An industrial classification and accounting method

Abstract

The invention discloses an industry classification and accounting method, which relates to the technical field of industry classification and accounting, and comprises the following steps: obtaining relevant business activities and industry-related information of typical enterprises, extracting preliminary keywords and constructing a preliminary keyword library; sorting out relevant industrial chains and expanding them into a number of industry groups; refining, supplementing, merging and deleting the industry groups to preliminarily form an industry classification; matching national economic industry codes corresponding to field subcategories with enterprises in a regional directory library to obtain a preliminary library of industry-related units; performing a vocabulary matching search on industry keywords and relevant business activities of enterprises, and preliminarily identifying the enterprises as industry-related enterprises if the vocabulary matching is successful; finally obtaining a list of enterprises classified by industry; selecting enterprises from the list of enterprises for investigation, and calculating the industry stripping coefficient of the field subcategory; calculating the industry scale by using the industry stripping coefficient and basic data of various industries in the industry; the invention is simple, highly operable and easy to promote.

Description

Technical Field

[0001] The present invention relates to the field of industrial classification and accounting technology, and in particular to an industrial classification and accounting method. Background Art

[0002] Industry represents the meso-level of the national economy, and is inextricably linked to national economic development. In a market economy, industries require appropriate, rational, and effective management to achieve optimized industrial structure, rational distribution, and healthy development. First, industrial classification is essential for industrial management; it enables categorized management. Second, industrial classification is a prerequisite for scientific industrial research and the foundation for establishing the concept of industrial structure and conducting research on it. Finally, sound industrial classification is the primary task of industrial research. Given the diverse perspectives and objectives for studying industrial structure and industrial development, and the varying purposes of industrial research and analysis, the methods of industrial classification also vary. To correctly understand and grasp the connotations and characteristics of industries, addressing the issue of industrial classification is essential.

[0003] Industrial accounting is an effective tool for monitoring and analyzing the performance of the national economy, a crucial foundation for strengthening macroeconomic management and regulation, and a key component of industrial research. With the booming rise of mass entrepreneurship and innovation, new technologies, products, new business formats, and new models are emerging in large numbers. Scientifically, truthfully, and accurately reflecting and monitoring the development of emerging industries has become a crucial topic. Emerging industries are intertwined and inseparable from traditional industries, making industrial classification and sector definition challenging. Scientifically and accurately analyzing the current status of relevant industries helps leaders at all levels understand and grasp the state of industrial development, formulate timely industrial policies, and drive regional economic growth through economic restructuring. Therefore, by leveraging big data crawling technology and applying a sequential exploratory strategy, we have established industry classification standards. Based on this, we further determine the industry peeling coefficient through enterprise surveys to calculate industry scale.

[0004] Among existing industrial classification technologies, the supervised classification method based on SVM requires a large number of samples to be labeled in advance, which places high demands on computer computing power. The industrial classification method that uses the distance between word vectors as the measure of word similarity requires the use of complex models and the calculation process is relatively cumbersome. The industrial scale accounting usually includes the income of all enterprises, resulting in inaccurate accounting of industrial scale. Summary of the Invention

[0005] The purpose of the present invention is to solve the problem that the current supervised classification method based on SVM requires a large number of samples to be manually labeled in advance and requires high computer computing power. At the same time, it overcomes the limitations of the industry classification method that uses the distance between word vectors as the measure of word similarity, which requires the use of complex models and tedious calculations. In addition, the traditional industry scale accounting method usually includes all the revenue of the enterprise, resulting in inaccurate accounting results. In order to solve the above problems, the present invention provides a new industry classification and accounting method with forward-looking and practical value.

[0006] In order to achieve the above-mentioned purpose, the present invention specifically adopts the following technical solutions:

[0007] An industry classification and accounting method includes the following steps:

[0008] Determine the concept and connotation of the industry, obtain relevant business activities of typical enterprises, industry-related information, extract preliminary keywords and build a preliminary keyword database;

[0009] Based on the preliminary keyword database, the relevant industrial chains were sorted out and expanded into several industrial groups;

[0010] Obtain a certain number of enterprises for investigation, and refine, add, merge, and delete industry groups according to the actual business scope of the enterprises to form a preliminary industry classification;

[0011] Based on the initially formed industrial classification, the national economic industry codes corresponding to its field subcategories are matched with the enterprises in the regional directory database to obtain a preliminary database of industry-related units;

[0012] Combined with the characteristics of industrial development and the relevant business activities of typical enterprises, industry keywords are summarized and sorted, and a vocabulary matching search is conducted between industry keywords and the relevant business activities of enterprises. If the vocabulary matching is successful, the enterprise is preliminarily identified as an industry-related enterprise;

[0013] De-duplicate the industry-related enterprises and the preliminary database of industry-related units to obtain a list of enterprises classified by industry;

[0014] Select companies from the list of companies to conduct surveys, aggregate the survey results belonging to the same sub-category, and calculate the industry stripping coefficient of the sub-category;

[0015] The industry scale is calculated using the industry stripping coefficient and basic data of each industry in the industry.

