Intersection similarity matching-based national-enterprise-affiliated and central-enterprise-affiliated identification method and device, and storage medium

Abstract

The invention relates to a national enterprise and central enterprise affiliated identification method based on cross similarity matching, and the method comprises the steps: S1, collecting a national asset member unit list from a group enterprise management system, constructing a national enterprise and central enterprise member unit data set which comprises the Chinese and English names of the national enterprise and central enterprise member units; s2, constructing a data set of full names of the national enterprise and central enterprise member units and corresponding abbreviations of the national enterprise and central enterprise member units; s3: aiming at the company name of the civil enterprise, screening in the industry and commerce information base, and forming a data set of suspected civil enterprise affiliated to the national capital, the screening result including the full name and short name keywords of the member units of the national enterprise and the central enterprise included in the data sets of S1 and S2; s4, vectorizing enterprise names and abbreviations of all national resource member units of the national enterprise and the central enterprise; vectorizing the name and the abbreviation of the suspected civil enterprise affiliated to the national asset; s5, calling a cross similarity algorithm, calculating the vector similarity between the name of the suspected civil enterprise affiliated national capital enterprise and the name and abbreviation of the member unit of the national enterprise and the central enterprise, and retaining the suspected civil enterprise with the similarity score greater than 0.7; and S6, according to the similarity score in the S5, generating a suspected affiliation relation table.

Description

Technical Field

[0001] The present invention belongs to the field of computer technology, and in particular relates to a method, device and storage medium for identifying affiliated state-owned enterprises and central enterprises based on cross-similarity matching. Background Art

[0002] The behavior of affiliated enterprises, especially private enterprises affiliated with state-owned enterprises and central enterprises, is likely to cause reputation, credibility and economic losses to the affiliated state-owned enterprises and central enterprises, and damage the legitimate interests of the affiliated state-owned enterprises and central enterprises. At present, the identification of affiliated enterprises mainly adopts the traditional manual screening method, which has low identification efficiency and low accuracy. Summary of the invention

[0003] The purpose of the present invention is to provide a method, device and storage medium for identifying affiliated state-owned enterprises and central enterprises based on cross-similarity matching, so as to solve the problems existing in the prior art.

[0004] In order to achieve the above object, the present invention adopts the following technical solutions:

[0005] A method for identifying affiliated state-owned enterprises and central enterprises based on cross-similarity matching:

[0006] S1: Collect the list of all state-owned member units in one or more state-owned enterprises or central enterprises management groups from the group enterprise management system to construct a state-owned enterprise or central enterprise member unit data set; the state-owned enterprise or central enterprise member unit data set includes the Chinese and English names of the state-owned enterprise or central enterprise member units;

[0007] S2: Construct a dataset of the full names of state-owned enterprises and central enterprises and their corresponding abbreviations, so that state-owned enterprises and central enterprises can be mapped one by one with their corresponding abbreviations;

[0008] S3: For the company names of private enterprises, we screened them in the industrial and commercial information database. The screening results contained the full names and abbreviations of the state-owned enterprises and central enterprises in the S1 and S2 data sets. If they were not in the state-owned enterprises and central enterprises data sets, there was a possibility of misusing the names and being affiliated with state-owned enterprises, forming a data set of suspected private enterprises affiliated with state-owned assets.

[0009] S4: Quantify the company names and abbreviations of all state-owned enterprises and central enterprises; quantify the company names and abbreviations of suspected private enterprises affiliated with state-owned assets;

[0010] S5: Call the cross-similarity algorithm to calculate the vector similarity between the names of suspected private enterprises affiliated with state-owned enterprises and the names and abbreviations of state-owned enterprises and central enterprises, perform similarity scoring, and retain suspected private enterprises with a similarity score greater than 0.7.

[0011] S6: Generate a suspected affiliation relationship table based on the similarity score described in S5.

[0012] Furthermore, the data set of state-owned enterprises and member units of central enterprises described in S1 removes the strings of “Limited Company”, “Liability Company” and “Limited Liability Company”.

[0013] Furthermore, the corresponding abbreviations of the state-owned enterprises and central enterprises member units described in S2 are abbreviations extracted based on industrial and commercial information.

