, 231120, and Peter' are taken as the corresponding entities, and the relationship between the three indexes and the entities is established by recording the entity numbers in the index table.

[0062] Continue reading Figure 2 In a possible implementation, step S200 may include:

[0063] Step S210, based on the feature information corresponding to the candidate applicant and the applicant to be admitted, generate feature vectors corresponding to the candidate applicant and the applicant to be admitted. Exemplarily, the above feature vectors can be realized by feature extraction methods in related technologies, and the embodiments of the present disclosure are not limited here. For example, the feature information corresponding to the candidate applicant and the applicant to be admitted can be converted into feature vectors in vector space by VSM (i.e., Vector Space Model).

[0064] Step S220, based on the feature vectors corresponding to the candidate applicant and the applicant to be admitted, the similarity corresponding to the candidate applicant and the applicant to be admitted is generated through a machine learning model. Exemplarily, the input of the machine learning model in the training stage may be the feature vectors corresponding to the candidate applicant and the applicant to be admitted, and the output is the predicted similarity between the two. The predicted similarity between the two is compared with the actual similarity between the two to adjust the internal parameters of the machine learning model until the accuracy of the predicted similarity converges. The disclosed embodiment does not limit the type of machine learning model used. Exemplarily, the machine learning model can use the LightGBM model (a machine learning model based on a decision tree algorithm), which has the advantages of faster training speed, lower memory consumption, and higher accuracy. Exemplarily, if there are multiple candidate applicants, the feature vector corresponding to each candidate applicant and the feature vector corresponding to the applicant to be admitted are combined into multiple data pairs, and each data pair is input into the machine learning model to obtain the similarity between each candidate applicant and the applicant to be admitted.

[0065] In a possible implementation, step S300 may include: taking the candidate applicant corresponding to the highest similarity as the result applicant who successfully matches the applicant to be admitted, and determining that the applicant to be admitted is successfully matched; or taking the candidate applicant corresponding to the highest similarity and the highest similarity higher than a preset similarity threshold as the result applicant who successfully matches the applicant to be admitted, and determining that the applicant to be admitted is successfully matched. The above-mentioned preset similarity threshold is not limited in the embodiments of the present disclosure. For example, the higher the above-mentioned preset similarity threshold, the higher the matching accuracy rate of the candidate applicant who successfully matches the applicant to be admitted.

[0066] In a possible implementation, the matching method may further include a pre-step to establish the above-mentioned preset database, including: obtaining characteristic information corresponding to the first applicant. Using at least one characteristic information corresponding to the first applicant as an index, using all characteristic information corresponding to the first applicant as entities corresponding to the index, and establishing an index table in the preset database.

[0067] Exemplarily, the obtaining of characteristic information corresponding to the first applicant includes at least one of the following:

[0068] The first address information corresponding to the first applicant is divided into second address information according to a hierarchical relationship, and the area number corresponding to the second address information is used as the address information corresponding to the first applicant. For example, the number of the above address information can be determined according to actual conditions, and the embodiment of the present disclosure is not limited here.

[0069] The patent family number corresponding to the first applicant in the first database is obtained as the patent family number corresponding to the first applicant. For example, the number of the patent family numbers can be determined according to actual conditions, and the embodiment of the present disclosure is not limited thereto.

[0070] A second preset number of patent classification numbers corresponding to the first applicant in the first database is obtained as classification information corresponding to the first applicant. Exemplarily, the second preset number may be determined according to actual conditions, for example, the five largest patent classification numbers among all patent classification numbers corresponding to the first applicant may be selected in the first database as classification information corresponding to the first applicant.

[0071] A third preset number of inventor information corresponding to the first applicant in the first database is obtained as the inventor information corresponding to the first applicant. Exemplarily, the third preset number may be determined according to actual conditions, for example, the five with the largest number of inventor information corresponding to the first applicant may be selected in the first database as the inventor information corresponding to the first applicant.