[0016] Furthermore, relevant business activities and industry-related information of typical enterprises are obtained through data crawler technology.

[0017] Furthermore, the Textrank algorithm is used to extract preliminary keywords of relevant business activities and industry-related information of typical enterprises to build a preliminary keyword library.

[0018] Furthermore, based on the preliminary keyword library, industries are classified according to the number of new industrial formats to form several industrial groups as major industrial categories. The preliminary keywords are classified using the Textcnn text classification algorithm to obtain several keyword groups. The word frequency statistics of the keyword groups are performed to obtain the topN keywords as industrial subgroups, where N is a natural number greater than zero.

[0019] Furthermore, the preliminary keywords are matched with the national economic industries for text similarity, and the corresponding national economic industries are found and used as industry subcategories in turn;

[0020] Through surveys of local businesses, the industry subcategories corresponding to the preliminary keywords that are most closely related to new industrial formats are subdivided;

[0021] Add new industry formats corresponding to newly emerged keywords and form industry subcategories;

[0022] Merge industry subcategories corresponding to keywords with high similarity;

[0023] Delete industry subcategories that do not conform to the operating conditions of local businesses.

[0024] Furthermore, the coefficient of proportion of industry-related enterprises in the recovered samples is calculated;

[0025] If the proportion coefficient is greater than or equal to the preset threshold, the industry stripping coefficient is 1;

[0026] If the proportion coefficient is less than the preset threshold, the proportion coefficient of the company's products and / or services is calculated, and the proportion coefficient of the company's products and / or services and the proportion coefficient of the company's income in the subcategory of this field are combined to calculate the industry stripping coefficient of the subcategory in this field.

[0027] The beneficial effects of the present invention are as follows:

[0028] The present invention provides an innovative method for industrial classification and accounting, and its main beneficial effects are as follows: First, by clearly defining the concept and connotation of the industry, obtaining typical enterprise industry-related information, business activity information to extract keywords, sorting out the industrial chain, and using survey methods to expand and revise the industrial group to obtain an accurate industrial classification. Secondly, by matching the preliminary keyword library with the relevant business operations of the enterprise, the relevant unit directory library is supplemented and improved to improve the accuracy and completeness of the name. Finally, the industry stripping coefficient is obtained through investigation, and the industry scale under the same industry subcategory is calculated, and it is highly operational and easy to promote. In addition, the present invention is particularly suitable for the classification and accounting of emerging industries, and is also effective for traditional industries. By establishing industry groups and formulating industry classifications, and using industry stripping coefficients, the current status of the industry can be effectively understood, providing a scientific basis for industrial planning. In short, the present invention has foresight and practical value, and can effectively improve the accuracy and efficiency of industry classification and accounting. Description of the accompanying drawings

[0029] Figure 1 It is a schematic flow chart of the method of the present invention. DETAILED DESCRIPTION

[0030] To make the objectives, technical solutions, and advantages of the embodiments of the present invention more clear, the technical solutions of the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Generally, the components of the embodiments of the present invention described and shown in the drawings herein can be arranged and designed in various different configurations.

[0031] Therefore, the following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the invention as claimed, but rather merely represents selected embodiments of the present invention. All other embodiments derived by persons of ordinary skill in the art based on the embodiments of the present invention without creative effort shall fall within the scope of protection of the present invention.

[0032] Example 1

[0033] like Figure 1 As shown, the present invention provides an industry classification and accounting method, comprising the following steps:

[0034] Determine the concept and connotation of the industry based on its economic attributes, production and commercial organizations, social attributes, and development characteristics. Obtain relevant business activities and industry-related information from typical companies to extract preliminary keywords and build a preliminary keyword database.

[0035] During implementation, data crawling technology was used to obtain relevant business activities and industry-related information of typical enterprises. The Textrank algorithm was used to extract preliminary keywords related to the relevant business activities and industry-related information of typical enterprises to build a preliminary keyword library.