[0014] Furthermore, the dataset of suspected private enterprises relying on state-owned assets as described in S3 is formed by removing the strings of "Limited Company", "Liability Company" and "Limited Liability Company" to preliminarily determine the dataset of suspected private enterprises relying on state-owned assets;

[0015] Furthermore, the company name of the private enterprise described in S3 includes Chinese and English names.

[0016] Furthermore, the vectorization of the data set described in S4 uses the language model Bert to construct the vectorization of the names and abbreviations of all state-owned member units of state-owned enterprises and central enterprises; and uses the language model Bert to construct the vectorization of the names and abbreviations of state-owned enterprises and central enterprises that are suspected to be affiliated with private enterprises.

[0017] Furthermore, the cross-similarity algorithm described in S5, i.e., based on cosine similarity calculation and minimum edit distance similarity calculation, uses two similarity algorithms to respectively calculate the similarity of suspected private enterprises affiliated with state-owned enterprises and member units of central enterprises, and then obtains the final similarity score after weighted average.

[0018] Furthermore, the generation of the affiliation relationship table described in S6 is to retain the data sets with a similarity score greater than 0.9 to generate a suspected affiliation relationship table; if there is no data set with a similarity score greater than 0.9, the data sets with the top two similarity scores are retained to generate a suspected affiliation relationship table; for the same private enterprise suspected of being a private enterprise affiliated with state-owned assets, corresponding to multiple member units, the group with the largest similarity score is selected to generate a suspected affiliation relationship table;

[0019] The present invention also provides a device for identifying affiliated state-owned enterprises and central enterprises based on cross-similarity matching, the identification device comprising:

[0020] Construct a state-owned enterprise and central enterprise member unit data set module, collect the list of all state-owned member units in one or more state-owned enterprises and central enterprise management groups from the group enterprise management system, and construct a state-owned enterprise and central enterprise member unit data set; the state-owned enterprise and central enterprise member unit data set includes the Chinese and English names of the state-owned enterprise and central enterprise member units;

[0021] Construct a data set module for the abbreviations of state-owned enterprises and central enterprises, and construct a data set for the full names of state-owned enterprises and central enterprises and the abbreviations of state-owned enterprises and central enterprises, so that state-owned enterprises and central enterprises can be mapped to their abbreviations one by one;

[0022] Construct a data set module for suspected private enterprises affiliating with state-owned assets. For the company names of private enterprises, screen them in the industrial and commercial information database. The screening results contain the full names and abbreviations of state-owned enterprises and central enterprises contained in the S1 and S2 data sets. If they are not in the state-owned enterprises and central enterprises data sets, there is a possibility of misusing the name and affiliating with state-owned enterprises, thus forming a data set of suspected private enterprises affiliating with state-owned assets.

[0023] The dataset vectorization module vectorizes the company names and abbreviations of all state-owned enterprises and central enterprises; and vectorizes the company names and abbreviations of suspected private enterprises affiliated with state-owned assets;

[0024] The similarity calculation module between the names of suspected private enterprises affiliated with state-owned enterprises and their member units calls the cross-similarity algorithm to calculate the vector similarity between the names of suspected private enterprises affiliated with state-owned enterprises and the names and abbreviations of member units of state-owned enterprises and central enterprises, performs similarity scoring, and retains suspected private enterprises with similarity scores greater than 0.7;

[0025] A module for generating affiliation relationship table: generating a suspected affiliation relationship table according to the similarity score described in S5.

[0026] The present invention also provides a computer-readable storage medium, which stores computer program instructions. When the computer program instructions are executed by a computer, the computer executes the steps of any of the above-mentioned methods for identifying affiliated state-owned enterprises and central enterprises based on cross-similarity matching.

[0027] By adopting the above technical solution, the present invention has the following beneficial effects:

[0028] The present invention is based on a machine learning algorithm. First, keywords are used to preliminarily screen out suspected affiliated enterprises from the entire industrial and commercial information. Then, a vector similarity algorithm is used to calculate the similarity between the suspected affiliated enterprises and the member units. Finally, a threshold is set to obtain the final screening result, which greatly improves the efficiency and accuracy of identifying affiliated enterprises. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] In order to more clearly illustrate the specific implementation methods of the present invention or the technical solutions in the prior art, the drawings required for use in the specific implementation methods or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some implementation methods of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0030] Figure 1 It is a flow chart of the method for identifying affiliated state-owned enterprises and central enterprises based on cross-similarity matching of the present invention;