[0072] Normalize the foreign applicant name corresponding to the first applicant in the first database as the foreign information corresponding to the first applicant. Exemplarily, the number of the above foreign applicant names can be determined according to the actual situation, and the above foreign applicant names can be normalized after data cleaning to facilitate comparison with the applicants to be included in the database. The disclosed embodiment avoids the differences caused by special symbols in the foreign applicant names through normalization. For example: invalid text differences caused by special symbols in the foreign applicant names corresponding to the same applicant can be avoided by normalizing separators such as '·' and spaces in English to the same separator. In other words, 'David Robert', 'David·Robert', and 'David.Robert' can be normalized to 'David Robert'. For another example: 'Pineapple Computer.Inc' and 'Pineapple Computer Inc' are normalized to 'Pineapple Computer Inc'.

[0073] According to the semantic vector corresponding to the first applicant in the first database, the patent semantic information corresponding to the first applicant is determined. Exemplarily, the above semantic vector may be the average value of multiple patent semantic vectors corresponding to the first applicant. The above patent semantic vector is used to summarize the content of the first applicant in the stored patent documents. For example: the patent semantic vector may be the text features such as the title and abstract of the stored patent documents.

[0074] The name of the first applicant is classified as identification information corresponding to the first applicant, wherein the identification information includes at least one of the applicant entity, field name, and company category.

[0075] See also Figure 3 As shown, Figure 3 FIG. 1 is a block diagram of a matching device provided by the patent applicant according to an embodiment of the present disclosure, such as Figure 3 As shown, in a possible implementation, the embodiment of the present disclosure further provides a matching device 100 of the patent applicant, and the matching device includes: a candidate generation module 110, a similarity generation module 120, and a matching result generation module 130.

[0076] The candidate generation module 110 is used to retrieve the candidate applicant corresponding to the applicant in the preset database based on the characteristic information corresponding to the applicant; the similarity generation module 120 is used to determine the similarity between the candidate applicant and the applicant; the matching result generation module 130 is used to determine whether the applicant is successfully matched with the candidate applicant based on the similarity. Wherein, the candidate generation module 110 is used to retrieve the candidate applicant corresponding to the applicant to be stored in the preset database based on the characteristic information corresponding to the applicant to be stored in the patent document to be stored; the similarity generation module 120 is used to determine the similarity between the candidate applicant and the applicant to be stored; the matching result generation module 130 is used to determine whether the applicant to be stored is successfully matched with the candidate applicant based on the similarity; wherein the characteristic information includes: at least one of the address information corresponding to the applicant to be stored, the patent family number corresponding to the applicant to be stored, the classification information corresponding to the applicant to be stored, the inventor information corresponding to the applicant to be stored, the foreign language information corresponding to the applicant to be stored, the patent semantic information corresponding to the applicant to be stored, and the identification information corresponding to the applicant to be stored.

[0077] In a possible implementation, the matching device is also used to perform at least one of the following: divide the first address information corresponding to the patent document to be stored into second address information according to a hierarchical relationship, and use the region number corresponding to the second address information as the address information corresponding to the applicant to be stored; obtain the patent family number corresponding to the patent document to be stored as the patent family number corresponding to the applicant to be stored; obtain the patent classification number corresponding to the patent document to be stored as the classification information corresponding to the applicant to be stored; obtain the inventor information corresponding to the patent document to be stored as the inventor information corresponding to the applicant to be stored; convert the name of the applicant to be stored into a foreign applicant name, and normalize the foreign applicant name as the foreign information corresponding to the applicant to be stored; determine the patent semantic information corresponding to the applicant to be stored according to the semantic vector corresponding to the patent document to be stored; divide the name of the applicant to be stored into the identification information corresponding to the applicant to be stored, wherein the identification information includes: at least one of the applicant subject, field name, and company category.

[0078] In a possible implementation, based on the characteristic information corresponding to the applicant to be included in the patent document to be included, the candidate applicant corresponding to the applicant to be included is retrieved in a preset database, including: dividing the characteristic information corresponding to the applicant to be included into at least one query field, and querying the at least one query field in the existing index table in the preset database; and selecting the first preset number of first applicants with the highest matching degree in the preset database as the candidate applicant.

[0079] In a possible implementation, when it is determined that the applicant to be admitted is successfully matched, each query field is used as an index, and all query fields are used as entities corresponding to the index, and the index table is updated.