[0036] Based on the preliminary keyword database, the relevant industrial chains were sorted out and expanded into several industrial groups;

[0037] During implementation, based on the preliminary keyword library, industries are classified according to the number of new industrial formats to form several industrial groups as major industrial categories. The Textcnn text classification algorithm is used to classify the preliminary keywords to obtain several keyword groups. The word frequency statistics of the keyword groups are performed to obtain the topN keywords as industrial subgroups, where N is a natural number greater than zero.

[0038] A certain number of enterprises are surveyed, and industry groups are refined, supplemented, merged, and deleted based on their actual business scope to ensure the accuracy and completeness of the industry groups, thereby forming a preliminary industry classification. During implementation, preliminary keywords are matched with national economic industries for text similarity to find the corresponding national economic industries, which are then used as industry subcategories. Through surveys of local enterprises, industry subcategories corresponding to preliminary keywords with a high degree of connection to new industrial formats are further subdivided. New industrial formats corresponding to newly emerging keywords are added to form industry subcategories. Industry subcategories corresponding to keywords with high similarity are merged. Industry subcategories that do not conform to the operating conditions of local enterprises are deleted.

[0039] Based on the initially formed industrial classification, the national economic industry codes corresponding to its field subcategories are matched with the enterprises in the regional directory database to obtain a preliminary database of industry-related units;

[0040] Combined with the characteristics of industrial development and the relevant business activities of typical enterprises, industry keywords are summarized and sorted, and a vocabulary matching search is conducted between industry keywords and the relevant business activities of enterprises. If the vocabulary matching is successful, the enterprise is preliminarily identified as an industry-related enterprise;

[0041] De-duplicate the industry-related legal entities and the preliminary database of industrial activity units to obtain a list of enterprises classified by industry;

[0042] Enterprises will be randomly selected from the enterprise list for investigation. Enterprise information will be collected through a combination of telephone interviews, manual distribution of questionnaires and direct reporting of online questionnaires. The survey results belonging to the same field subcategory will be summarized to calculate the industry stripping coefficient of the field subcategory. The industry stripping coefficient refers to the proportion of participation in activities in different industries under the same industry subcategory. The calculation formula is: Industry stripping coefficient = the proportion of the industry's revenue in the industry to the total revenue of the industry / the proportion of the industry's revenue outside the industry to the total revenue of the industry. Among them, the industry's revenue in the industry refers to the industry's operating income in the industry under study; the industry's revenue outside the industry refers to the industry's operating income outside the industry under study. This indicator reflects the importance of the industry in the industry. The larger the value, the greater the contribution of the industry in the industry, and vice versa. The calculation of the industry stripping coefficient is of great significance to the accuracy and completeness of industry classification. It can help us better understand the contribution of different industries in different industries and provide strong support for the formulation of industrial development strategies.

[0043] During implementation, the proportion coefficient of industry-related enterprises in the recovered samples is calculated; if the proportion coefficient is greater than or equal to the preset threshold, the industry stripping coefficient is 1; if the proportion coefficient is less than the preset threshold, the proportion coefficient of the enterprise's products and / or services is calculated, and the proportion coefficient of the enterprise's products and / or services and the proportion coefficient of the enterprise's income in the subcategory of this field are combined to calculate the industry stripping coefficient of the subcategory in this field.

[0044] The industry scale is calculated using the industry stripping coefficient and basic data of each industry in the industry.

[0045] In summary, the present invention is simple, highly operational, and easy to promote. It has a more obvious effect on the classification and accounting of emerging industries. The scale of the industry can be calculated by establishing industry groups, formulating industry classifications, and using industry stripping coefficients. For traditional industries with clear industry classifications, this method is also effective for industry accounting. According to the existing industry classifications, the industry stripping coefficients are obtained through investigations, and then the scale of the industry is calculated.

[0046] Example 2

[0047] Taking the cultural and technological industries as an example, we define the cultural and technological industries and construct a statistical classification standard system for the industries, and then calculate the scale of the industries. The specific steps are as follows:

[0048] Obtain relevant business activities and industry-related information from typical enterprises in a certain region through data crawling technology;

[0049] Use the Textrank algorithm to extract the relevant business activities and industry-related information of typical enterprises and build a preliminary keyword library;

[0050] On this basis, through a survey of typical cultural and technological enterprises, the keyword library was expanded to three industry groups, including traditional cultural industry transformation and upgrading, new cultural formats, and technological business expansion, as well as 18 sub-groups;

[0051] Several companies were selected for investigation to understand typical businesses and products that integrate corporate culture and technology. Meanwhile, based on the characteristics of regional industrial development and the need for regular industrial monitoring, industry subgroups were supplemented, deleted, merged, and adjusted to form 8 major categories, 24 medium categories, and 65 minor categories.

