Method and device for identifying persons with the same name, storage medium and electronic device

Abstract

The embodiment of the present invention discloses a method and device for identifying persons with duplicate names, as well as a storage medium and an electronic device. The method includes: for two target objects with duplicate names, respectively obtaining first feature data samples and second feature data samples of the two target objects, determining first feature values ​​of multiple first features according to the first feature data samples; and determining second feature values ​​of multiple second features according to the second feature data samples; finally, determining whether the persons with duplicate names of the two target objects are the same natural person according to the correlation values ​​of the two target objects determined by the multiple first feature values ​​and the multiple second feature values ​​and a preset correlation threshold. The embodiment of the present invention expands the recognition scope of persons with duplicate names, and by filtering the collected feature data and calculating the feature values ​​of multiple first features and second features to comprehensively judge the degree of correlation between the two target objects, the accuracy of identifying persons with duplicate names is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of data analysis, and more particularly, to a method and apparatus for identifying persons with the same name, a storage medium, and an electronic device. Background Art

[0002] In daily life and various business activities, people often need to query information about specific persons. However, since different natural persons may have the same name, it has become a rather troublesome problem for users to determine whether two natural persons with the same name among two target objects are the same natural person during the query process. Summary of the Invention

[0003] The problem to be solved by the present invention is how to determine whether the persons with the same name among two target objects are the same natural person for two target objects with persons having the same name. To solve the technical problem in the prior art that it is impossible to identify whether the persons with the same name among two target objects are the same natural person, embodiments of the present invention provide a method and apparatus for identifying persons with the same name, as well as a storage medium and an electronic device.

[0004] According to one aspect of an embodiment of the present invention, a method for identifying persons with the same name is provided, the method comprising:

[0005] For two target objects with persons having the same name, respectively obtain a first feature data sample and a second feature data sample of the two target objects, wherein the first feature data sample includes a plurality of first features, the second feature data sample includes a plurality of second features, the first features are features for determining the correlation relationship between the target objects, and the second features are features for determining the similarity relationship between the target objects;

[0006] Determine a first feature value of the plurality of first features according to the first feature data sample of the two target objects; and determine a second feature value of the plurality of second features according to the second feature data sample of the two target objects;

[0007] Determine an association degree value of the two target objects according to the plurality of first feature values and the plurality of second feature values;

[0008] Determine whether the persons with the same name of the two target objects are the same natural person according to the association degree value and a preset association degree threshold.

[0009] Optionally, in the above method embodiments of the present invention, for two target objects with persons having the same name, respectively obtaining a first feature data sample and a second feature data sample of the two target objects includes:

[0010] Respectively obtain initial first feature data and initial second feature data of the two target objects;

[0011] Filter the feature data of the first feature to be filtered in the initial first feature data based on a preset first feature blacklist to generate a first feature data sample;

[0012] Filter the feature data of the second feature to be filtered in the initial second feature data based on a preset second feature blacklist to generate a second feature data sample.

[0013] Optionally, in the above method embodiments of the present invention, determining the first feature values of multiple first features according to the first feature data samples of two target objects includes:

[0014] Match the first feature information items of each first feature in the first feature data samples of the two target objects. When the first feature information items of the first feature do not match, confirm that the feature value of the first feature is 0; when the first feature information items of the first feature match, calculate the first feature value of the first feature according to the first feature occurrence times item corresponding to the first feature information item and a preset first feature occurrence times threshold.

[0015] Optionally, in the above method embodiments of the present invention, the calculation formula for calculating the first feature value of the first feature according to the first feature occurrence times item corresponding to the first feature information item and a preset first feature occurrence times threshold is specifically:

[0016] r = (y - x) / y

[0017] In the formula, y is the preset first feature occurrence times threshold, x is the first feature occurrence times item; r is the first feature value of the first feature.

[0018] Optionally, in the above method embodiments of the present invention, the method further includes that when each first feature in the first feature data sample includes multiple first feature information items and each first feature information item matches, calculating the first feature value of the first feature according to the first feature occurrence times item corresponding to each first feature information item, specifically:

[0019] Calculate the feature values corresponding to each first feature information item according to the first feature occurrence times corresponding to each first feature information item of the first feature and a preset first feature occurrence times threshold, and use the sum of the feature values corresponding to multiple first feature information items as the first feature value of the first feature.

[0020] Optionally, in the above method embodiments of the present invention, determining the second feature values of multiple second features according to the second feature data samples of two target objects includes:

[0021] Calculate the text similarity values of each second feature in the second feature data samples of the two target objects;

[0022] Take the text similarity value of each second feature as the second feature value corresponding to the second feature.

[0023] Optionally, in the above method embodiments of the present invention, determining the second feature values of multiple second features according to the second feature data samples of two target objects includes:

[0024] Extract the text of each second feature in the second feature data samples of the two target objects respectively according to a preset rule to obtain the core text of each second feature of the two target objects;

[0025] Calculate the text similarity value of the core text of each second feature of the two target objects;

[0026] Take the text similarity value of the core text of each second feature as the second feature value corresponding to the second feature.

[0027] Optionally, in the above method embodiments of the present invention, determining the association degree value of two target objects according to multiple first feature values and multiple second feature values includes:

[0028] Sum the first feature values of multiple first features to obtain the first feature association degree value;

[0029] Sum the second feature values of multiple second features to obtain the second feature association degree value;

[0030] Sum the first feature association degree value and the second feature association degree value to obtain the association degree value of the two target objects.

[0031] Optionally, in the above method embodiments of the present invention, determining the association degree value of two target objects according to multiple first feature values and multiple second feature values includes:

[0032] Set the weight coefficients of each first feature and each second feature;

[0033] Multiply the first feature value of each first feature by its corresponding weight coefficient to obtain the first weighted feature value of each first feature;

[0034] Sum the first weighted feature values of multiple first features to obtain the first feature association degree value;

[0035] Multiply the second feature value of each second feature by its corresponding weight coefficient to obtain the second weighted feature value of each second feature;

[0036] Sum the second weighted feature values of multiple second features to obtain the second feature association degree value;

[0037] Sum the first feature association degree value and the second feature association degree value to obtain the association degree value of the two target objects.