[0080] In a possible implementation, determining whether the applicant to be admitted has been successfully matched with the candidate applicant based on the similarity includes: taking the candidate applicant corresponding to the highest similarity as the result applicant who has been successfully matched with the applicant to be admitted, and determining that the applicant to be admitted has been successfully matched; or taking the candidate applicant corresponding to the highest similarity and the highest similarity being higher than a preset similarity threshold as the result applicant who has been successfully matched with the applicant to be admitted, and determining that the applicant to be admitted has been successfully matched.

[0081] In one possible implementation, determining the similarity between the candidate applicant and the applicant to be admitted includes: generating feature vectors corresponding to the candidate applicant and the applicant to be admitted based on feature information corresponding to the candidate applicant and the applicant to be admitted; and generating the similarity between the candidate applicant and the applicant to be admitted through a machine learning model based on the feature vectors corresponding to the candidate applicant and the applicant to be admitted.

[0082] In a possible implementation, the preset database is an Elasticsearch database.

[0083] In a possible implementation, the matching device is further used to execute: establishing a link between the applicant to be included in the database and the result applicant, and saving the patent document to be included corresponding to the applicant to be included in the database into the first database.

[0084] In a possible implementation, the matching device is further used to execute: obtaining characteristic information corresponding to the first applicant; using at least one characteristic information corresponding to the first applicant as an index and all characteristic information corresponding to the first applicant as entities corresponding to the index, to establish an index table in the preset database.

[0085] In a possible implementation, the obtaining of characteristic information corresponding to the first applicant includes at least one of the following: dividing the first address information corresponding to the first applicant into second address information according to a hierarchical relationship, and using the region code corresponding to the second address information as the address information corresponding to the first applicant; obtaining the patent family number corresponding to the first applicant in the first database as the patent family number corresponding to the first applicant; obtaining a second preset number of patent classification numbers corresponding to the first applicant in the first database as the classification information corresponding to the first applicant; obtaining a third preset number of inventor information corresponding to the first applicant in the first database as the inventor information corresponding to the first applicant; normalizing the foreign applicant name corresponding to the first applicant in the first database as the foreign information corresponding to the first applicant; determining the patent semantic information corresponding to the first applicant based on the semantic vector corresponding to the first applicant in the first database; and dividing the name of the first applicant into identification information corresponding to the first applicant, wherein the identification information includes at least one of the applicant entity, field name, and company category.

[0086] In some embodiments, the functions or modules included in the device provided by the embodiments of the present disclosure can be used to execute the method described in the above method embodiments. The specific implementation can refer to the description of the above method embodiments, and for the sake of brevity, it will not be repeated here.

[0087] The embodiment of the present disclosure also provides a computer-readable storage medium on which computer program instructions are stored, and when the computer program instructions are executed by a processor, the above method is implemented. The computer-readable storage medium may be a non-volatile computer-readable storage medium.

[0088] An embodiment of the present disclosure further provides an electronic device, comprising: a processor; and a memory for storing instructions executable by the processor; wherein the processor is configured to call the instructions stored in the memory to execute the above method.

[0089] The embodiments of the present disclosure also provide a computer program product, including a computer-readable code, or a non-volatile computer-readable storage medium carrying the computer-readable code. When the computer-readable code runs in a processor of an electronic device, the processor in the electronic device executes the above method.

[0090] The electronic device may be provided as a terminal, a server, or a device in other forms.

[0091] See also Figure 4 , Figure 4 A block diagram of an electronic device provided according to an embodiment of the present disclosure is shown.

[0092] For example, the electronic device 1900 may be provided as a terminal device or a server. Figure 4 , the electronic device 1900 includes a processing component 1922, which further includes one or more processors, and a memory resource represented by a memory 1932 for storing instructions executable by the processing component 1922, such as an application. The application stored in the memory 1932 may include one or more modules, each corresponding to a set of instructions. In addition, the processing component 1922 is configured to execute instructions to perform the above method.