[0031] Figure 2It is a schematic diagram of the method for identifying state-owned enterprises and central enterprises affiliated to affiliated enterprises based on cross-similarity matching of the present invention;

[0032] Figure 3 It is a schematic diagram of the device for identifying affiliated state-owned enterprises and central enterprises based on cross-similarity matching of the present invention. DETAILED DESCRIPTION

[0033] The technical solution of the present invention will be described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0034] Combination Figure 1 As shown in the figure, a method for identifying affiliated state-owned enterprises and central enterprises based on cross-similarity matching is as follows:

[0035] S1: Collect the list of all state-owned member units in one or more state-owned enterprises or central enterprises management groups from the group enterprise management system to construct a state-owned enterprise or central enterprise member unit data set; the state-owned enterprise or central enterprise member unit data set includes the Chinese and English names of the state-owned enterprise or central enterprise member units;

[0036] S2: Construct a dataset of the full names of state-owned enterprises and central enterprises and their corresponding abbreviations, so that state-owned enterprises and central enterprises can be mapped one by one with their corresponding abbreviations;

[0037] S3: For the company names of private enterprises, we screened them in the industrial and commercial information database. The screening results contained the full names and abbreviations of the state-owned enterprises and central enterprises in the S1 and S2 data sets. If they were not in the state-owned enterprises and central enterprises data sets, there was a possibility of misusing the names and being affiliated with state-owned enterprises, forming a data set of suspected private enterprises affiliated with state-owned assets.

[0038] S4: Quantify the company names and abbreviations of all state-owned enterprises and central enterprises; quantify the company names and abbreviations of suspected private enterprises affiliated with state-owned assets;

[0039] S5: Call the cross-similarity algorithm to calculate the vector similarity between the names of suspected private enterprises affiliated with state-owned enterprises and the names and abbreviations of state-owned enterprises and central enterprises, perform similarity scoring, and retain suspected private enterprises with a similarity score greater than 0.7.

[0040] S6: Generate a suspected affiliation relationship table based on the similarity score described in S5.

[0041] Of course, private enterprises that are not included in the data of state-owned enterprises and central enterprises and are still in existence may use false names and register with state-owned enterprises.

[0042] Specifically, the data set of state-owned enterprises and member units of central enterprises described in S1 removes the strings of “Limited Liability Company”, “Liability Company”, and “Limited Liability Company”.

[0043] Specifically, the corresponding abbreviations of the state-owned enterprises and central enterprises member units described in S2 are abbreviations extracted based on industrial and commercial information.

[0044] Specifically, the dataset of suspected private enterprises relying on state-owned assets as described in S3 is formed by removing the strings of "Limited Company", "Liability Company" and "Limited Liability Company" to preliminarily determine the dataset of suspected private enterprises relying on state-owned assets;

[0045] Specifically, the company names of the private enterprises described in S3 include Chinese and English names.

[0046] Specifically, the vectorization of the data set described in S4 uses the language model Bert to construct the vectorization of the names and abbreviations of all state-owned member units of state-owned enterprises and central enterprises; and uses the language model Bert to construct the vectorization of the names and abbreviations of state-owned enterprises and central enterprises that are suspected to be affiliated with private enterprises.

[0047] Specifically, the cross-similarity algorithm described in S5, namely, cosine similarity calculation and minimum edit distance similarity calculation, uses two similarity algorithms to respectively calculate the similarity of suspected private enterprises affiliated with state-owned enterprises and member units of central enterprises, and then obtains the final similarity score after weighted average.

[0048] This part uses text similarity algorithms, including: cosine similarity calculation and minimum edit distance technology. Cosine similarity calculates the spatial similarity distance between the suspected affiliated company name and the member unit name after vectorization, and minimum edit distance calculates the semantic similarity distance between the suspected affiliated company name and the member unit name after vectorization.

[0049] Cross-similarity refers to the weighted average of the final results of the above two vector similarity algorithms according to the algorithm properties and accuracy to obtain the final score.

