Enterprise industry information determination method and device, and electronic device

Abstract

The present disclosure provides an industry information determination method and device for an enterprise and an electronic device, and relates to the technical fields of knowledge graph and big data. The method comprises: obtaining enterprise operation information of a target enterprise; reorganizing text in the enterprise operation information to obtain a plurality of text clauses; matching the plurality of text clauses with pre-acquired industry information of a plurality of industries respectively to obtain matching results; and determining the industry information of the target enterprise according to the matching results. The technical scheme of the present disclosure reorganizes the text in the enterprise operation information, matches the plurality of reorganized text clauses with the pre-acquired industry information of the plurality of industries respectively, and determines the industry information of the target enterprise. The matching accuracy is high, no classifier needs to be trained, and manpower, material resources and computing resources are saved.

Description

Technical Field

[0001] This disclosure relates to the fields of knowledge graph and big data technology, and in particular to a method, apparatus and electronic device for determining industry information of an enterprise. Background Technology

[0002] In the context of spatiotemporal big data, smart parks are a sub-direction of smart city construction. The healthy development of a park requires cooperation and appropriate competition among different enterprises within the park. Therefore, during the construction of a park, park managers need to know the industry to which each enterprise in the park belongs.

[0003] To categorize companies into different industries, classifiers are typically trained. However, achieving good classification accuracy requires extensive manual data annotation for model training, consuming significant human and material resources. Furthermore, the number of industry categories is vast; national standard industry classification systems contain thousands of categories, making training such a large classifier extremely time-consuming and computationally expensive. Summary of the Invention

[0004] This disclosure provides a method, apparatus, and electronic device for determining industry information of an enterprise.

[0005] According to one aspect of this disclosure, a method for determining industry information of an enterprise is provided, comprising:

[0006] Obtain the target company's business information;

[0007] The text in the enterprise's business information is reorganized to obtain multiple text clauses;

[0008] Multiple text clauses are matched with industry information from multiple pre-acquired industries to obtain matching results;

[0009] Based on the matching results, determine the industry information of the target company.

[0010] According to another aspect of this disclosure, an apparatus for determining industry information of an enterprise is provided, comprising:

[0011] The acquisition module is used to obtain the target company's business information;

[0012] The restructuring module is used to restructure the text in the enterprise's business information to obtain multiple text clauses; the matching module is used to match the multiple text clauses with industry information from multiple pre-acquired industries to obtain matching results.

[0013] The determination module is used to determine the industry information of the target company based on the matching results.

[0014] According to another aspect of this disclosure, an electronic device is provided, comprising:

[0015] At least one processor; and

[0016] The memory is communicatively connected to the at least one processor; wherein,

[0017] The memory stores instructions that can be executed by the at least one processor to enable the at least one processor to perform the methods in any embodiment of this disclosure.

[0018] According to another aspect of this disclosure, a non-transitory computer-readable storage medium is provided storing computer instructions for causing a computer to perform the methods of any embodiment of this disclosure.

[0019] According to another aspect of this disclosure, a computer program product is provided, including a computer program that, when executed by a processor, implements the methods of any embodiment of this disclosure.

[0020] In this embodiment, the accuracy of matching target enterprise and industry information is high, and there is no need to train a classifier, saving manpower, material resources and computing resources.

[0021] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of this disclosure, nor is it intended to limit the scope of this disclosure. Other features of this disclosure will become readily apparent from the following description. Attached Figure Description

[0022] The accompanying drawings are provided to better understand this solution and do not constitute a limitation of this disclosure. Wherein:

[0023] Figure 1 This is a flowchart of a method for determining industry information of an enterprise in one embodiment of this disclosure;

[0024] Figure 2 This is a flowchart of a method for determining industry information of an enterprise in one embodiment of this disclosure;

[0025] Figure 3 This is a schematic diagram of a method for determining industry information of an enterprise in one embodiment of this disclosure;

[0026] Figure 4 This is a schematic diagram illustrating the recombination of text in enterprise business information into text clauses in one embodiment of this disclosure;

[0027] Figure 5 This is a schematic diagram illustrating the matching process between text clauses and industry information in one embodiment of this disclosure;

[0028] Figure 6This is a schematic diagram of an enterprise industry information determination device in one embodiment of the present disclosure;

[0029] Figure 7 This is a block diagram of an electronic device used to implement the method for determining industry information of an enterprise according to embodiments of the present disclosure. Detailed Implementation

[0030] The exemplary embodiments of this disclosure are described below with reference to the accompanying drawings, including various details of the embodiments to aid understanding, and should be considered merely exemplary. Therefore, those skilled in the art will recognize that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of this disclosure. Similarly, for clarity and brevity, descriptions of well-known functions and structures are omitted in the following description.