[0052] Establish a corresponding relationship with the "National Economic Industry Classification" (GB / T 4754-2017) to form a cultural and technological classification system with unified classification norms and standards;

[0053] According to the statistical classification standards of the cultural and technological industries, the national economic industry codes corresponding to the subcategories of the fields are matched with the regional enterprise database to obtain a preliminary database of cultural and technological industry related units;

[0054] The keywords are searched for word matching with the business activities of the enterprise. If the word matching is successful, it is preliminarily identified as a cultural and technological enterprise;

[0055] Deduplication is performed between the aforementioned cultural and technological enterprises identified through keyword matching and the preliminary database of cultural and technological enterprises to ultimately obtain a list of cultural and technological enterprises; preferably, the list of cultural and technological enterprises can be determined after seeking expert opinions;

[0056] Enterprises in the fields of digital content services, Internet culture and entertainment platforms, etc., whose proportion coefficient is greater than or equal to the preset threshold through sampling survey, will be directly included, and the industry stripping coefficient is 1;

[0057] For enterprises in the fields of Internet education and training, Internet e-commerce, etc., if the proportion coefficient is less than the preset threshold through sampling survey, the proportion of the enterprise's cultural technology products and / or services in the total operating income of the enterprise is calculated to obtain the cultural technology proportion coefficient of the field;

[0058] Summarize the survey results belonging to the same sub-category and calculate the proportion coefficient of cultural and technological products and / or services in the same sub-category in the successfully recovered samples;

[0059] The industry stripping coefficient is calculated by combining the cultural technology proportion coefficient with the proportion coefficient of the company's cultural technology products and / or services. The scale of the cultural technology industry is calculated through the company's financial data (basic data).

Claims

1. An industry classification and accounting method, characterized in that: The following steps are involved: Determine the industry concept and connotation, use the Textrank algorithm to obtain relevant business activities of typical enterprises, industry-related information to extract preliminary keywords and build a preliminary keyword library; Based on the preliminary keyword database, industries were classified based on the number of new business formats, and related industrial chains were sorted out to form several industrial groups as major industry categories. The preliminary keywords were classified using the TextCNN text classification algorithm to obtain several keyword groups, which were then expanded into several industrial groups. The keyword groups were counted for word frequency, and the top N keywords were obtained as industrial subgroups, where N is a natural number greater than zero. Obtain a certain number of enterprises for investigation, and refine, add, merge, and delete industry groups according to the actual business scope of the enterprises to form a preliminary industry classification; Based on the initially formed industrial classification, the national economic industry codes corresponding to the field subcategories are matched with the enterprises in the regional directory database to obtain a preliminary database of industry-related units; when matching the national economic industries, the preliminary keywords are matched with the national economic industries for text similarity to find the corresponding national economic industries, which are then used as industry subcategories; through surveys of local enterprises, the industry subcategories corresponding to the preliminary keywords with a high degree of connection with new industrial formats are subdivided; new industrial formats corresponding to newly emerging keywords are added to form industry subcategories; industry subcategories corresponding to keywords with high similarity are merged; and industry subcategories that do not conform to the operating conditions of local enterprises are deleted; Combined with the characteristics of industrial development and the relevant business activities of typical enterprises, industry keywords are summarized and sorted, and a vocabulary matching search is conducted between industry keywords and the relevant business activities of enterprises. If the vocabulary matching is successful, the enterprise is preliminarily identified as an industry-related enterprise; De-duplicate the industry-related enterprises and the preliminary database of industry-related units to obtain a list of enterprises classified by industry; Select companies from the list of companies to conduct surveys, aggregate the survey results belonging to the same sub-category, and calculate the industry stripping coefficient of the sub-category; The method for calculating the industry stripping coefficient is as follows: calculate the proportion coefficient of industry-related enterprises in the recovered samples; if the proportion coefficient is greater than or equal to the preset threshold, the industry stripping coefficient is 1; If the proportion coefficient is less than the preset threshold, the proportion coefficient of the enterprise's products and / or services is calculated, and the industry stripping coefficient of the sub-category in the field is calculated by combining the proportion coefficient of the enterprise's products and / or services and the proportion coefficient of the enterprise's income in the sub-category in the field; The industry scale is calculated using the industry stripping coefficient and basic data of each industry in the industry.

2. The industrial classification and accounting method according to claim 1, characterized in that: Obtain relevant business activities and industry-related information of typical enterprises through data crawling technology.

Images