[0050] Specifically, the generation of the affiliation relationship table described in S6 is to retain the data sets with a similarity score greater than 0.9 to generate a suspected affiliation relationship table; if there is no data set with a similarity score greater than 0.9, the data sets with the top two similarity scores are retained to generate a suspected affiliation relationship table; for the same private enterprise suspected of being a private enterprise affiliated with state-owned assets, corresponding to multiple member units, the group with the largest similarity score is selected to generate a suspected affiliation relationship table;

[0051] By adopting the technical solution of the present invention, the machine automatically completes:

[0052] 1. Use cosine similarity to calculate the spatial similarity distance between the suspected affiliated enterprise name and the member unit name after vectorization.

[0053] 2. Use the minimum edit distance to calculate the semantic similarity distance between the suspected affiliated enterprise name and the member unit name after vectorization.

[0054] Manual completion:

[0055] 1. In the entire industrial and commercial data, search for private enterprises that are still in existence and have used the names of member units through the abbreviations of the member units and keywords in their names.

[0056] 2. The final results of the two types of vector similarity algorithms, cosine similarity and minimum edit distance, are weighted averaged according to the weighted ratios set by expert experience.

[0057] See also Figure 2 As shown, go to the group enterprise management system, read the member units of the party, remove the strings of "Limited Liability Company", "Liability Company", and "Limited Liability Company", and build the member unit data set, such as:

[0058] [Company A, Company B, Company C, …];

[0059] The member unit and abbreviation mapping database formed:

[0060] [Company A’s abbreviation,

[0061] Company B,

[0062] C Company C abbreviation, …];

[0063] The company names and abbreviations of all state-owned enterprises and central enterprises are quantified.

[0064] [Company A[0.93,0.41,0.32,0.331.0.944..0935,……]

[0065] A is abbreviated as [0.93, 0.41, 0.32, 0.331. 0.944.. 0935, ...],

[0066] ……]

[0067] From the industrial and commercial information database, we collect private enterprise information. Based on the company abbreviation and full name + rules (including the name of the state-owned member unit, keyword information and its abbreviation), and the nature of the enterprise is a private enterprise, we extract the list of suspected private enterprise affiliations, and remove the strings such as "limited liability company", "liability company", and "limited liability company" to form a suspected private enterprise affiliation data set

[0068] [Company A’s abbreviation,

[0069] Company B is short for

[0070] C Company C abbreviation, ...]

[0071] Further, vector data of suspected affiliated enterprise names and abbreviations are formed.

[0072] [Company A[0.93,0.41,0.32.0.331,0.944..0935,……]

[0073] A is abbreviated as [0.93, 0.41, 0.32, 0.331. 0.944.. 0935, ...],

[0074] ……]

[0075] After vector similarity calculation, the similarity score between the suspected affiliated enterprise and the member unit is calculated.

[0076] [Similarity score of A private enterprise N member units, …]

[0077] Finally, generate a table of suspected affiliated companies and member units’ affiliation relationships

[0078] [Similarity score of A private enterprise N member units, …]

[0079] In yet another embodiment, see Figure 3 As shown, the present invention also provides a device for identifying affiliated state-owned enterprises and central enterprises based on cross-similarity matching, and the identification device includes:

[0080] Construct a state-owned enterprise and central enterprise member unit data set module, collect the list of all state-owned member units in one or more state-owned enterprises and central enterprise management groups from the group enterprise management system, and construct a state-owned enterprise and central enterprise member unit data set; the state-owned enterprise and central enterprise member unit data set includes the Chinese and English names of the state-owned enterprise and central enterprise member units;

[0081] Construct a data set module for the abbreviations of state-owned enterprises and central enterprises, and construct a data set for the full names of state-owned enterprises and central enterprises and the abbreviations of state-owned enterprises and central enterprises, so that state-owned enterprises and central enterprises can be mapped to their abbreviations one by one;

[0082] Construct a data set module for suspected private enterprises affiliating with state-owned assets. For the company names of private enterprises, screen them in the industrial and commercial information database. The screening results contain the full names and abbreviations of state-owned enterprises and central enterprises contained in the S1 and S2 data sets. If they are not in the state-owned enterprises and central enterprises data sets, there is a possibility of misusing the name and affiliating with state-owned enterprises, thus forming a data set of suspected private enterprises affiliating with state-owned assets.