[0031] This disclosure provides a method for determining industry information of an enterprise. Figure 1 This is a flowchart of a method for determining industry information of an enterprise according to an embodiment of the present disclosure. This method can be applied to an enterprise industry information determination device, which can be deployed in a terminal device, server, or other processing device. In some possible implementations, the method can also be implemented by a processor calling computer-readable instructions stored in memory. For example... Figure 1 As shown, it includes:

[0032] Step S101: Obtain the target company's business information;

[0033] The business information may include descriptive information about the company's business direction or the products it deals in. For example, the business scope description may include: production and sales of textiles and clothing; production and sales of raw materials and accessories for clothing.

[0034] The target company's business information can be found on the Internet through search engines or in a pre-set database.

[0035] Step S102: Reorganize the text in the enterprise's business information to obtain multiple text clauses;

[0036] Because the text statements in business information are often quite long, in order to improve the accuracy of matching, the text in the business information is reorganized in a certain way to obtain multiple shorter text clauses. For example, reorganizing the text "production and sales of textiles and clothing; production and sales of raw materials and accessories for clothing" in the business information can yield text clauses such as "production of textiles", "production of clothing", "sales of textiles", and "sales of clothing".

[0037] Before reorganizing the text in the company's business information, a preprocessing step is also included. Preprocessing includes, but is not limited to, noise reduction and replacement.

[0038] Company business information often contains noisy text such as "business projects" and "this company's main business." To obtain more accurate industry information about target companies, noise reduction processing is performed on their business information. The specific process includes: First, maintaining a noise word list by continuously adding common noise words and phrases, and deleting these noise words from the business information. Second, most business information contains parentheses, and the content within these parentheses is mostly disclaimers or legal notices. Therefore, during the noise reduction process, these parentheses and their contents are deleted.

[0039] In addition, punctuation marks will be uniformly replaced, with English punctuation marks being replaced with Chinese punctuation marks. Furthermore, in business information, individual sentences typically end with semicolons and commas, while entire paragraphs of business information end with periods. Therefore, for ease of subsequent processing, semicolons and commas will be uniformly replaced with periods.

[0040] It should be noted that the noise reduction process can also be other noise information, and the replacement process is not limited to punctuation marks; it can be the replacement of other information. This disclosure does not limit this aspect.

[0041] Step S103: Match the multiple text clauses with the industry information of multiple pre-acquired industries to obtain the matching results;

[0042] Industry information from multiple pre-acquired industries is used. This industry information can represent industry categories, such as manufacturing, food manufacturing, canned food manufacturing, and canned seafood manufacturing. For each text clause, the similarity between the text clause and each piece of industry information is calculated, and the matching result is determined based on multiple similarity scores.

[0043] Step S104: Determine the industry information of the target company based on the matching results.

[0044] After obtaining the matching results of each text clause and each industry information, the industry information that meets the preset conditions in the matching results of the text clauses is taken as the industry information of the target company.

[0045] This disclosed technical solution can be applied to the construction of smart parks in the context of spatiotemporal big data smart city scenarios. By accurately matching the industry of enterprises, it can provide data support for investment promotion in smart parks.

[0046] In related technologies, determining the industry information of enterprises requires a large amount of manually labeled training data. When there are many enterprise categories, the classification model is difficult to train and requires a lot of human and material resources. Moreover, when there are many categories, the classification model is difficult to classify enterprises with fine granularity, resulting in poor matching results.

[0047] The method for determining industry information of an enterprise provided in this disclosure involves: acquiring the business information of a target enterprise; reorganizing the text in the business information to obtain multiple text clauses; matching each of the multiple text clauses with pre-acquired industry information from multiple industries to obtain matching results; and determining the industry information of the target enterprise based on the matching results. This technical solution, by reorganizing the text in the business information and matching the reorganized text clauses with pre-acquired industry information from multiple industries, determines the industry information of the target enterprise. It achieves high matching accuracy, eliminates the need for classifier training, and saves manpower, material resources, and computational resources.

[0048] In one possible implementation, the text in the enterprise's business information is reorganized to obtain multiple text clauses, including:

[0049] The text in the company's business information is broken down to obtain multiple keywords;

[0050] Multiple keywords are recombined to obtain multiple text clauses.

[0051] In practical applications, each long text statement in a company's business information can be broken down into multiple words. Multiple keywords are then identified from these words according to preset rules. These keywords are then recombined, specifically by concatenating the keywords from each text statement to obtain multiple text clauses. The preset rules can be set according to specific needs, such as setting rules based on the different parts of speech of the words.

