A method for industry classification of listed companies

Abstract

The present invention relates to the field of industry classification of listed companies, and specifically to a method for industry classification of listed companies. The method divides the classified industries into several levels and the corresponding revenue product types, calculates the purity of the classified industries at each level according to the turnover proportion, uses the purity to reflect the turnover proportion occupied by the classified industries at each level, thereby screening out the classified industries with higher purity as the classified industries to which the listed companies belong, and calculates the total purity of the last level of classified industries to determine whether data omissions or repeated statistics occur in the process of extracting the turnover proportions corresponding to each revenue product type of the listed companies. The data omissions can be solved by adding keywords, and the repeated statistics can be solved by modifying keywords, thereby ensuring the accuracy of the data. The method solves the problem that the current one-to-one industry classification will cause inaccurate industry classification and reduce the comparability of listed companies.

Description

Technical Field

[0001] The present invention relates to the field of industry classification of listed companies, and in particular to a method for industry classification of listed companies. Background Art

[0002] In securities investment, listed companies will be classified according to different dimensions to facilitate investors to classify and compare when investing. The industry classification of listed companies refers to the classification of listed companies according to the industry in which their main business is located. It is also the most widely used classification, because different industries mean different business models, which will have a decisive impact on the fundamentals of listed companies.

[0003] At present, the more commonly used industry classifications in the market include the China Securities Regulatory Commission's industry classification, the Shenwan industry classification, and the industry classifications of various software vendors. However, all current industry classifications are one-to-one industry classifications, that is, one listed company corresponds to one industry. However, in actual production, many companies' main business is not single, and many companies even have two or more main businesses. This makes the one-to-one industry classification lead to inaccurate industry classification. The final result is that the comparability of listed companies in the same industry classification is reduced, which loses the original intention of the classification. Summary of the invention

[0004] In view of the deficiencies of the prior art, the present invention provides a method for industry classification of listed companies, which solves the problem that the current one-to-one industry classification will cause inaccurate industry classification and reduce the comparability of listed companies.

[0005] To achieve the above object, the present invention provides the following technical solutions:

[0006] A method for classifying industries of listed companies, the method comprising the following steps:

[0007] S1. Divide each industry into several levels of classified industries according to actual conditions, and classify different types of revenue products into different classified industries at the last level;

[0008] S2. Obtain the periodic reports disclosed by listed companies, and obtain the turnover proportion corresponding to each revenue product type of the listed company based on the preset keywords;

[0009] S3. Calculate the purity of each level of classified industry based on the turnover proportion corresponding to each revenue product type of the listed company;

[0010] S4. Determine whether the data in the operating income statement is abnormal based on whether the total purity of the last level of classification industry is within the purity threshold range;

[0011] If yes, then the data in the operating income table is normal and the process goes to step S6;

[0012] If not, proceed to step S5;

[0013] S5. Output the data anomaly type of the operating income table according to the total purity of the last level classification industry, and return to step S1;

[0014] S6. Calculate the corresponding purity difference according to the purity of each level of classified industries, and output several classified industries with the largest purity as the industries corresponding to listed companies.

[0015] Preferably, in step S2, the following steps are specifically included:

[0016] S21. Obtain periodic reports disclosed by listed companies;

[0017] S22. Obtaining an operating income statement from a periodic report according to a preset keyword;

[0018] S23. Obtain the corresponding turnover ratio of the total operating income from the operating income table according to the preset revenue product type.

[0019] Preferably, in step S23, the following steps are specifically included:

[0020] S231. Establish a two-dimensional coordinate system using any horizontal border and vertical border of the operating income table;

[0021] S232, obtaining the center coordinates of a preset keyword according to a two-dimensional coordinate system;

[0022] S233, extending two horizontal and vertical pointing lines from the central coordinates of the keyword and penetrating the table area of ​​the operating income table;

[0023] S234, determining whether the pointing line passes through the center coordinates of the preset keyword;

[0024] If so, delete the pointing line;

[0025] If not, keep the pointing line;

[0026] S235, obtaining the coordinates of the first intersection point where all the pointing lines intersect;

[0027] S236, extracting the content in the cell containing the first intersection from the operating income table;

[0028] S237. Obtain the proportion of the turnover of the revenue product type to the total operating income based on the extracted content and the two keywords corresponding to the coordinates of the first intersection point.