[0038] Optionally, in the above method embodiments of the present invention, determining the correlation value of two target objects according to a plurality of first eigenvalue and a plurality of second eigenvalues includes:

[0039] Inputting the plurality of first eigenvalues and the plurality of second eigenvalues into a target object correlation model to obtain the correlation value of the two target objects; wherein, the target object correlation model is a model for determining the degree of correlation between target objects, which takes the first eigenvalues and the second eigenvalues of known target objects with the same name as input and the correlation value between the target objects as output, and is trained by using a machine learning algorithm.

[0040] Optionally, in the above method embodiments of the present invention, determining whether the persons with the same name of the two target objects are the same natural person according to the correlation value and a preset correlation threshold includes:

[0041] When the correlation value is not less than the preset correlation threshold, determining that the persons with the same name of the two target objects are the same natural person;

[0042] When the correlation value is less than the preset correlation threshold, determining that the persons with the same name of the two target objects are not the same natural person.

[0043] According to another aspect of the embodiments of the present invention, the present invention provides a device for identifying persons with the same name, the device includes:

[0044] A data sample module, configured to respectively obtain a first feature data sample and a second feature data sample of two target objects for two target objects with the same name, wherein the first feature data sample includes a plurality of first features, the second feature data sample includes a plurality of second features, the first feature is a feature for determining the correlation relationship between target objects, and the second feature is a feature for determining the similarity relationship between target objects;

[0045] An eigenvalue module, configured to determine a first eigenvalue of a plurality of first features according to the first feature data sample of the two target objects; and determine a second eigenvalue of a plurality of second features according to the second feature data sample of the two target objects;

[0046] A correlation value module, configured to determine the correlation value of the two target objects according to the plurality of first eigenvalues and the plurality of second eigenvalues;

[0047] A result determination module, configured to determine whether the persons with the same name of the two target objects are the same natural person according to the correlation value and a preset correlation threshold.

[0048] Optionally, in the above device embodiments of the present invention, the data sample module includes:

[0049] A data acquisition unit for respectively obtaining initial first feature data and initial second feature data of two target objects;

[0050] A first filtering unit for filtering the feature data of the first features to be filtered in the initial first feature data based on a preset first feature blacklist to generate a first feature data sample;

[0051] A second filtering unit for filtering the feature data of the second features to be filtered in the initial second feature data based on a preset second feature blacklist to generate a second feature data sample.

[0052] Optionally, in the above device embodiments of the present invention, the eigenvalue module is used to determine the first eigenvalues of multiple first features according to the first feature data samples of two target objects, including:

[0053] Matching each first feature information item of the first feature data samples of the two target objects, and when the first feature information items of the first feature do not match, confirming that the first eigenvalue of the first feature is 0; when the first feature information items of the first feature match, calculating the eigenvalue of the first feature according to the first feature occurrence times item corresponding to the first feature information item of the first feature and a preset first feature occurrence times threshold.

[0054] Optionally, in the above device embodiments of the present invention, the specific formula for the eigenvalue module to calculate the first eigenvalue of the first feature according to the first feature occurrence times item corresponding to the first feature information item of the first feature and a preset first feature occurrence times threshold is:

[0055] r = (y - x) / y

[0056] In the formula, y is the preset first feature occurrence times threshold, x is the first feature occurrence times item; r is the first eigenvalue of the first feature.

[0057] Optionally, in the above device embodiments of the present invention, the eigenvalue module is further used to, when each first feature in the first feature data sample includes multiple first feature information items and each first feature information item matches, calculate the first eigenvalue of the first feature according to the first feature occurrence times item corresponding to each first feature information item, specifically:

[0058] Calculating the eigenvalue corresponding to each first feature information item according to the first feature occurrence times item corresponding to each first feature information item of the first feature and a preset first feature occurrence times threshold, and taking the sum of the eigenvalues corresponding to multiple first feature information items as the first eigenvalue of the first feature.

[0059] Optionally, in the above device embodiments of the present invention, the eigenvalue module determining the second eigenvalue of multiple second features based on the second feature data samples of two target objects includes:

[0060] Calculating the text similarity value of each second feature in the second feature data samples of two target objects;

[0061] Taking the text similarity value of each second feature as the second eigenvalue of the corresponding second feature.

[0062] Optionally, in the above device embodiments of the present invention, the eigenvalue module determining the second eigenvalue of multiple second features based on the second feature data samples of two target objects includes:

[0063] Respectively extracting the text of each second feature in the second feature data samples of two target objects according to a preset rule to obtain the core text of each second feature of the two target objects;

[0064] Calculating the text similarity value of the core text of each second feature of the two target objects;

[0065] Taking the text similarity value of the core text of each second feature as the second eigenvalue of the corresponding second feature.

[0066] Optionally, in the above device embodiments of the present invention, the correlation value module determining the correlation value of two target objects based on multiple first eigenvalues and multiple second eigenvalues includes:

[0067] Summing the first eigenvalues of multiple first features to obtain the first feature correlation value;

[0068] Summing the second eigenvalues of multiple second features to obtain the second feature correlation value;

[0069] Summing the first feature correlation value and the second feature correlation value to obtain the correlation value of two target objects.

[0070] Optionally, in the above device embodiments of the present invention, the correlation value module determining the correlation value of two target objects based on multiple first eigenvalues and multiple second eigenvalues includes:

[0071] Setting the weight coefficients of each first feature and each second feature;

[0072] Multiplying the first eigenvalue of each first feature by its corresponding weight coefficient to obtain the first weighted feature value of each first feature;

[0073] Summing the first weighted feature values of multiple first features to obtain the first feature correlation value;

[0074] Multiply the second eigenvalue of each second feature by its corresponding weight coefficient to obtain the second weighted eigenvalue of each second feature;

[0075] Sum the second weighted eigenvalues of multiple second features to obtain the second feature correlation value;

[0076] Sum the first feature correlation value and the second feature correlation value to obtain the correlation value of the two target objects.