[0052] In this embodiment of the disclosure, since the text statements in the enterprise business information are long, the accuracy of text matching can be improved by reorganizing the text in the enterprise business information and using the reorganized multiple text clauses as the data basis for matching with industry information.

[0053] In one possible implementation, the text in the company's business information is split to obtain multiple keywords, including:

[0054] The text in the company's business information is split into multiple short sentences;

[0055] For each short text sentence, determine the part of speech of each word in the short text sentence and the dependency relationships between each word. Based on the part of speech of each word and the dependency relationships between each word, determine the multiple keywords corresponding to the short text sentence.

[0056] In practical applications, the text in enterprise operation information is generally a long text. In order to extract clauses more easily, it is necessary to split the long text into sentences according to certain rules, process the long text based on punctuation marks, and obtain multiple short text sentences.

[0057] After analyzing a large amount of data, it is found that: each short sentence generally ends with a semicolon, comma or period. Through preprocessing, these punctuation marks are normalized and replaced, and commas and semicolons are replaced with periods. The long text is split into sentences through periods. For example, for the long text: "Production and sales of textiles and clothing; production and sales of raw materials and accessories for clothing.", the short sentences obtained are: "Production and sales of textiles and clothing", "Production and sales of raw materials and accessories for clothing".

[0058] After obtaining the short text sentences, for each short text sentence, the syntactic and morphological analysis model tool can be used to analyze the syntax and morphology of this short text sentence, so as to obtain the dependency relationship between different words in the short text sentence. For example, verb-object relationship, parallel relationship, etc. At the same time, the词性 of each word can be marked, and nouns, verbs, auxiliary words, etc. in the sentence can be found. After having the dependency relationship between words and the词性 of each word in the sentence, these dependency relationships and词性 can be used to disassemble this short sentence to obtain multiple keywords.

[0059] Among them, the syntactic and morphological analysis model can be a customized natural language processing model (Natural Language Processing Customization, NLPC), or other models with syntactic and morphological analysis functions. The embodiments of the present disclosure do not limit this.

[0060] For example, for the short text sentence "Production and sales of textiles and clothing", after being processed by the NLPC model, the nouns "textiles", "clothing"; the verbs "production", "sales"; the auxiliary word "of"; the conjunction "and" and other words'词性 can be obtained. At the same time, the dependency relationship between each word can be obtained: the relationship between "textiles" and "clothing" is a parallel relationship, the relationship between "production" and "sales" is a parallel relationship, and the relationship between "clothing" and "production" is a verb-object relationship. After analyzing the enterprise operation information, it is found that most of the enterprise operation information is mainly composed of combinations of verbs and nouns. For example: "Sales and maintenance of RVs, cars and spare parts; automobile interior decoration; management and information consultation of RV camping sites." Therefore, the verbs and nouns in the short text sentence are determined as keywords.

[0061] In this embodiment of the disclosure, multiple keywords corresponding to the text phrase are determined based on the part of speech of each word in the text phrase and the dependency relationship between each word. The keywords obtained through this processing method are reorganized to obtain text clauses, which have a higher accuracy rate when matching with industry information.

[0062] In one possible implementation, multiple keywords are recombined to obtain multiple text clauses, including:

[0063] For each short text sentence, multiple keywords are reorganized based on their parts of speech and dependencies to obtain multiple text clauses.

[0064] In practical applications, after obtaining multiple keywords corresponding to a short text sentence, the keywords are reorganized into text clauses based on their parts of speech and the dependency relationships between them. For example, after obtaining keywords such as verbs and nouns in a short text sentence, as well as the dependency relationships between them, verbs and nouns can be combined to obtain multiple text clauses corresponding to the short text sentence. For instance, after obtaining the dependency relationships and parts of speech of each word, the nouns "textiles" and "clothing" can be reorganized with the verbs "production" and "sales" based on parallel and verb-object relationships to form clauses: "production of textiles," "production of clothing," "sales of textiles," and "sales of clothing." After processing all the short text sentences, a set of text clauses can be obtained, where each text clause is a sample describing business information.

[0065] In this embodiment of the disclosure, multiple keywords are recombined based on their parts of speech and dependencies, resulting in multiple text clauses that can serve as samples describing business information. This approach achieves higher accuracy when matching with industry information.

[0066] In one possible implementation, multiple text clauses are matched against pre-acquired industry information from multiple industries to obtain matching results, including:

[0067] For each text clause, determine the first similarity between the text clause and industry information from multiple industries. Based on the multiple first similarities, determine multiple candidate industry information corresponding to the text clause. Based on the text clause and the multiple candidate industry information, determine the matching result.