[0083] The dataset vectorization module vectorizes the company names and abbreviations of all state-owned enterprises and central enterprises; and vectorizes the company names and abbreviations of suspected private enterprises affiliated with state-owned assets;

[0084] The similarity calculation module between the names of suspected private enterprises affiliated with state-owned enterprises and their member units calls the cross-similarity algorithm to calculate the vector similarity between the names of suspected private enterprises affiliated with state-owned enterprises and the names and abbreviations of member units of state-owned enterprises and central enterprises, performs similarity scoring, and retains suspected private enterprises with similarity scores greater than 0.7;

[0085] A module for generating affiliation relationship table: generating a suspected affiliation relationship table according to the similarity score described in S5.

[0086] In another embodiment, the present invention also provides a computer-readable storage medium, which stores computer program instructions. When the computer program instructions are executed by a computer, the computer executes the steps of any of the above-mentioned methods for identifying affiliated state-owned enterprises and central enterprises based on cross-similarity matching.

[0087] It should be noted that the storage medium shown in the embodiment of the present invention may be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM), a 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. In the present invention, a computer-readable storage medium may be any tangible medium containing or storing a program, which may be used by or in combination with an instruction execution system, device or device. In the present invention, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, which carries a computer-readable program code. Such propagated data signals may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. Computer-readable signal media may also be any storage medium other than computer-readable storage media, which may send, propagate, or transmit programs for use by or in conjunction with an instruction execution system, apparatus, or device. The program code contained on the storage medium may be transmitted using any appropriate medium, including but not limited to: wireless, wired, etc., or any suitable combination of the above.

[0088] The accompanying drawings illustrate the possible implementation architecture, functions and operations of the system, method and computer program product according to various embodiments of the present invention. Each box in the figure can represent a module, program segment, or a part of the code, and the above-mentioned module, program segment, or a part of the code contains one or more executable instructions for implementing the specified logical function. It should also be noted that each box in the block diagram or flow chart, and the combination of boxes in the block diagram or flow chart can be implemented by a dedicated hardware-based system that performs the specified function or operation, or can be implemented by a combination of dedicated hardware and computer instructions.

[0089] The units involved in the embodiments of the present invention may be implemented by software or hardware, and the units described may also be arranged in a processor. The names of these units do not, in some cases, limit the units themselves.

[0090] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for identifying affiliated state-owned enterprises and central enterprises based on cross-similarity matching, characterized in that S1: Collect the list of all state-owned asset member units within a certain or multiple state-owned enterprises and central enterprises management groups from the group enterprise management system, and construct a dataset of state-owned enterprises and central enterprises member units; the dataset of state-owned enterprises and central enterprises member units includes the Chinese and English names of state-owned enterprises and central enterprises member units. S2: Construct a dataset of the full names of state-owned enterprises and central enterprises member units and the corresponding abbreviations of state-owned enterprises and central enterprises member units, so that the state-owned enterprises and central enterprises member units are in one-to-one mapping with the corresponding abbreviations of state-owned enterprises and central enterprises members. S3: Screen the company names of private enterprises in the industrial and commercial information database. If the screening results contain the full names and abbreviated keywords of state-owned enterprises and central enterprises member units included in the datasets of S1 and S2 and are not in the dataset of state-owned enterprises and central enterprises member units, there may be a case of misusing the name and affiliating with state-owned enterprises, and a dataset of suspected private enterprises affiliating with state-owned assets is formed. S4: Vectorize the enterprise names and abbreviations of all state-owned asset member units of state-owned enterprises and central enterprises; vectorize the enterprise names and abbreviations of suspected private enterprises affiliating with state-owned assets. S5: Call the cross-similarity algorithm to calculate the vector similarity between the names of suspected private enterprises affiliating with state-owned assets and the names and abbreviations of state-owned enterprises and central enterprises member units, perform similarity scoring, and retain the suspected private enterprises with a similarity score greater than 0.

7. S6: Generate a suspected affiliation relationship table according to the similarity scores described in S5.

2. According to the method of identifying affiliated state-owned enterprises and central enterprises based on cross-similarity matching according to claim 1, it is characterized in that: For the dataset of state-owned enterprises and central enterprises member units described in S1, remove the strings "Co., Ltd.", "Limited Liability Company", and "Company Limited by Shares".