[0093] The electronic device 1900 may also include a power supply component 1926 configured to perform power management of the electronic device 1900, a wired or wireless network interface 1950 configured to connect the electronic device 1900 to a network, and an input / output (I / O) interface 1958. The electronic device 1900 may operate based on an operating system stored in the memory 1932, such as a Microsoft Server operating system (Windows Server 2003). TM ), a graphical user interface operating system launched by Apple (Mac OSX TM ), a multi-user, multi-process computer operating system (Unix TM ), a free and open source Unix-like operating system (Linux TM), an open source Unix-like operating system (FreeBSD TM ) or similar.

[0094] In an exemplary embodiment, a non-volatile computer-readable storage medium is also provided, such as a memory 1932 including computer program instructions, which can be executed by the processing component 1922 of the electronic device 1900 to perform the above method.

[0095] The present disclosure may be a system, a method and / or a computer program product. The computer program product may include a computer-readable storage medium carrying computer-readable program instructions for causing a processor to implement various aspects of the present disclosure.

[0096] Computer readable storage medium can be a tangible device that can hold and store instructions used by an instruction execution device. Computer readable storage medium can be, for example, (but not limited to) an electrical storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any suitable combination thereof. More specific examples (non-exhaustive list) of computer readable storage medium include: a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), a static random access memory (SRAM), a portable compact disk read-only memory (CD-ROM), a digital versatile disk (DVD), a memory stick, a floppy disk, a mechanical encoding device, for example, a punch card or a convex structure in a groove on which instructions are stored, and any suitable combination thereof. The computer readable storage medium used here is not interpreted as a transient signal itself, such as a radio wave or other freely propagating electromagnetic wave, an electromagnetic wave propagated by a waveguide or other transmission medium (for example, a light pulse by an optical fiber cable), or an electrical signal transmitted by a wire.

[0097] The computer-readable program instructions described herein can be downloaded from a computer-readable storage medium to each computing / processing device, or downloaded to an external computer or external storage device via a network, such as the Internet, a local area network, a wide area network, and / or a wireless network. The network can include copper transmission cables, optical fiber transmissions, wireless transmissions, routers, firewalls, switches, gateway computers, and / or edge servers. The network adapter card or network interface in each computing / processing device receives the computer-readable program instructions from the network and forwards the computer-readable program instructions for storage in the computer-readable storage medium in each computing / processing device.

[0098] The computer program instructions for performing the operation of the present disclosure may be assembly instructions, instruction set architecture (ISA) instructions, machine instructions, machine-related instructions, microcode, firmware instructions, state setting data, or source code or object code written in any combination of one or more programming languages, including object-oriented programming languages, such as Smalltalk, C++, etc., and conventional procedural programming languages, such as "C" language or similar programming languages. Computer-readable program instructions may be executed completely on a user's computer, partially on a user's computer, as an independent software package, partially on a user's computer, partially on a remote computer, or completely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer via any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., using an Internet service provider to connect via the Internet). In some embodiments, an electronic circuit, such as a programmable logic circuit, a field programmable gate array (FPGA), or a programmable logic array (PLA), may be customized by utilizing the state information of the computer-readable program instructions, and the electronic circuit may execute the computer-readable program instructions, thereby realizing various aspects of the present disclosure.

[0099] Various aspects of the present disclosure are described herein with reference to the flowcharts and / or block diagrams of the methods, devices (systems) and computer program products according to the embodiments of the present disclosure. It should be understood that each box in the flowchart and / or block diagram and the combination of each box in the flowchart and / or block diagram can be implemented by computer-readable program instructions.

[0100] These computer-readable program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, thereby producing a machine, so that when these instructions are executed by the processor of the computer or other programmable data processing device, a device that implements the functions / actions specified in one or more boxes in the flowchart and / or block diagram is generated. These computer-readable program instructions can also be stored in a computer-readable storage medium, and these instructions cause the computer, programmable data processing device, and / or other equipment to work in a specific manner, so that the computer-readable medium storing the instructions includes a manufactured product, which includes instructions for implementing various aspects of the functions / actions specified in one or more boxes in the flowchart and / or block diagram.

[0101] Computer-readable program instructions may also be loaded onto a computer, other programmable data processing apparatus, or other device so that a series of operating steps are performed on the computer, other programmable data processing apparatus, or other device to produce a computer-implemented process, thereby causing the instructions executed on the computer, other programmable data processing apparatus, or other device to implement the functions / actions specified in one or more boxes in the flowchart and / or block diagram.