[0068] In practical applications, a text similarity model can be used to calculate the similarity between text clauses and industry information, i.e., the first similarity score. Multiple similarity scores are obtained through this calculation. The industry information corresponding to the top few positions of these similarity scores is selected as candidate industry information, representing the most similar industry information for each text clause. Then, based on the similarity between the text clause and the multiple candidate industry information, the matching result is determined. This can specifically include: using the similarity between the text clause and the multiple candidate industry information as the matching result, or further determining the matching result based on the similarity between the text clause and the multiple candidate industry information.

[0069] The text similarity model can be the ERNIE-Sim matching model. The ERNIE-Sim matching model is pre-trained using massive amounts of data and then fine-tuned for the similarity task, thus possessing powerful semantic matching capabilities and yielding relatively accurate similarity results between two sentences. It should be noted that the text similarity model can also be any other model with text similarity calculation functionality; this embodiment does not limit its application.

[0070] In this embodiment of the disclosure, candidate industry information is determined by similarity calculation, and then the matching result is determined. The calculation is simple and the result is accurate.

[0071] In one possible implementation, the matching result is determined based on the text clause and multiple candidate industry information, including:

[0072] Determine the part-of-speech tag for each word in the text clauses and multiple candidate industry information;

[0073] Determine the second similarity between text clauses and words with the same part of speech in multiple candidate industry information;

[0074] The matching result is determined based on multiple first similarity scores and multiple second similarity scores.

[0075] In practical applications, the matching result can be further determined based on the similarity between the text clause and multiple candidate industry information. This involves determining the part-of-speech (POS) of each word in the text clause and the multiple candidate industry information, calculating the similarity (second similarity) of words with the same POS in the text clause and the multiple candidate industry information, and determining the final matching result based on the similarity of words with the same POS and the similarity between the text clause and the multiple candidate industry information. Optionally, the average of the similarity of words with the same POS and the similarity between the text clause and the multiple candidate industry information can be calculated as the final matching result.

[0076] For example, the text clause "clothing sales" corresponds to two candidate industry information "clothing manufacturing" and "clothing retail," with corresponding similarity scores of [0.725, 0.712], respectively. After part-of-speech tagging, the similarity between words with the same part of speech is calculated. The similarity between "clothing" and "clothing" is 0.91, and the similarity between "sales" and "manufacturing" is 0.27; the similarity between "clothing" and "clothing" is 1.0, and the similarity between "sales" and "retail" is 0.83. Therefore, the final similarity scores are: the similarity between the text clauses "clothing sales" and "clothing manufacturing" is (0.725 + 0.91 + 0.27) / 3 = 0.635; and the similarity between "clothing sales" and "clothing retail" is (0.712 + 1.0 + 0.83) / 3 = 0.847.

[0077] In this embodiment of the disclosure, the final matching result is determined by the similarity between the text clause and multiple candidate industry information, as well as the similarity between words of the same part of speech in the text clause and multiple candidate industry information. This method has good discriminative power and can thus obtain more accurate matching results.

[0078] One possible implementation also includes:

[0079] Obtain standard industry classification information, and construct an industry tree map based on the standard industry classification information. The industry tree map includes multiple industry information and corresponding industry levels.

[0080] Industry information that meets preset conditions at the industry level in the industry tree map is identified as industry information for multiple industries.

[0081] In practical applications, to formally classify enterprises, standard industry classification information at the national standard level can be obtained. Optionally, standard industry classification information can be determined based on the "National Economic Industry Classification" published by the National Bureau of Statistics. Each major industry in the standard industry classification information is divided into four levels, for example: "Manufacturing - Food Manufacturing - Canned Food Manufacturing - Canned Aquatic Products Manufacturing". Some industries do not have a fourth level, such as "Mining - Oil and Gas Extraction - Oil Extraction". Therefore, for uniform processing, all industries can be read and divided into four levels. For industries without a fourth level, its fourth level is set to empty. After processing such as table extraction, noise reduction, and classification, a national standard industry tree map is constructed.

[0082] In one example, by processing the standard industry classification information, a level 4 industry tree map is obtained. Through analysis, it is found that the level 3 and level 4 industry levels in the industry tree map are relatively moderate in terms of granularity and text length. Therefore, the industry information at the level 3 and level 4 industry levels is selected as the industry information used for matching calculation.