[0077] Optionally, in each of the above device embodiments of the present invention, the correlation value module determines the correlation value of two target objects according to multiple first feature values and multiple second feature values, including:

[0078] Input multiple first feature values and multiple second feature values into the target object correlation model to obtain the correlation value of the two target objects; wherein, the target object correlation model is a model for determining the degree of association between target objects, which is trained by a machine learning algorithm with the first feature values and second feature values of known target objects with the same name as input and the correlation value between target objects as output.

[0079] Optionally, in each of the above device embodiments of the present invention, the result determination module determines whether the persons with the same name of the two target objects are the same natural person according to the correlation value and a preset correlation threshold, including:

[0080] When the correlation value is not less than the preset correlation threshold, determine that the persons with the same name of the two target objects are the same natural person;

[0081] When the correlation value is less than the preset correlation threshold, determine that the persons with the same name of the two target objects are not the same natural person.

[0082] According to another aspect of the embodiments of the present invention, the present invention provides a computer-readable storage medium, which stores a computer program for executing the method according to any one of the above embodiments of the present invention.

[0083] According to another aspect of the embodiments of the present invention, the present invention provides an electronic device, which includes:

[0084] A processor;

[0085] A memory for storing executable instructions of the processor;

[0086] The processor is configured to read the executable instructions from the memory and execute the executable instructions to implement the method according to any one of the above embodiments of the present invention.

[0087] In one aspect of the embodiments of the present invention, by comprehensively considering the characteristics of the relevant relationship and similarity relationship between two target objects with the same name, the identification scope of the people with the same name is expanded. On the other hand, by filtering the collected feature data, calculating the eigenvalues of multiple first features and the eigenvalues of second features, and comprehensively judging the degree of association between the two target objects, the possibility of misassociating the two target objects is reduced, thereby better avoiding identifying two natural persons with the same name as the same natural person and improving the accuracy of identifying people with the same name.

[0088] The technical solutions of the present invention will be further described in detail below with reference to the drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0089] By referring to the following drawings, the exemplary embodiments of the present invention can be more completely understood:

[0090] Figure 1 It is a flowchart of a method for identifying people with the same name provided by an exemplary embodiment of the present invention;

[0091] Figure 2 It is a schematic structural diagram of a device for identifying people with the same name provided by an exemplary embodiment of the present invention;

[0092] Figure 3 It is a schematic structural diagram of a data sample module provided by an exemplary embodiment of the present invention;

[0093] Figure 4 It is a schematic structural diagram of an electronic device provided by an exemplary embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0094] Next, the exemplary embodiments of the present invention will be described in detail with reference to the drawings. Obviously, the described exemplary embodiments are only a part of the embodiments of the present invention, rather than all the embodiments of the present invention. It should be understood that the present invention is not limited by the exemplary embodiments described herein.

[0095] It should be noted that: unless otherwise specifically stated, the relative arrangements, numerical expressions and numerical values of the components and steps set forth in these exemplary embodiments do not limit the scope of the present invention.

[0096] Those skilled in the art can understand that the terms "first", "second", etc. in the exemplary embodiments of the present invention are only used to distinguish different steps, devices or modules, etc., and neither represent any specific technical meaning nor indicate an inevitable logical order between them.

[0097] It should also be understood that in the exemplary embodiments of the present invention, "a plurality" may refer to two or more, and "at least one" may refer to one, two or more.

[0098] It should also be understood that for any component, data, or structure mentioned in the exemplary embodiments of the present invention, in the absence of explicit definition or contrary indication in the context, it is generally understood as one or more.

[0099] In addition, the term "and / or" in the present invention is merely a description of the association relationship between associated objects, indicating that three relationships can exist. For example, A and / or B can represent three situations: A exists alone, A and B exist simultaneously, and B exists alone. In addition, the character " / " in the present invention generally represents an "or" relationship between the associated objects before and after.

[0100] It should also be understood that the present invention emphasizes the differences between the various exemplary embodiments, and their similarities can be referred to each other. For the sake of brevity, they will not be elaborated one by one.

[0101] At the same time, it should be understood that for the convenience of description, the dimensions of the various parts shown in the drawings are not drawn in actual proportional relationship.

[0102] The following description of at least one exemplary embodiment is actually merely illustrative and in no way restricts the present invention and its application or use.

[0103] Well-known technologies, methods, and devices for those of ordinary skill in the relevant art may not be discussed in detail, but in appropriate cases, the technologies, methods, and devices should be regarded as part of the specification.

[0104] It should be noted that similar reference numerals and letters represent similar items in the following drawings. Therefore, once an item is defined in one drawing, it does not need to be further discussed in subsequent drawings.

[0105] The exemplary embodiments of the present invention can be applied to electronic devices such as terminal devices, computer systems, servers, etc., which can operate together with many other general-purpose or special-purpose computing system environments or configurations. Examples of well-known terminal devices, computing systems, environments, and / or configurations suitable for use with electronic devices such as terminal devices, computer systems, servers, etc. include, but are not limited to: personal computer systems, server computer systems, thin clients, thick clients, handheld or laptop devices, microprocessor-based systems, set-top boxes, programmable consumer electronics, network personal computers, minicomputer systems, mainframe computer systems, and distributed cloud computing technology environments including any of the above systems, and so on.

[0106] Electronic devices such as terminal devices, computer systems, and servers can be described in the general context of computer system-executable instructions (such as program modules) executed by a computer system. Generally, program modules can include routines, programs, target programs, components, logics, data structures, etc., which perform specific tasks or implement specific abstract data types. The computer system / server can be implemented in a distributed cloud computing environment where tasks are executed by remote processing devices linked through a communication network. In a distributed cloud computing environment, program modules can be located on local or remote computing system storage media including storage devices.

[0107] Exemplary method

[0108] Figure 1 The flowchart of the method for identifying persons with the same name provided for an exemplary embodiment of the present invention can be applied to an electronic device. As Figure 1 shown, the method 100 for identifying persons with the same name described in this exemplary embodiment starts from step 101.

[0109] In step 101, for two target objects with persons having the same name, respectively obtain a first feature data sample and a second feature data sample of the two target objects. Among them, the first feature data sample includes multiple first features, the second feature data sample includes multiple second features, the first feature is a feature for determining the correlation relationship between the target objects, and the second feature is a feature for determining the similarity relationship between the target objects.