[0102] The flow chart and block diagram in the accompanying drawings show the possible architecture, function and operation of the system, method and computer program product according to multiple embodiments of the present disclosure. In this regard, each square box in the flow chart or block diagram can represent a part of a module, program segment or instruction, and a part of the module, program segment or instruction includes one or more executable instructions for realizing the specified logical function. In some alternative implementations, the function marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two continuous square boxes can actually be executed substantially in parallel, and they can sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs the specified function or action, or can be implemented with a combination of special hardware and computer instructions.

[0103] The computer program product may be implemented in hardware, software or a combination thereof. In one optional embodiment, the computer program product is embodied as a computer storage medium, and in another optional embodiment, the computer program product is embodied as a software product, such as a software development kit (SDK) and the like.

[0104] The flow chart and block diagram in the accompanying drawings show the possible architecture, function and operation of the system, method and computer program product according to multiple embodiments of the present disclosure. In this regard, each square box in the flow chart or block diagram can represent a part of a module, program segment or instruction, and a part of the module, program segment or instruction includes one or more executable instructions for realizing the specified logical function. In some alternative implementations, the function marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two continuous square boxes can actually be executed substantially in parallel, and they can sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs the specified function or action, or can be implemented with a combination of special hardware and computer instructions.

[0105] The embodiments of the present disclosure have been described above, and the above description is exemplary, not exhaustive, and is not limited to the disclosed embodiments. Many modifications and changes will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the described embodiments. The selection of terms used herein is intended to best explain the principles of the embodiments, practical applications, or improvements to the technology in the market, or to enable other persons of ordinary skill in the art to understand the embodiments disclosed herein.

Claims

1. A method for matching patent applicants, characterized in that: The matching method comprises: Based on the characteristic information corresponding to the applicant to be included in the patent document to be included, the candidate applicant corresponding to the applicant to be included is retrieved in the preset database; Determining the similarity between the candidate applicant and the applicant to be admitted; Based on the similarity, determine whether the applicant to be admitted successfully matches the candidate applicant; wherein the characteristic information includes: at least one of the address information corresponding to the applicant to be admitted, the patent family number corresponding to the applicant to be admitted, the classification information corresponding to the applicant to be admitted, the inventor information corresponding to the applicant to be admitted, the foreign language information corresponding to the applicant to be admitted, the patent semantic information corresponding to the applicant to be admitted, and the identification information corresponding to the applicant to be admitted; Wherein, the determining of the similarity between the candidate applicant and the applicant to be included in the database includes: Based on the feature information corresponding to the candidate applicants and the applicants to be admitted, generating feature vectors corresponding to the candidate applicants and the applicants to be admitted; Based on the feature vectors corresponding to the candidate applicants and the applicants to be admitted, the similarity between the candidate applicants and the applicants to be admitted is generated through a machine learning model.

2. The matching method according to claim 1, characterized in that: The matching method includes at least one of the following: Divide the first address information corresponding to the patent document to be stored into second address information according to a hierarchical relationship, and use the region number corresponding to the second address information as the address information corresponding to the applicant to be stored; Obtain the patent family number corresponding to the patent document to be stored as the patent family number corresponding to the applicant to be stored; Obtaining the patent classification number corresponding to the patent document to be stored as the classification information corresponding to the applicant to be stored; Obtaining the inventor information corresponding to the patent document to be stored as the inventor information corresponding to the applicant to be stored; Convert the name of the applicant to be added to the database into a foreign-language applicant name, and normalize the foreign-language applicant name as the foreign-language information corresponding to the applicant to be added to the database; Determine the patent semantic information corresponding to the applicant to be stored according to the semantic vector corresponding to the patent document to be stored; The names of the applicants to be admitted are classified as identification information corresponding to the applicants to be admitted, wherein the identification information includes at least one of the applicant subject, field name, and company category.

3. The matching method according to claim 1, characterized in that: The step of retrieving candidate applicants corresponding to the applicants to be included in the preset database based on the characteristic information corresponding to the applicants to be included in the patent documents to be included, comprises: Dividing the characteristic information corresponding to the applicant to be admitted into at least one query field, and searching the at least one query field in the existing index table in the preset database; A first preset number of first applicants with the highest matching degree in the preset database are selected as the candidate applicants.