[0083] It should be noted that when selecting an industry level for matching calculation, it is not limited to industry levels 3 and 4. You can set preset conditions according to specific needs and select industry information at other industry levels.

[0084] In this embodiment of the disclosure, an industry tree graph is constructed using standard industry classification information. Industry information in the industry tree graph that meets preset conditions is used as the industry information for matching calculation, which can make the industry classification of enterprises more standardized.

[0085] In one possible implementation, the industry information of the target company is determined based on the matching results, including:

[0086] When the matching results include multiple similarities, the industry information corresponding to the similarity exceeding the similarity threshold is identified as the industry information of the target company.

[0087] In practical applications, a similarity threshold is pre-set according to specific needs. When the matching results include multiple similarities, the industry information corresponding to similarities exceeding the similarity threshold is identified as the target company's industry information. After matching all text clauses of the target company, multiple matching industry information is obtained. After deduplicating these matching industry information, the industry information corresponding to the target company is obtained.

[0088] In this embodiment of the disclosure, industry information corresponding to similarity exceeding a similarity threshold is determined as the industry information of the target enterprise, which can enable the target enterprise to match industry information that meets its actual needs.

[0089] Figure 2 This is a flowchart of a method for determining industry information of an enterprise according to an embodiment of this disclosure. Figure 2 As shown, the method includes:

[0090] Step S201: Obtain the target company's business information;

[0091] Step S202: The text in the enterprise's business information is split into multiple short text sentences;

[0092] Step S203: For each text sentence, determine the part of speech of each word in the text sentence and the dependency relationship between each word. Based on the part of speech of each word and the dependency relationship between each word, determine multiple keywords corresponding to the text sentence.

[0093] Step S204: For each text phrase corresponding to multiple keywords, the multiple keywords are reorganized according to their parts of speech and the dependency relationship between them to obtain multiple text clauses;

[0094] Step S205: For each text clause, determine the first similarity between the text clause and industry information of multiple industries, and determine multiple candidate industry information corresponding to the text clause based on the multiple first similarities.

[0095] Step S206: Determine the part-of-speech of each word in the text clause and multiple candidate industry information respectively;

[0096] Step S207: Determine the second similarity between the text clause and words with the same part of speech in multiple candidate industry information;

[0097] Step S208: Determine the matching result based on multiple first similarities and multiple second similarities;

[0098] Step S209: When the matching results include multiple similarities, the industry information corresponding to the similarity exceeding the similarity threshold is determined as the industry information of the target enterprise.

[0099] The method for determining industry information of an enterprise provided in this disclosure reorganizes the text in the enterprise's business information and matches the reorganized text clauses with the industry information of multiple pre-acquired industries to determine the industry information of the target enterprise. The matching accuracy is high, and there is no need to train a classifier, saving manpower, material resources and computing resources.

[0100] Figure 3This is a schematic diagram of a method for determining industry information of an enterprise in one embodiment of this disclosure. In this embodiment, enterprise business information is a text describing the enterprise's business scope, industry information is an industry node, and standard industry classification information is an official document. The preprocessing of the enterprise business scope based on syntactic and lexical analysis includes: splitting the text in the enterprise business information to obtain multiple text sentences; for each text sentence, determining the part-of-speech of each word and the dependency relationships between words; and determining multiple keywords corresponding to the text sentence based on the part-of-speech of each word and the dependency relationships between words; and recombining the multiple keywords corresponding to each text sentence based on the part-of-speech of each keyword and the dependency relationships between the multiple keywords to obtain multiple text clauses. An industry tree graph is constructed using the official documents, and industry information in the industry tree graph that meets preset conditions is determined as industry information for multiple industries. Coarse similarity matching includes: determining the first similarity between the text clauses and the industry information of multiple industries; and determining multiple candidate industry information corresponding to the text clauses based on the multiple first similarities. Fine-grained matching based on syntactic analysis includes: determining the part-of-speech of each word in the text clause and multiple candidate industry information; determining the second similarity of words with the same part of speech in the text clause and multiple candidate industry information; and determining the matching result based on multiple first similarities and multiple second similarities. When the matching result includes multiple similarities, the industry information corresponding to similarities exceeding a similarity threshold is identified as the industry information of the target company, as shown in the "Company and Industry Node Results" figure.