[0029] Preferably, in step S236, the following steps are specifically included:

[0030] S2361, obtaining the coordinates of all second intersections of the two pointing lines corresponding to the first intersection and the table border;

[0031] S2362, obtaining coordinates of four second intersection points closest to the first intersection point;

[0032] S2363. Establish an extraction rectangle according to the coordinates of the four second intersection points, and obtain the extraction content within the extraction rectangle.

[0033] Preferably, in step S3, the following steps are specifically included:

[0034] S31. Calculate the purity of each last-level classification industry in turn based on the turnover proportion of each revenue product type; the calculation formula for the purity of each last-level classification industry is:

[0035]

[0036] In the above formula, It represents the purity of the last level of the i-th classification industry, and the last level of the classification industry is the k-th level of the classification industry. represents the turnover proportion of the jth revenue product type belonging to the i-th last-level classification industry, where the total number of revenue product types in the i-th last-level classification industry is n;

[0037] S32. Calculate the purity of each level of classification industry based on the purity of each last level of classification industry.

[0038] Preferably, in step S5, the following steps are specifically included:

[0039] S51. Add the purity of each classified industry at the last level to obtain the total purity;

[0040] S52, judging the data anomaly type according to the purity threshold interval and the total purity;

[0041] If the total purity is greater than the maximum value of the purity threshold interval, proceed to step S53;

[0042] If the total purity is less than the minimum value of the purity threshold interval, a prompt indicating that the keyword is abnormal is output, and the process returns to step S1;

[0043] S53, select and output the revenue product types with the same turnover ratio, and return to step S1.

[0044] Preferably, in step S6, the following steps are specifically included:

[0045] S61. Calculate the purity difference of each adjacent level of classified industries in turn according to the purity of each level of classified industries, and mark each classified industry as a pre-selected industry;

[0046] S62. Determine whether the purity difference of each classified industry is less than or equal to the purity difference threshold;

[0047] If so, the higher-level classified industries in the adjacent-level classified industries will be removed from the pre-selected industries;

[0048] If not, both adjacent level classification industries will be retained in the pre-selected industries;

[0049] S63. Set a purity range, and select several parallel classified industries with the highest purity within the purity range from the pre-selected industries as the industries corresponding to the listed companies.

[0050] Preferably, in step S63, the following steps are specifically included:

[0051] S631. Select several classified industries with the highest purity that belong to the purity range from the classified industries corresponding to the first level;

[0052] S632. Determine whether the quantity of the classified industry is equal to the rated quantity;

[0053] If so, output several classified industries with the highest purity;

[0054] If not, proceed to step S633;

[0055] S633. According to the level from high to low, a number of classified industries with the highest purity within the purity range are obtained from the classified industries of each level, until the number of classified industries obtained is equal to the rated number or the level corresponding to the classified industry is the last level;

[0056] S634, determining whether the level of the currently acquired classified industry is the last level, and whether the number of the acquired classified industries is less than the rated number;

[0057] If yes, then output several classified industries that are within the purity range and have the highest purity;

[0058] If not, output a number of classified industries equal to the rated number.

[0059] Preferably, in step S32, the purity calculation formula for each level of classification industry is:

[0060]

[0061] In the above formula, Indicates The purity of the i-th classification industry in the level classification industry, Indicates The purity of the i-th classification industry in the level classification industry, There are 20,000 enterprises in the industry Classified industries.

[0062] Preferably, in step S61, the purity difference between adjacent level classification industries is calculated as follows:

[0063]

[0064] In the above formula, It represents the purity difference between two adjacent level classification industries, that is, The i-th classification industry in the classification industry of the level The purity difference of the jth classification industry in the classification industries of the level, Indicates The purity of the i-th classification industry in the classification industry of the level, Indicates The purity of the jth classification industry among the classification industries at the level.