[0110] In daily life and business activities, people often need to query certain specific persons. In one embodiment, taking two companies with persons having the same name as target objects, the phone numbers, email addresses, first addresses, other persons with the same name, and investment relationships of the two companies are used as the first features for determining the correlation relationship between the two companies, and the company types, company names, industries to which they belong, business scopes, and second addresses of the two companies are used as the second features for determining the similarity relationship between the two companies. Among them, the first address is the country and province where the company is registered, and the second address is the detailed address of the company's registration place.

[0111] Optionally, for two target objects with persons having the same name, respectively obtaining a first feature data sample and a second feature data sample of the two target objects includes:

[0112] Respectively obtain the initial first feature data and the initial second feature data of the two target objects;

[0113] Based on a preset first feature blacklist, filter the feature data of the first features to be filtered in the initial first feature data to generate a first feature data sample;

[0114] Filter the feature data of the second feature to be filtered in the initial second feature data based on the pre-set second feature blacklist to generate a second feature data sample.

[0115] In one embodiment, after collecting the data of the first features such as the email addresses, phone numbers, first addresses, and other persons with the same name of two companies, as well as the data of the second features such as the company types, names, industries, business scopes, and second addresses of the two companies, the original first feature data and the original second feature data are obtained.

[0116] Since in the company registration process, sometimes some pseudo-registration information is used, such as using common numbers to replace phone information, such as "123456", "111111", etc., these pseudo-registration information can be generated into a blacklist; there are also some registration information that is registered in groups, and the number of registrations involved is very large. By counting the number of registrations to determine whether to add it to the blacklist. For example, when performing data statistics, it is found that more than 100 companies have registered the same phone number, then it is considered that this phone number is group registration information and should be added to the blacklist. By statistically analyzing the first feature and the second feature respectively, a first feature blacklist and a second feature blacklist are generated. Filter the collected original first feature data through the first feature blacklist to obtain a first feature data sample, and filter the collected original second feature data through the second feature blacklist to obtain a second feature data sample.

[0117] In step 102, determine the first feature values of multiple first features according to the first feature data samples of the two target objects; and determine the second feature values of multiple second features according to the second feature data samples of the two target objects.

[0118] Optionally, determining the first feature values of multiple first features according to the first feature data samples of the two target objects includes:

[0119] Match the first feature information items of each first feature in the first feature data samples of the two target objects. When the first feature information items of the first feature do not match, determine that the first feature value of the first feature is 0; when the first feature information items of the first feature match, calculate the first feature value of the first feature according to the first feature occurrence times item corresponding to the first feature information item and the preset first feature occurrence times threshold.

[0120] Optionally, the calculation formula for calculating the first feature value of the first feature according to the first feature occurrence times item corresponding to the first feature information item and the preset first feature occurrence times threshold is specifically:

[0121] r = (y - x) / y

[0122] Where y is a preset threshold for the number of occurrences of the first feature, x is the term for the number of occurrences of the first feature; r is the first feature value of the first feature.

[0123] In one embodiment, the first feature includes a first feature information item and a first feature occurrence number item. For example, when the first feature is a phone, the feature data of the phone includes "12345678" and "20", where the former is the information item and the latter is the occurrence number item. Match the phone numbers of the first feature phone in the first feature data samples of the two companies after filtering. When the phones are different, determine that the feature value of the first feature phone is 0. When the phone numbers are the same, then calculate the feature value of the first feature phone by taking the number of occurrences of the phone number of any one company. Since the number of occurrences of the phone number item is 20 and the set threshold for the number of occurrences is 100, then according to the calculation formula

[0124] it can be known that r = (100 - 20) / 100 = 0.8, so the feature value of the first feature phone of the two companies is 0.8.

[0125] Optionally, the method further includes that when each first feature in the first feature data sample includes multiple first feature information items and each first feature information item matches, calculate the first feature value of the first feature according to the first feature occurrence number item corresponding to each first feature information item. Specifically:

[0126] Calculate the feature value corresponding to each first feature information item according to the first feature occurrence number item corresponding to each first feature information item of the first feature and the preset threshold for the number of occurrences of the first feature, and take the sum of the feature values corresponding to the multiple first feature information items as the first feature value of the first feature.

[0127] In one embodiment, when the first feature phone of the two companies has 2 phone numbers and both phone numbers match, then calculate the feature values corresponding to the phone numbers respectively for these two phone numbers. When it is calculated that r 1 = 0.8, r 2 = 0.9, then the feature value r corresponding to the first feature phone is 0.8 + 0.9 = 1.7.

[0128] Optionally, determining the second feature values of multiple second features according to the second feature data samples of two target objects includes:

[0129] Calculate the text similarity value of each second feature in the second feature data samples of the two target objects;

[0130] Take the text similarity value of each second feature as the second feature value corresponding to the second feature.

[0131] In one embodiment, for the second feature, i.e., the company name, text similarity can be used to calculate the similarity of the second feature (company name) of two companies. Specifically, text matching can be performed on the feature data of the company names of the two companies, and the feature value of the second feature (company name) is determined by dividing the sum of the lengths of the matching texts by the sum of the lengths of the texts of the company names of the two companies. For example, if the company name of one company is Beijing Tianyancha Technology Co., Ltd. and the company name of another company is Haikou Tianyancha Technology Co., Ltd., when these two company names are matched, the length of the matching text "Tianyancha Technology Co., Ltd." is 9, so the sum of the lengths of the matching texts is 9 * 2 = 18, and the sum of the lengths of the texts of the company names is 11 * 2 = 22. Then the text similarity value of the company names is 18 / 22 = 0.82, and this text similarity value 0.82 is the feature value of the second feature (company name) of the two companies.

[0132] Optionally, determining the second feature values of multiple second features according to the second feature data samples of two target objects includes:

[0133] Respectively extract texts for each second feature in the second feature data samples of the two target objects according to a preset rule to obtain the core texts of each second feature of the two target objects;

[0134] Calculate the text similarity values of the core texts of each second feature of the two target objects;

[0135] Use the text similarity value of the core text of each second feature as the second feature value corresponding to the second feature.