[0101] Figure 4 This is a schematic diagram illustrating the reorganization of text from enterprise business information into text clauses in one embodiment of this disclosure. In this embodiment, the enterprise business information is a description of the enterprise's business scope. Text denoising includes: denoising the enterprise business scope description text to remove noisy words. Long text segmentation includes: splitting the long text into multiple short text clauses based on punctuation marks. Enterprise business content clause extraction based on syntactic and lexical analysis includes: using the NLPC lexical and syntactic analysis model to perform syntactic and lexical analysis on the short text clauses, thereby obtaining the dependency relationships between different words in the short text clauses. At the same time, each word can be tagged with its part of speech, and the short text clauses can be decomposed according to the dependency relationships and parts of speech to obtain multiple keywords. For the multiple keywords corresponding to each short text clause, the multiple keywords are reorganized according to the parts of speech of the multiple keywords and the dependency relationships between the multiple keywords to obtain multiple text clauses, forming a clause set.

[0102] Figure 5 This is a schematic diagram illustrating the matching process between text clauses and industry information in one embodiment of this disclosure. For example... Figure 5As shown, the clause set consists of multiple text clauses obtained based on the target company's business information. The ERNIE-Sim model is used to calculate the similarity between the text clauses and the industry information corresponding to the 3rd and 4th level nodes of the industry tree. Based on multiple similarities, the Top-n candidate industry nodes are determined. The NLPC part-of-speech tagging model is used to tag the parts of speech of each word in the text clauses and industry nodes, obtaining multiple word pairs with the same part of speech. A word similarity model is used to calculate the similarity of the same part of speech. The average of the similarity between the text clauses and the Top-n candidate industry nodes and the similarity of the same part of speech is calculated, resulting in multiple average similarity scores S. The industry information corresponding to the industry node with the highest similarity score S is taken as the industry information matched to the target company.

[0103] Figure 6 This is a schematic diagram of an industry information determination device for an enterprise according to an embodiment of this disclosure. Figure 6 As shown, the device for determining an enterprise's industry information may include:

[0104] Module 601 is used to acquire the business information of the target company.

[0105] Reorganization module 602 is used to reorganize the text in the enterprise's business information to obtain multiple text clauses;

[0106] The matching module 603 is used to match multiple text clauses with industry information from multiple pre-acquired industries to obtain matching results;

[0107] The determination module 604 is used to determine the industry information of the target company based on the matching results.

[0108] The industry information determination device for an enterprise provided in this disclosure acquires the business information of a target enterprise; reconstructs the text in the business information to obtain multiple text clauses; matches each of the multiple text clauses with pre-acquired industry information of multiple industries to obtain matching results; and determines the industry information of the target enterprise based on the matching results. This technical solution, by reconstructing the text in the business information and matching the reconstructed multiple text clauses with pre-acquired industry information of multiple industries, determines the industry information of the target enterprise. The matching accuracy is high, and it does not require training a classifier, saving manpower, material resources, and computing resources.

[0109] In one possible implementation, the reassembly module 602 includes a splitting unit and a reassembly unit;

[0110] The splitting unit is used to break down text in enterprise business information into multiple keywords;

[0111] The recombination unit is used to recombine multiple keywords to obtain multiple text clauses.

[0112] In one possible implementation, the unit is split, specifically for:

[0113] The text in the company's business information is split into multiple short sentences;

[0114] For each short text sentence, determine the part of speech of each word in the short text sentence and the dependency relationships between each word. Based on the part of speech of each word and the dependency relationships between each word, determine the multiple keywords corresponding to the short text sentence.

[0115] In one possible implementation, the reorganization unit is specifically used for:

[0116] For each short text sentence, multiple keywords are reorganized based on their parts of speech and dependencies to obtain multiple text clauses.

[0117] In one possible implementation, the matching module 603 is specifically used for:

[0118] For each text clause, determine the first similarity between the text clause and industry information from multiple industries. Based on the multiple first similarities, determine multiple candidate industry information corresponding to the text clause. Based on the text clause and the multiple candidate industry information, determine the matching result.

[0119] In one possible implementation, the matching module 603, when determining the matching result based on the text clause and multiple candidate industry information, is used to:

[0120] Determine the part-of-speech tag for each word in the text clauses and multiple candidate industry information;

[0121] Determine the second similarity between text clauses and words with the same part of speech in multiple candidate industry information;

[0122] The matching result is determined based on multiple first similarity scores and multiple second similarity scores.

[0123] In one possible implementation, a building module is also included for:

[0124] Obtain standard industry classification information, and construct an industry tree map based on the standard industry classification information. The industry tree map includes multiple industry information and corresponding industry levels.

[0125] Industry information that meets preset conditions at the industry level in the industry tree map is identified as industry information for multiple industries.

[0126] In one possible implementation, module 604 is specifically used for:

[0127] When the matching results include multiple similarities, the industry information corresponding to the similarity exceeding the similarity threshold is identified as the industry information of the target company.