[0065] Compared with the prior art, the present invention provides a method for industry classification of listed companies, which has the following beneficial effects:

[0066] 1. By dividing the classified industries into several levels and the corresponding revenue product types, and calculating the purity of each level of the classified industries based on the proportion of turnover, the purity is used to reflect the proportion of turnover occupied by each level of the classified industries, so as to screen out the classified industries with higher purity as the classified industries to which the listed companies belong.

[0067] 2. By calculating the total purity of the last level of classification industry, it is used to determine whether there are data omissions or repeated statistics in the process of extracting the proportion of turnover corresponding to each revenue product type of listed companies. The problem of data omissions can be solved by adding keywords, and the problem of repeated statistics can be solved by modifying keywords. In this way, the accuracy of the data can be ensured, and the subsequent calculation of the degree of attribution and the purpose of selecting the industry to which the listed company belongs can be carried out.

[0068] 3. By calculating the purity difference of each adjacent level classified industry, the difference between adjacent levels of classified industries can be determined. When the difference, that is, the purity difference, is less than the purity difference threshold, the lower level classified industry is used as the pre-selected industry to which the listed company belongs, thereby making the industry classification of listed companies more accurate. BRIEF DESCRIPTION OF THE DRAWINGS

[0069] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0070] Figure 1A flowchart of the industry classification method for listed companies of the present invention;

[0071] Figure 2 A schematic diagram of different levels of classification of industries in the present invention;

[0072] Figure 3 This is a schematic diagram of the operating income table in the periodic report of the listed company of the present invention. DETAILED DESCRIPTION

[0073] In order to make the above-mentioned purposes, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific implementation methods, so that the implementation process of how the present application uses technical means to solve technical problems and achieve technical effects can be fully understood and implemented accordingly.

[0074] Those skilled in the art can understand that all or part of the steps in the following embodiments can be completed by instructing the relevant hardware through a program, so the present application can be in the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present application can be in the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes.

[0075] Listed companies are classified into different industries, which means different business models, which will have a decisive impact on the fundamentals of listed companies. For example, different industries will have different industry cycles, different profit levels, different inventory turnover rates, etc. Take a listed company as an example, its main business is water affairs (60.08%) and gas sales (37.25%). Shenwan Industry Classification classifies it as "Water Affairs and Water Management", and the China Securities Regulatory Commission industry classification classifies it as "Water Production and Supply Industry", so it cannot fully reflect the gas business that accounts for nearly 40% of the business, resulting in a decrease in the comparability of the listed company within the same industry classification, affecting the outside world's judgment on the situation of the listed company, and thus affecting the implementation of investment and other decisions. For this reason, the present invention provides a method for classifying listed companies in the industry, classifying the same listed company into different industries, so as to more objectively reflect the comprehensive situation of the listed company. Industry classification can help us understand the market environment and industry trends, and provide a reference for the company's strategic decision-making. Secondly, industry classification can help companies better compete and cooperate with companies in the same industry. Finally, industry classification is also an important basis for the government to conduct macro-control and policy formulation. The classification method includes the following steps:

[0076] S1. Divide each industry into several levels of classified industries according to actual conditions, and classify different types of revenue products into different classified industries at the last level; e.g. Figure 2As shown, a classification industry is shown which divides various industries into three levels according to actual conditions.