[0136] In another embodiment, for the second feature (company name) of two companies, their feature data are respectively Beijing Tianyancha Technology Co., Ltd. and Haikou Tianyancha Technology Co., Ltd. According to the text extraction rule, after removing the address information "Beijing, Haikou" in front of the company name and the registration type information "Co., Ltd., Co., Ltd." behind the company name respectively, the core texts of the company names "Tianyancha Technology, Tianyancha Technology" are obtained. Calculate the text similarity value for the core texts of the company names "Tianyancha Technology" and "Tianyancha Technology". Since the texts are exactly the same, it can be known that the similarity value of the core texts of the company names is 1, that is, the feature value of the second feature (company name) determined according to the core text of the company name is 1. Using the core text to calculate the feature value of the second feature can discard other interference factors and more accurately judge the similarity of the second feature of the target object, so as to more accurately judge the degree of association of the target object.

[0137] In step 103, determine the association degree value of the two target objects according to the multiple first feature values and the multiple second feature values.

[0138] Optionally, determining the correlation degree value of two target objects according to a plurality of first eigenvalue and a plurality of second eigenvalues includes:

[0139] Summing the first eigenvalue of a plurality of first features to obtain a first feature correlation degree value;

[0140] Summing the second eigenvalue of a plurality of second features to obtain a second feature correlation degree value;

[0141] Summing the first feature correlation degree value and the second feature correlation degree value to obtain the correlation degree value of two target objects.

[0142] In one embodiment, when two companies have i first eigenvalues r 1 to r i , and j second eigenvalues s 1 to s j , the correlation degree value z of the two companies is obtained by summing the i first eigenvalues and the j second eigenvalues. The calculation formula is:

[0143] z = r 1 + r 2 + … + r i + s 1 + s 2 + … + s j .

[0144] Optionally, determining the correlation degree value of two target objects according to a plurality of first eigenvalue and a plurality of second eigenvalues includes:

[0145] Setting the weight coefficients of each first feature and each second feature;

[0146] Multiplying the first eigenvalue of each first feature by its corresponding weight coefficient to obtain the first weighted eigenvalue of each first feature;

[0147] Summing the first weighted eigenvalues of a plurality of first features to obtain a first feature correlation degree value;

[0148] Multiplying the second eigenvalue of each second feature by its corresponding weight coefficient to obtain the second weighted eigenvalue of each second feature;

[0149] Summing the second weighted eigenvalues of a plurality of second features to obtain a second feature correlation degree value;

[0150] Summing the first feature correlation degree value and the second feature correlation degree value to obtain the correlation degree value of two target objects.

[0151] In another embodiment, different weight coefficients can be set for each first eigenvalue and each second eigenvalue of two companies. At this time, the calculation formula for calculating the correlation degree value z of two companies is:

[0152] z = r 1 * w 1 + r 2 * w 2 + … + r i * w i + s 1 * w i+1 + s 2 * w i+2 + … + s j * w n

[0153] Wherein, w is the weight coefficient, and n = i + j.

[0154] According to the influence of different features on the correlation degree between two companies, different weight coefficients are set for different feature values, and the weighted sum of the first eigenvalue and the second eigenvalue of the two companies is calculated to determine the correlation degree value of a personal company. This method is more accurate than directly summing the first eigenvalue and the second eigenvalue of the two companies to obtain the correlation degree between the two companies, and can more effectively improve the accuracy of identifying people with the same name.

[0155] Optionally, determining the correlation degree value of two target objects according to multiple first eigenvalues and multiple second eigenvalues includes:

[0156] Inputting multiple first eigenvalues and multiple second eigenvalues into the target object correlation degree model to obtain the correlation degree value of the two target objects; wherein, the target object correlation degree model is a model for determining the correlation degree between target objects, which is trained by a machine learning algorithm with the first eigenvalue and the second eigenvalue of known target objects with people having the same name as the input and the correlation degree value between the target objects as the output.

[0157] In one embodiment, before identifying people with the same name for two target objects with people having the same name, a model for determining the correlation degree between target objects is first trained by a machine learning algorithm with the first eigenvalue and the second eigenvalue of known target objects with people having the same name as the input and the correlation degree value between the target objects as the output. The machine learning algorithm can be a neural network algorithm, a support vector machine algorithm, a decision tree algorithm, etc. By using the target object correlation degree model to judge the correlation degree of the two target objects, the target object correlation degree model can be continuously iteratively optimized by using the known eigenvalue and correlation degree value of the target object, so that the judgment of the correlation degree between the two target objects by the target object correlation degree model is more and more accurate, effectively improving the accuracy of identifying people with the same name.

[0158] In step 104, it is determined whether the people with the same name of the two target objects are the same natural person according to the correlation degree value and a preset correlation degree threshold.

[0159] Optionally, determining whether the persons with the same name of two target objects are the same natural person according to the relevance value and a preset relevance threshold includes:

[0160] When the relevance value is not less than the preset relevance threshold, determining that the persons with the same name of the two target objects are the same natural person;

[0161] When the relevance value is less than the preset relevance threshold, determining that the persons with the same name of the two target objects are not the same natural person.

[0162] Compared with the prior art, for two target objects with persons having the same name, this embodiment not only includes the feature that there is a correlation relationship between the target objects, but also includes the feature that there is a similarity relationship, effectively expanding the identification range of persons with the same name. Moreover, for the collected feature data, false registration information and group registration information are filtered, and the feature values of multiple first features and the feature values of second features are calculated according to the filtered feature data, and the correlation degree of the two target objects is comprehensively judged, reducing the possibility of misassociating the two target objects, thereby better avoiding identifying two natural persons with the same name as the same natural person and improving the accuracy of identifying persons with the same name.

[0163] Exemplary apparatus

[0164] Figure 2 is a schematic structural diagram of a device for identifying persons with the same name provided by an exemplary embodiment of the present invention. As Figure 2 shown, the device for identifying persons with the same name in this embodiment includes:

[0165] A data sample module 201, configured to, for two target objects with persons having the same name, respectively obtain a first feature data sample and a second feature data sample of the two target objects, where the first feature data sample includes multiple first features, the second feature data sample includes multiple second features, the first feature is a feature for determining the correlation relationship between the target objects, and the second feature is a feature for determining the similarity relationship between the target objects;

[0166] A feature value module 202, configured to determine a first feature value of multiple first features according to the first feature data sample of the two target objects; and determine a second feature value of multiple second features according to the second feature data sample of the two target objects;

[0167] A relevance value module 203, configured to determine a relevance value of the two target objects according to multiple first feature values and multiple second feature values;

[0168] A result determination module 204, configured to determine whether the persons with the same name of the two target objects are the same natural person according to the relevance value and a preset relevance threshold.