3. According to claim 2, a method for identifying affiliated state-owned enterprises and central enterprises based on cross-similarity matching is characterized in that: The corresponding abbreviations of state-owned enterprises and central enterprises member units described in S2 are abbreviations extracted based on industrial and commercial information.

4. According to the method of identifying affiliated state-owned enterprises and central enterprises based on cross-similarity matching according to claim 1, it is characterized in that: For the dataset of suspected private enterprises affiliating with state-owned assets formed in S3, remove the strings "Co., Ltd.", "Limited Liability Company", and "Company Limited by Shares" to preliminarily determine the dataset of suspected private enterprises affiliating with state-owned assets.

5. According to the method of identifying affiliated state-owned enterprises and central enterprises based on cross-similarity matching according to claim 1, it is characterized in that: The company names of private enterprises described in S3 include Chinese and English names.

6. The method for identifying affiliated state-owned enterprises and central enterprises based on cross-similarity matching according to claim 1 is characterized in that: For the vectorization of the dataset described in S4, use the language model Bert to vectorize the names and abbreviations of all state-owned asset member units of state-owned enterprises and central enterprises; use the language model Bert to vectorize the names and abbreviations of suspected private enterprises affiliating with state-owned enterprises and central enterprises.

7. The method for identifying affiliated state-owned enterprises and central enterprises based on cross-similarity matching according to claim 6 is characterized in that: The cross-similarity algorithm described in S5, that is, based on cosine similarity calculation and minimum edit distance similarity calculation, after using the two similarity algorithms to calculate the similarity between suspected private enterprises affiliating with state-owned enterprises and central enterprises member units respectively, the final similarity score is obtained by weighted average.

8. The method for identifying affiliated state-owned enterprises and central enterprises based on cross-similarity matching according to claim 7 is characterized in that: For the generation of the affiliation relationship table described in S6, retain the dataset with a similarity score greater than 0.9 to generate a suspected affiliation relationship table; if there is no dataset with a similarity score greater than 0.9, retain the top two datasets with similarity scores to generate a suspected affiliation relationship table; for a private enterprise corresponding to multiple member units suspected of affiliating with state-owned assets, select the group with the largest similarity score to generate a suspected affiliation relationship table.

9. A device for identifying affiliated state-owned enterprises and central enterprises based on cross-similarity matching, characterized in that: The identification device includes: Construct a state-owned enterprise and central enterprise member unit data set module, collect the list of all state-owned member units in one or more state-owned enterprises and central enterprise management groups from the group enterprise management system, and construct a state-owned enterprise and central enterprise member unit data set; the state-owned enterprise and central enterprise member unit data set includes the Chinese and English names of the state-owned enterprise and central enterprise member units; Construct a data set module for the abbreviations of state-owned enterprises and central enterprises, and construct a data set for the full names of state-owned enterprises and central enterprises and the abbreviations of state-owned enterprises and central enterprises, so that state-owned enterprises and central enterprises can be mapped to their abbreviations one by one; Construct a data set module for suspected private enterprises affiliating with state-owned assets. For the company names of private enterprises, screen them in the industrial and commercial information database. The screening results contain the full names and abbreviations of state-owned enterprises and central enterprises contained in the S1 and S2 data sets. If they are not in the state-owned enterprises and central enterprises data sets, there is a possibility of misusing the name and affiliating with state-owned enterprises, thus forming a data set of suspected private enterprises affiliating with state-owned assets. The dataset vectorization module vectorizes the company names and abbreviations of all state-owned enterprises and central enterprises; and vectorizes the company names and abbreviations of suspected private enterprises affiliated with state-owned assets; The similarity calculation module between the names of suspected private enterprises affiliated with state-owned enterprises and their member units calls the cross-similarity algorithm to calculate the vector similarity between the names of suspected private enterprises affiliated with state-owned enterprises and the names and abbreviations of member units of state-owned enterprises and central enterprises, performs similarity scoring, and retains suspected private enterprises with similarity scores greater than 0.7; A module for generating affiliation relationship table: generating a suspected affiliation relationship table according to the similarity score described in S5.

10. A computer-readable storage medium storing computer program instructions. When the computer program instructions are executed by a computer, the computer executes the steps of the method for identifying affiliated state-owned enterprises and central enterprises based on cross-similarity matching as described in any one of claims 1 to 8.

Images