[0128] The functions of each unit, module, or sub-module in the various devices of this disclosure embodiment can be found in the corresponding descriptions in the above method embodiments, and will not be repeated here.

[0129] The acquisition, storage, and application of user personal information involved in the technical solution disclosed herein comply with the provisions of relevant laws and regulations and do not violate public order and good morals.

[0130] According to another aspect of this disclosure, an electronic device is provided, comprising:

[0131] At least one processor; and

[0132] The memory is communicatively connected to the at least one processor; wherein,

[0133] The memory stores instructions that can be executed by the at least one processor to enable the at least one processor to perform the methods in any embodiment of this disclosure.

[0134] According to another aspect of this disclosure, a non-transitory computer-readable storage medium is provided storing computer instructions for causing a computer to perform the methods of any embodiment of this disclosure.

[0135] According to another aspect of this disclosure, a computer program product is provided, including a computer program that, when executed by a processor, implements the methods of any embodiment of this disclosure.

[0136] Figure 7 A schematic block diagram of an example electronic device 700 that can be used to implement embodiments of the present disclosure is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device may also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the present disclosure described and / or claimed herein.

[0137] like Figure 7As shown, device 700 includes a computing unit 701, which can perform various appropriate actions and processes based on a computer program stored in read-only memory (ROM) 702 or a computer program loaded into random access memory (RAM) 703 from storage unit 708. RAM 703 may also store various programs and information required for the operation of device 700. The computing unit 701, ROM 702, and RAM 703 are interconnected via bus 704. Input / output (I / O) interface 705 is also connected to bus 704.

[0138] Multiple components in device 700 are connected to I / O interface 705, including: input unit 706, such as keyboard, mouse, etc.; output unit 707, such as various types of monitors, speakers, etc.; storage unit 708, such as disk, optical disk, etc.; and communication unit 709, such as network card, modem, wireless transceiver, etc. Communication unit 709 allows device 700 to exchange information with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0139] The computing unit 701 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of the computing unit 701 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. The computing unit 701 performs the various methods and processes described above, such as the enterprise industry information determination method. For example, in some embodiments, the enterprise industry information determination method may be implemented as a computer software program tangibly contained in a machine-readable medium, such as storage unit 708. In some embodiments, part or all of the computer program may be loaded and / or installed on device 700 via ROM 702 and / or communication unit 709. When the computer program is loaded into RAM 703 and executed by the computing unit 701, one or more steps of the enterprise industry information determination method described above may be performed. Alternatively, in other embodiments, the computing unit 701 may be configured to perform the enterprise industry information determination method by any other suitable means (e.g., by means of firmware).

[0140] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SoCs), payload-programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving information and instructions from a storage system, at least one input device, and at least one output device, and transmitting information and instructions to the storage system, the at least one input device, and the at least one output device.

[0141] The program code used to implement the methods of this disclosure may be written in any combination of one or more programming languages. This program code may be provided to a processor or controller of a general-purpose computer, special-purpose computer, or other programmable information processing apparatus, such that when executed by the processor or controller, the program code causes the functions / operations specified in the flowcharts and / or block diagrams to be implemented. The program code may be executed entirely on a machine, partially on a machine, as a standalone software package partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0142] In the context of this disclosure, a machine-readable medium can be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, apparatus, or device. A machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium can be, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0143] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device for displaying information to the user (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor); and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the computer. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).

[0144] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as an information server), or computing systems that include middleware components (e.g., an application server), or computing systems that include frontend components (e.g., a user computer with a graphical user interface or web browser through which a user can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital information communication (e.g., a communication network) of any form or medium. Examples of communication networks include local area networks (LANs), wide area networks (WANs), and the Internet.

[0145] Computer systems can include clients and servers. Clients and servers are generally located far apart and typically interact via communication networks. Client-server relationships are created by computer programs running on the respective computers and having a client-server relationship with each other. Servers can be cloud servers, servers in distributed systems, or servers incorporating blockchain technology.

[0146] It should be understood that the various forms of processes shown above can be used to rearrange, add, or delete steps. For example, the steps described in this disclosure can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution disclosed in this disclosure can be achieved, and this is not limited herein.

[0147] The specific embodiments described above do not constitute a limitation on the scope of protection of this disclosure. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this disclosure should be included within the scope of protection of this disclosure.