[0169] Figure 3 This is a schematic structural diagram of a data sample module provided by an exemplary embodiment of the present invention. As Figure 3 shown, the data sample module 201 includes:

[0170] A data acquisition unit 211, configured to respectively obtain initial first feature data and initial second feature data of two target objects;

[0171] A first filtering unit 212, configured to filter the feature data of the first feature to be filtered in the initial first feature data based on a preset first feature blacklist, and generate a first feature data sample;

[0172] A second filtering unit 213, configured to filter the feature data of the second feature to be filtered in the initial second feature data based on a preset second feature blacklist, and generate a second feature data sample.

[0173] Optionally, the feature value module 202 is configured to determine first feature values of multiple first features according to the first feature data samples of two target objects, including:

[0174] Matching each first feature information item in the first feature data samples of the two target objects. When the first feature information items of the first feature do not match, determining the first feature value of the first feature to be 0; when the first feature information items of the first feature match, calculating the first feature value of the first feature according to the first feature occurrence times item corresponding to the first feature information item and a preset first feature occurrence times threshold.

[0175] Optionally, the formula for the feature value module 202 to calculate the first feature value of the first feature according to the first feature occurrence times item corresponding to the first feature information item and a preset first feature occurrence times threshold is specifically:

[0176] r = (y - x) / y

[0177] In the formula, y is the preset first feature occurrence times threshold, x is the first feature occurrence times item; r is the first feature value of the first feature.

[0178] Optionally, when each first feature in the first feature data sample includes multiple first feature information items and each first feature information item matches, the feature value module 202 is further configured to calculate the first feature value of the first feature according to the first feature occurrence times item corresponding to each first feature information item, specifically:

[0179] Calculate the feature value corresponding to each first feature information item according to the number of occurrences of the first feature corresponding to each first feature information item of the first feature and a preset first feature occurrence threshold, and use the sum of the feature values corresponding to multiple first feature information items as the first feature value of the first feature.

[0180] Optionally, the feature value module 202 determines the second feature values of multiple second features according to the second feature data samples of two target objects, including:

[0181] Calculate the text similarity value of each second feature in the second feature data samples of two target objects;

[0182] Use the text similarity value of each second feature as the second feature value of the corresponding second feature.

[0183] Optionally, the feature value module 202 determines the second feature values of multiple second features according to the second feature data samples of two target objects, including:

[0184] Extract the text of each second feature in the second feature data samples of two target objects respectively according to a preset rule to obtain the core text of each second feature of the two target objects;

[0185] Calculate the text similarity value of the core text of each second feature of the two target objects;

[0186] Use the text similarity value of the core text of each second feature as the second feature value of the corresponding second feature.

[0187] Optionally, the correlation value module 203 determines the correlation value of two target objects according to multiple first feature values and multiple second feature values, including:

[0188] Sum the first feature values of multiple first features to obtain the first feature correlation value;

[0189] Sum the second feature values of multiple second features to obtain the second feature correlation value;

[0190] Sum the first feature correlation value and the second feature correlation value to obtain the correlation value of two target objects.

[0191] Optionally, the correlation value module 203 determines the correlation value of two target objects according to multiple first feature values and multiple second feature values, including:

[0192] Set the weight coefficients of each first feature and each second feature;

[0193] Multiply the first feature value of each first feature by its corresponding weight coefficient to obtain the first weighted feature value of each first feature;

[0194] Sum the first weighted eigenvalue of multiple first features to obtain the first feature correlation degree value;

[0195] Multiply the second eigenvalue of each second feature by its corresponding weight coefficient to obtain the second weighted eigenvalue of each second feature;

[0196] Sum the second weighted eigenvalue of multiple second features to obtain the second feature correlation degree value;

[0197] Sum the first feature correlation degree value and the second feature correlation degree value to obtain the correlation degree value of the two target objects.

[0198] Optionally, the correlation degree value module 203 determines the correlation degree value of the two target objects according to multiple first eigenvalues and multiple second eigenvalues, including:

[0199] Input multiple first eigenvalues and multiple second eigenvalues into the target object correlation degree model to obtain the correlation degree value of the two target objects; wherein, the target object correlation degree model takes the first eigenvalue and the second eigenvalue of the target object with known persons having the same name as input, and takes the correlation degree value between the target objects as output, and is a model for determining the correlation degree between target objects trained by using a machine learning algorithm.

[0200] Optionally, the result determination module 204 determines whether the persons with the same name of the two target objects are the same natural person according to the correlation degree value and a preset correlation degree threshold, including:

[0201] When the correlation degree value is not less than the preset correlation degree threshold, determine that the persons with the same name of the two target objects are the same natural person;

[0202] When the correlation degree value is less than the preset correlation degree threshold, determine that the persons with the same name of the two target objects are not the same natural person.

[0203] For two target objects with persons having the same name, the device for identifying persons having the same name provided in this embodiment, after respectively obtaining the first feature data sample and the second feature data sample of the two target objects, calculates the first eigenvalue of multiple first features and the second eigenvalue of multiple second features, and then determines the correlation degree between the two target objects according to the multiple first eigenvalues and multiple second eigenvalues to determine whether the persons having the same name are the same natural person. The steps are the same as those of the method for identifying persons having the same name provided in this embodiment, and the achieved technical effects are also the same, which will not be elaborated here.

[0204] Exemplary electronic device

[0205] Figure 4It is a schematic structural diagram of an electronic device provided by an exemplary embodiment of the present invention. The electronic device can be any one or both of the first device and the second device, or a stand-alone device independent of them, and the stand-alone device can communicate with the first device and the second device to receive the input signals collected from them. Figure 4 The block diagram of the electronic device according to an embodiment of the present disclosure is illustrated. As Figure 4 shown, the electronic device includes one or more processors 401 and a memory 402.