[0077] S2. Obtain the periodic reports disclosed by listed companies, and obtain the turnover proportion corresponding to each revenue product type of the listed company based on the preset keywords; Figure 3 The figure shows a schematic diagram of the operating income table regarding the operating income composition in the periodic report disclosed by the listed company. The text corresponding to the boxed area in the figure contains the required keywords, such as Internet financial information, value-added telecommunications, software sales, business promotion, fund agency sales, other transaction services and proportion of operating income, etc. The position of the operating income table can be located by the operating income composition or other keywords, and then the data in the operating income table can be extracted according to the keywords of the boxed area in the figure. In step S2, the following steps are specifically included:

[0078] S21. Obtain the periodic reports disclosed by the above-mentioned company;

[0079] S22. Obtaining an operating income statement from a periodic report according to a preset keyword;

[0080] S23, according to the preset revenue product type, obtain the corresponding turnover ratio of the total operating income from the operating income table; as for how to accurately extract the data in the operating income table, since the shape and lines of the operating income table are relatively regular, a relatively simple method is adopted to realize fast and accurate data extraction, in step S23, specifically including the following steps:

[0081] S231. Establish a two-dimensional coordinate system using any horizontal border and vertical border of the operating income table;

[0082] S232, obtaining the center coordinates of a preset keyword according to a two-dimensional coordinate system;

[0083] S233, extending two horizontal and vertical pointing lines from the central coordinates of the keyword and penetrating the table area of ​​the operating income table;

[0084] S234, determining whether the pointing line passes through the center coordinates of the preset keyword;

[0085] If so, delete the pointing line;

[0086] If not, keep the pointing line;

[0087] S235, obtaining the coordinates of the first intersection point where all the pointing lines intersect;

[0088] S236, extracting the extracted content in the cell containing the first intersection from the business income table; as for how to extract the corresponding extracted content according to the first intersection, here, in combination with the two-dimensional coordinate system, the coordinate system is divided into regions by the lines on the business income table, and at this time, by identifying the position of the first intersection in different regions of the business income table, the extracted content located in different regions can be correspondingly extracted. The extracted content is the data in the table. In step S236, the following steps are specifically included:

[0089] S2361, obtaining the coordinates of all second intersections of the two pointing lines corresponding to the first intersection and the table border;

[0090] S2362, obtaining coordinates of four second intersection points closest to the first intersection point;

[0091] S2363, establishing an extraction rectangle according to the coordinates of the four second intersection points, and obtaining the extraction content within the extraction rectangle;

[0092] S237. Obtain the proportion of the turnover of the revenue product type to the total operating income based on the extracted content and the two keywords corresponding to the coordinates of the first intersection point.

[0093] S3. Calculate the purity of each level of classified industries according to the turnover proportion corresponding to each revenue product type of the listed company; in order to facilitate the subsequent judgment of whether the classified industry is the classified industry of the listed company according to the turnover proportion corresponding to each revenue product type and the classified industry to which it belongs, the turnover proportion corresponding to each revenue product type in each classified industry is accumulated, so as to know which classified industry is the main industry of the listed company, so as to facilitate the subsequent further classification of the industry of the listed company. In step S3, the following steps are specifically included:

[0094] S31. Calculate the purity of each last-level classification industry in turn based on the turnover proportion of each revenue product type; the calculation formula for the purity of each last-level classification industry is:

[0095]

[0096] In the above formula, It represents the purity of the last level of the i-th classification industry, and the last level of the classification industry is the k-th level of the classification industry. It represents the proportion of the turnover of the jth revenue product type belonging to the ith last-level classification industry, and the total number of revenue product types in the ith last-level classification industry is n.

[0097] S32. Calculate the purity of each level of classification industry based on the purity of each last level of classification industry; the purity calculation formula for each level of classification industry is:

[0098]

[0099] In the above formula, Indicates The purity of the i-th classification industry in the level classification industry, Indicates The purity of the i-th classification industry in the level classification industry, There are 20,000 enterprises in the industry Classified industries.

[0100] S4. Determine whether the data in the operating income table is abnormal based on whether the total purity of the last level of classification industry is within the purity threshold range. This is because some revenue product types in the operating income table may not be identified, resulting in the total purity of the last level of classification industry not being 100%. Generally, the purity threshold range is 90% to 100%. If it is outside this range, it indicates that the data is abnormal and needs to be resolved by re-adding keywords, etc.