Claims

1. A method for determining industry information of an enterprise, the method comprising: Obtain the target company's business information; The text in the enterprise's business information is reorganized to obtain multiple text clauses, including: splitting the text in the enterprise's business information to obtain keywords for each of the multiple text clauses; for each text clause, obtaining noun keywords with a parallel relationship and verb keywords with a parallel relationship, and reorganizing the noun keywords and verb keywords with a verb-object relationship to obtain multiple text clauses; The multiple text clauses are matched with industry information from multiple pre-acquired industries to obtain matching results; Based on the matching results, the industry information of the target company is determined.

2. The method according to claim 1, wherein, The process of splitting the text in the enterprise's business information yields keywords for each of several short text phrases, including: The text in the enterprise's business information is split into multiple short text sentences; For each text phrase, determine the part of speech of each word in the text phrase and the dependency relationship between each word. Based on the part of speech of each word and the dependency relationship between each word, determine multiple keywords corresponding to the text phrase.

3. The method according to claim 1, wherein, The step of matching the multiple text clauses with pre-acquired industry information from multiple industries to obtain matching results includes: For each text clause, a first similarity is determined between the text clause and the industry information of the multiple industries. Based on the multiple first similarities, multiple candidate industry information corresponding to the text clause is determined. Based on the text clause and the multiple candidate industry information, a matching result is determined.

4. The method according to claim 3, wherein, The step of determining the matching result based on the text clause and the multiple candidate industry information includes: Determine the part-of-speech tag of each word in the text clause and the multiple candidate industry information respectively; Determine the second similarity between the text clause and words with the same part of speech in the multiple candidate industry information; The matching result is determined based on multiple first similarities and multiple second similarities.

5. The method according to any one of claims 1-4, further comprising: Obtain standard industry classification information, and construct an industry tree map based on the standard industry classification information. The industry tree map includes multiple industry information and corresponding industry levels. Industry information in the industry tree map that meets preset conditions is identified as industry information for the multiple industries.

6. The method according to any one of claims 1-4, wherein, Determining the industry information of the target enterprise based on the matching result includes: When the matching results include multiple similarities, the industry information corresponding to the similarity exceeding the similarity threshold is determined as the industry information of the target enterprise.

7. An industry information determination device for an enterprise, the device comprising: The acquisition module is used to obtain the target company's business information; The restructuring module is used to restructure the text in the enterprise business information to obtain multiple text clauses. The restructuring module includes: a splitting unit, used to split the text in the enterprise business information to obtain the keywords of each of the multiple text clauses; and a restructuring unit, used to obtain noun keywords with a parallel relationship and verb keywords with a parallel relationship for each text clause, and to restructure the noun keywords and verb keywords with a verb-object relationship to obtain multiple text clauses. The matching module is used to match the multiple text clauses with industry information of multiple pre-acquired industries to obtain matching results; The determination module is used to determine the industry information of the target enterprise based on the matching results.

8. The apparatus according to claim 7, wherein, The splitting unit is specifically used for: The text in the enterprise's business information is split into multiple short text sentences; For each text phrase, determine the part of speech of each word in the text phrase and the dependency relationship between each word. Based on the part of speech of each word and the dependency relationship between each word, determine multiple keywords corresponding to the text phrase.

9. The apparatus according to claim 7, wherein, The matching module is specifically used for: For each text clause, a first similarity is determined between the text clause and the industry information of the multiple industries. Based on the multiple first similarities, multiple candidate industry information corresponding to the text clause is determined. Based on the text clause and the multiple candidate industry information, a matching result is determined.

10. The apparatus according to claim 9, wherein, When determining the matching result based on the text clause and the multiple candidate industry information, the matching module is used to: Determine the part-of-speech tag of each word in the text clause and the multiple candidate industry information respectively; Determine the second similarity between the text clause and words with the same part of speech in the multiple candidate industry information; The matching result is determined based on multiple first similarities and multiple second similarities.

11. The apparatus according to any one of claims 7-10, further comprising a construction module for: Obtain standard industry classification information, and construct an industry tree map based on the standard industry classification information. The industry tree map includes multiple industry information and corresponding industry levels. Industry information in the industry tree map that meets preset conditions is identified as industry information for the multiple industries.

12. The apparatus according to any one of claims 7-10, wherein, The determining module is specifically used for: When the matching results include multiple similarities, the industry information corresponding to the similarity exceeding the similarity threshold is determined as the industry information of the target enterprise.

13. An electronic device, characterized in that, include: At least one processor; as well as A memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor to enable the at least one processor to perform the method of any one of claims 1-6.

14. A non-transitory computer-readable storage medium storing computer instructions, characterized in that, The computer instructions are used to cause the computer to perform the method according to any one of claims 1-6.

15. A computer program product comprising a computer program that, when executed by a processor, implements the method according to any one of claims 1-6.

Images