[0206] The processor 401 can be a central processing unit (CPU) or other forms of processing units with data processing capabilities and / or instruction execution capabilities, and can control other components in the electronic device to perform desired functions.

[0207] The memory 402 can include one or more computer program products, and the computer program products can include various forms of computer-readable storage media, such as volatile memory and / or non-volatile memory. The volatile memory can include, for example, random access memory (RAM) and / or cache memory, etc. The non-volatile memory can include, for example, read-only memory (ROM), hard disk, flash memory, etc. One or more computer program instructions can be stored on the computer-readable storage medium, and the processor 401 can run the program instructions to implement the method for identifying duplicate-named persons of the software programs of the various embodiments of the present disclosure described above and / or other desired functions. In one example, the electronic device may further include: an input device 403 and an output device 404, and these components are interconnected through a bus system and / or other forms of connection mechanisms (not shown).

[0208] In addition, the input device 403 may further include, for example, a keyboard, a mouse, and so on.

[0209] The output device 404 can output various information to the outside. The output device 404 can include, for example, a display, a speaker, a printer, and a communication network and its connected remote output devices, and so on.

[0210] Of course, for simplicity, Figure 4 only some of the components related to the present disclosure in the electronic device are shown in, and components such as buses, input / output interfaces, etc. are omitted. In addition, according to specific application scenarios, the electronic device may further include any other appropriate components.

[0211] Exemplary computer program product and computer-readable storage medium

[0212] In addition to the above methods and devices, an embodiment of the present disclosure may also be a computer program product, which includes computer program instructions that, when run by a processor, cause the processor to execute the steps in the method for identifying persons with the same name according to various embodiments of the present disclosure described in the above "Exemplary Method" section of this specification.

[0213] The computer program product may be written in any combination of one or more programming languages for programming code to perform the operations of the embodiments of the present disclosure. The programming languages include object-oriented programming languages such as Java, C++, etc., and also include conventional procedural programming languages such as the "C" language or similar programming languages. The program code may be executed entirely on the user's computing device, partially on the user's device, executed as an independent software package, partially on the user's computing device and partially on a remote computing device, or entirely on a remote computing device or server.

[0214] In addition, an embodiment of the present disclosure may also be a computer-readable storage medium, on which computer program instructions are stored, and the computer program instructions, when run by a processor, cause the processor to execute the steps in the method for identifying persons with the same name according to various embodiments of the present disclosure described in the above "Exemplary Method" section of this specification.

[0215] The computer-readable storage medium may adopt any combination of one or more readable media. The readable medium may be a readable signal medium or a readable storage medium. The readable storage medium may, for example, include but is not limited to an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the above. More specific examples (non-exhaustive list) of the readable storage medium include: an electrical connection having one or more wires, a portable disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above.

[0216] The basic principles of the present disclosure have been described above in conjunction with specific embodiments. However, it should be noted that the advantages, benefits, effects, etc. mentioned in the present disclosure are only examples and not limitations, and it cannot be considered that these advantages, benefits, effects, etc. are essential for each embodiment of the present disclosure. In addition, the above specific details are only for the purposes of illustration and facilitating understanding, rather than limitations. The above details do not limit the present disclosure to necessarily adopt the above specific details for implementation.

[0217] In the description of the present specification, each embodiment is described in a progressive manner. What each embodiment focuses on is the difference from other embodiments. For the same or similar parts among the embodiments, reference can be made to each other. For the system embodiment, since it basically corresponds to the method embodiment, the description is relatively simple. For the relevant parts, reference can be made to the partial description of the method embodiment.

[0218] The block diagrams of the devices, apparatuses, equipment, and systems involved in the present disclosure are only illustrative examples and are not intended to require or imply that they must be connected, arranged, and configured in the manner shown in the block diagrams. As those skilled in the art will recognize, these devices, apparatuses, equipment, and systems can be connected, arranged, and configured in any manner. Words such as "including", "comprising", "having", etc. are open-ended words, meaning "including but not limited to", and can be used interchangeably with each other. The word "or" and "and" used herein refer to the word "and / or", and can be used interchangeably with each other, unless the context clearly indicates otherwise. The word "such as" used herein refers to the phrase "such as but not limited to", and can be used interchangeably with each other.

[0219] The methods and apparatuses of the present disclosure can be implemented in many ways. For example, the methods and apparatuses of the present disclosure can be implemented by software, hardware, firmware, or any combination of software, hardware, and firmware. The above order of the steps for the method is only for illustration, and the steps of the method of the present disclosure are not limited to the specific order described above, unless otherwise specifically stated. In addition, in some embodiments, the present disclosure can also be implemented as a program recorded in a recording medium, and these programs include machine-readable instructions for implementing the methods according to the present disclosure. Therefore, the present disclosure also covers a recording medium storing a program for executing the method according to the present disclosure.

[0220] It should also be noted that in the devices, equipment, and methods of the present disclosure, each component or each step can be decomposed and / or recombined. These decompositions and / or recombinations should be regarded as equivalent solutions of the present disclosure. The above description of the disclosed aspects is provided to enable any person skilled in the art to make or use the present disclosure. Various modifications to these aspects are very obvious to those skilled in the art, and the general principles defined herein can be applied to other aspects without departing from the scope of the present disclosure. Therefore, the present disclosure is not intended to be limited to the aspects shown herein, but rather to the broadest scope consistent with the principles and novel features disclosed herein.

[0221] The above description has been given for purposes of illustration and description. In addition, this description is not intended to limit the embodiments of the present disclosure to the forms disclosed herein. Although multiple example aspects and embodiments have been discussed above, those skilled in the art will recognize certain variations, modifications, changes, additions, and sub-combinations thereof.