4. The matching method according to claim 3, characterized in that: The matching method further comprises: When it is determined that the applicant to be admitted is matched successfully, each of the query fields is used as an index, and all of the query fields are used as entities corresponding to the indexes, and the index table is updated.

5. The matching method according to claim 1, characterized in that: The determining, based on the similarity, whether the applicant to be admitted successfully matches the candidate applicant includes: The candidate applicant corresponding to the highest similarity is taken as the result applicant who successfully matches the applicant to be admitted, and the applicant to be admitted is determined to be successfully matched; or the candidate applicant corresponding to the highest similarity and the highest similarity higher than a preset similarity threshold is taken as the result applicant who successfully matches the applicant to be admitted, and the applicant to be admitted is determined to be successfully matched.

6. The matching method according to any one of claims 1 to 5, characterized in that: The preset database is the Elasticsearch database.

7. The matching method according to any one of claims 1 to 6, characterized in that: The matching method further comprises: A link is established between the applicant to be included in the database and the result applicant, and the patent document to be included in the database corresponding to the applicant to be included in the database is saved in the first database.

8. The matching method according to any one of claims 1 to 7, characterized in that: The matching method further comprises: Obtaining characteristic information corresponding to the first applicant; At least one characteristic information corresponding to the first applicant is used as an index, and all characteristic information corresponding to the first applicant is used as an entity corresponding to the index, to establish an index table in the preset database.

9. The matching method according to claim 8, characterized in that: The obtaining of characteristic information corresponding to the first applicant includes at least one of the following: dividing the first address information corresponding to the first applicant into second address information according to a hierarchical relationship, and using the area code corresponding to the second address information as the address information corresponding to the first applicant; Obtain the patent family number corresponding to the first applicant in the first database as the patent family number corresponding to the first applicant; Obtaining a second preset number of patent classification numbers corresponding to the first applicant in the first database as classification information corresponding to the first applicant; Acquire information of a third preset number of inventors corresponding to the first applicant in the first database as the inventor information corresponding to the first applicant; Normalize the foreign-language applicant name corresponding to the first applicant in the first database as the foreign-language information corresponding to the first applicant; Determine patent semantic information corresponding to the first applicant based on the semantic vector corresponding to the first applicant in the first database; The name of the first applicant is classified as identification information corresponding to the first applicant, wherein the identification information includes at least one of the applicant entity, field name, and company category.

10. A matching device of the patent applicant, characterized in that: The matching device comprises: A candidate generation module is used to retrieve candidate applicants corresponding to the applicants to be stored in the preset database based on the characteristic information corresponding to the applicants to be stored in the patent documents to be stored; A similarity generation module, used to determine the similarity between the candidate applicant and the applicant to be admitted; A matching result generation module is used to determine whether the applicant to be admitted successfully matches the candidate applicant based on the similarity; wherein the characteristic information includes: at least one of the address information corresponding to the applicant to be admitted, the patent family number corresponding to the applicant to be admitted, the classification information corresponding to the applicant to be admitted, the inventor information corresponding to the applicant to be admitted, the foreign language information corresponding to the applicant to be admitted, the patent semantic information corresponding to the applicant to be admitted, and the identification information corresponding to the applicant to be admitted; Wherein, the determining of the similarity between the candidate applicant and the applicant to be included in the database includes: Based on the feature information corresponding to the candidate applicants and the applicants to be admitted, generating feature vectors corresponding to the candidate applicants and the applicants to be admitted; Based on the feature vectors corresponding to the candidate applicants and the applicants to be admitted, the similarity between the candidate applicants and the applicants to be admitted is generated through a machine learning model.

11. An electronic device, characterized in that: include: processor; a memory for storing processor-executable instructions; Wherein, the processor is configured to execute the matching method of the patent applicant as described in any one of claims 1 to 9.

12. A non-volatile computer-readable storage medium having computer program instructions stored thereon, characterized in that: When the computer program instructions are executed by a processor, the matching method of the patent applicant described in any one of claims 1 to 9 is implemented.

Images