[0101] If yes, then the data in the operating income table is normal and the process goes to step S6;

[0102] If not, proceed to step S5;

[0103] S5. Output the data anomaly type of the operating income table according to the total purity of the last level of classified industry, and return to step S1; automatic identification of data anomaly type helps to assist manual anomaly identification, thereby facilitating manual related adjustments and improvements, so that the classified industries of listed companies obtained subsequently are more accurate. In step S5, the following steps are specifically included:

[0104] S51. Add the purity of each classified industry at the last level to obtain the total purity;

[0105] S52, judging the data anomaly type according to the purity threshold interval and the total purity;

[0106] If the total purity is greater than the maximum value of the purity threshold interval, proceed to step S53;

[0107] If the total purity is less than the minimum value of the purity threshold interval, a prompt indicating that the keyword is abnormal is output, and the process returns to step S1;

[0108] S53, selecting and outputting the revenue product types with the same turnover ratio, and returning to step S1, which indicates that some turnover ratios are calculated repeatedly.

[0109] S6. Calculate the corresponding purity difference according to the purity of each level of classified industries, and output several classified industries with the largest purity as the industries corresponding to the listed companies, so as to select several most representative classified industries as the classified industries to which the listed companies belong. In step S6, the following steps are specifically included:

[0110] S61. According to the purity of each level of classification industry, the purity difference of each adjacent level of classification industry is calculated in turn, and each classification industry is marked as a pre-selected industry; the calculation formula for the purity difference of each adjacent level of classification industry is:

[0111]

[0112] In the above formula, It represents the purity difference between two adjacent level classification industries, that is, The i-th classification industry in the classification industry of the level The purity difference of the jth classification industry in the classification industries of the level, Indicates The purity of the i-th classification industry in the classification industry of the level, Indicates The purity of the jth classification industry in the classification industries of the level;

[0113] S62. Determine whether the purity difference of each classified industry is less than or equal to the purity difference threshold;

[0114] If yes, then remove the higher-level industries in the adjacent-level industries from the pre-selected industries; for example, the purity difference of the third-level industries is 70%, while the purity difference of the second-level industries is 75%. The higher the level, the more the high-level industries are generalized to the low-level industries, that is, the low-level industries are subdivided from the high-level industries. Here, the industries with a purity difference of 70% are selected, which is more accurate and representative, because the second-level industries contain a large number of third-level industries with lower purity, so the accuracy is poor.

[0115] If not, both adjacent level classification industries will be retained in the pre-selected industries;

[0116] S63, set the purity interval, and select several parallel classified industries with the largest purity within the purity interval from the pre-selected industries as the industries corresponding to the listed companies; after calculating the purity values ​​of each classified industry, it is necessary to further select the industries that can be classified as the listed companies by the purity values. Here, the method of screening the classified industries with larger purity values ​​layer by layer from high level to low level is adopted. In step S63, the following steps are specifically included:

[0117] S631. Select several classified industries with the highest purity that belong to the purity range from the classified industries corresponding to the first level;

[0118] S632. Determine whether the quantity of the classified industry is equal to the rated quantity;

[0119] If so, output several classified industries with the highest purity;

[0120] If not, proceed to step S633;

[0121] S633. According to the level from high to low, a number of classified industries with the highest purity within the purity range are obtained from the classified industries of each level, until the number of classified industries obtained is equal to the rated number or the level corresponding to the classified industry is the last level;

[0122] S634, determining whether the level of the currently acquired classified industry is the last level, and whether the number of the acquired classified industries is less than the rated number;

[0123] If yes, then output several classified industries that are within the purity range and have the highest purity;

[0124] If not, output a number of classified industries equal to the rated number.

[0125] When implementing the method mentioned in the present invention, it is necessary to first set up several levels of classified industries, and classify different revenue product types into different classified industries at the last level, then obtain the periodic reports disclosed by listed companies and the proportion of turnover corresponding to the revenue product type, and then calculate the purity of each level of classified industries, and then judge whether the acquired data has any abnormalities based on the purity. If there are any abnormalities, output the reasons according to the corresponding total purity, that is, the type of data abnormality. If there are no abnormalities in the data, calculate and screen out the classified industries to which the listed industries belong based on the purity of each level of classified industries.