Claims

1. A method for identifying persons with the same name, characterized in that: The method comprises: For two target objects with duplicated names, first feature data samples and second feature data samples of the two target objects are respectively obtained, wherein the first feature data sample includes a plurality of first features, and the second feature data sample includes a plurality of second features, the first feature is a feature for determining the correlation between the target objects, and the second feature is a feature for determining the similarity between the target objects; Determining first feature values ​​of a plurality of first features according to first feature data samples of two target objects includes: Matching the first feature information item of each first feature in the first feature data samples of the two target objects, and when the first feature information items of the first features do not match, determining that the first feature value of the first feature is 0; when the first feature information items of the first features match, calculating the first feature value of the first feature according to the first feature occurrence count item corresponding to the first feature information item of the first feature and a preset first feature occurrence count threshold; Determine second feature values ​​of a plurality of second features according to second feature data samples of the two target objects; Determine a correlation value between two target objects according to a plurality of first eigenvalues ​​and a plurality of second eigenvalues; It is determined whether the persons with the same name in two target objects are the same natural person according to the correlation value and a preset correlation threshold.

2. The method according to claim 1, characterized in that For two target objects with duplicated names, first feature data samples and second feature data samples of the two target objects are respectively obtained, including: Respectively obtain initial first feature data and initial second feature data of two target objects; Based on a preset first feature blacklist, filtering the feature data of the first feature to be filtered in the initial first feature data to generate a first feature data sample; Based on a preset second feature blacklist, the feature data of the second feature to be filtered in the initial second feature data is filtered to generate a second feature data sample.

3. The method according to claim 1, characterized in that The calculation formula for calculating the first feature value of the first feature according to the first feature occurrence count item corresponding to the first feature information item of the first feature and the preset first feature occurrence count threshold is specifically: r=(yx) / y In the formula, y is the preset threshold of the number of occurrences of the first feature, x is the number of occurrences of the first feature; and r is the first eigenvalue of the first feature.

4. The method according to claim 3, characterized in that The method further includes, when each first feature in the first feature data sample includes a plurality of first feature information items and each first feature information item matches, calculating a first feature value of the first feature according to a first feature occurrence count item corresponding to each first feature information item, specifically: The feature value corresponding to each first feature information item of the first feature is calculated according to the first feature occurrence count item corresponding to each first feature information item of the first feature and a preset first feature occurrence count threshold, and the sum of the feature values ​​corresponding to multiple first feature information items is used as the first feature value of the first feature.

5. The method according to claim 1, characterized in that Determining second feature values ​​of a plurality of second features according to second feature data samples of two target objects comprises: Calculating a text similarity value of each second feature in the second feature data samples of the two target objects; The text similarity value of each second feature is used as the second feature value corresponding to the second feature.

6. The method according to claim 1, characterized in that Determine the second feature data samples of the two target objects Determining the second feature values ​​of the plurality of second features comprises: Extract text from each second feature in the second feature data samples of the two target objects according to a preset rule to obtain the core text of each second feature of the two target objects; Calculate the text similarity value of the core text of each second feature of the two target objects; The text similarity value of the core text of each second feature is used as the second feature value corresponding to the second feature.

7. The method according to claim 1, characterized in that Determining the association value of two target objects according to a plurality of first eigenvalues ​​and a plurality of second eigenvalues ​​includes: Summing the first feature values ​​of the plurality of first features to obtain a first feature correlation value; Summing the second feature values ​​of the plurality of second features to obtain a second feature correlation value; The first feature association value and the second feature association value are summed to obtain association values ​​of the two target objects.

8. The method according to claim 1, characterized in that Determining the association value of two target objects according to a plurality of first eigenvalues ​​and a plurality of second eigenvalues ​​includes: Set the weight coefficient of each first feature and each second feature; Multiplying the first eigenvalue of each first feature by its corresponding weight coefficient to obtain a first weighted eigenvalue of each first feature; Summing the first weighted feature values ​​of the plurality of first features to obtain a first feature correlation value; Multiplying the second eigenvalue of each second feature by its corresponding weight coefficient to obtain a second weighted eigenvalue of each second feature; Summing the second weighted feature values ​​of the plurality of second features to obtain a second feature correlation value; The first feature association value and the second feature association value are summed to obtain association values ​​of the two target objects.

9. The method according to claim 1, characterized in that: Determining the association value of two target objects according to the plurality of first feature values ​​and the plurality of second feature values ​​includes: Inputting a plurality of first feature values ​​and a plurality of second feature values ​​into a target object association model to obtain association values ​​of two target objects; wherein the target object association model is based on the first feature values ​​of the target objects with known persons having the same name. The eigenvalue and the second eigenvalue are input, and the correlation value between the target objects is output. A model for determining the correlation degree between the target objects is obtained by training with a machine learning algorithm.

10. The method according to claim 1, characterized in that Determining whether the persons with the same name of two target objects are the same natural person according to the correlation value and a preset correlation threshold includes: When the correlation value is not less than a preset correlation threshold, it is determined that the persons with the same name in the two target objects are the same natural person; When the correlation value is less than a preset correlation threshold, it is determined that the persons with the same name in the two target objects are not the same natural person.

11. A device for identifying persons with the same name, characterized in that: The device comprises: A data sample module, for obtaining, for two target objects with duplicate names, first feature data samples and second feature data samples of the two target objects, respectively, wherein the first feature data sample includes a plurality of first features, and the second feature data sample includes a plurality of second features, the first feature is a feature for determining a correlation relationship between the target objects, and the second feature is a feature for determining a similarity relationship between the target objects; A feature value module, used to determine first feature values ​​of a plurality of first features according to first feature data samples of two target objects, comprising: Matching the first feature information item of each first feature in the first feature data samples of the two target objects, and when the first feature information items of the first features do not match, determining that the first feature value of the first feature is 0; when the first feature information items of the first features match, calculating the first feature value of the first feature according to the first feature occurrence count item corresponding to the first feature information item of the first feature and a preset first feature occurrence count threshold; The feature value module is also used to determine second feature values ​​of multiple second features according to the second feature data samples of the two target objects; A correlation value module, used to determine the correlation value of two target objects according to a plurality of first characteristic values ​​and a plurality of second characteristic values; The result determination module is used to determine whether the persons with the same name of two target objects are the same natural person according to the correlation value and a preset correlation threshold.

12. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method according to any one of claims 1 to 10 is implemented.

13. An electronic device, characterized in that: The electronic device comprises: processor; a memory for storing instructions executable by the processor; The processor is used to read the executable instructions from the memory and execute the executable instructions to implement any one of the methods 1 to 10 above.

Images