[0126] The above implementation methods have been described in detail. Specific examples are used herein to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only used to help understand the method of the present invention and its core idea. At the same time, for those skilled in the art, according to the idea of ​​the present invention, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as a limitation on the present invention.

Claims

1. A method for classifying listed companies by industry, characterized in that: The classification method includes the following steps: S1. Divide each industry into several levels of classified industries according to actual conditions, and classify different types of revenue products into different classified industries at the last level; S2. Obtain the periodic reports disclosed by listed companies, and obtain the turnover proportion corresponding to each revenue product type of the listed company based on the preset keywords; S3. Calculate the purity of each level of classified industry based on the turnover proportion corresponding to each revenue product type of the listed company; In step S3, the following steps are specifically included: S31. Calculate the purity of each last-level classification industry in turn based on the turnover proportion of each revenue product type; the calculation formula for the purity of each last-level classification industry is: In the above formula, It represents the purity of the last level of the i-th classification industry, and the last level of the classification industry is the k-th level of the classification industry. represents the turnover proportion of the ath revenue product type belonging to the ith last-level classification industry, where the total number of revenue product types in the ith last-level classification industry is n; S32. Calculate the purity of each level of classification industry based on the purity of each last level of classification industry; In step S32, the purity calculation formula for each level of classification industry is: In the above formula, Indicates The purity of the i-th classification industry in the level classification industry, Indicates The purity of the jth classification industry in the level classification industry, There are 20,000 enterprises in the industry Classified industries; S4. Determine whether the data in the operating income statement is abnormal based on whether the total purity of the last level of classification industry is within the purity threshold range; If yes, then the data in the operating income table is normal and the process goes to step S6; If not, proceed to step S5; S5. Output the data anomaly type of the operating income table according to the total purity of the last level classification industry, and return to step S1; In step S5, the following steps are specifically included: S51. Add the purity of each classified industry at the last level to obtain the total purity; S52, judging the data anomaly type according to the purity threshold interval and the total purity; If the total purity is greater than the maximum value of the purity threshold interval, proceed to step S53; If the total purity is less than the minimum value of the purity threshold interval, a prompt indicating that the keyword is abnormal is output, and the process returns to step S1; S53, selecting and outputting the revenue product types with the same turnover ratio, and returning to step S1; S6. Calculate the corresponding purity difference according to the purity of each level of classified industries, and output several classified industries with the largest purity as the industries corresponding to the listed companies; In step S6, the following steps are specifically included: S61. Calculate the purity difference of each adjacent level of classified industries in turn according to the purity of each level of classified industries, and mark each classified industry as a pre-selected industry; S62. Determine whether the purity difference of each classified industry is less than or equal to the purity difference threshold; If so, the higher-level classified industries in the adjacent-level classified industries will be removed from the pre-selected industries; If not, both adjacent level classification industries will be retained in the pre-selected industries; S63. Set a purity range, and select several parallel classified industries with the highest purity within the purity range from the pre-selected industries as the industries corresponding to the listed companies; In step S63, the following steps are specifically included: S631. Select several classified industries with the highest purity that belong to the purity range from the classified industries corresponding to the first level; S632. Determine whether the quantity of the classified industry is equal to the rated quantity; If so, output several classified industries with the highest purity; If not, proceed to step S633; S633. According to the level from high to low, a number of classified industries with the highest purity within the purity range are obtained from the classified industries of each level, until the number of classified industries obtained is equal to the rated number or the level corresponding to the classified industry is the last level; S634, determining whether the level of the currently acquired classified industry is the last level, and whether the number of the acquired classified industries is less than the rated number; If yes, then output several classified industries that are within the purity range and have the highest purity; If not, output a number of classified industries equal to the rated number.

2. The classification method according to claim 1, characterized in that: In step S2, the following steps are specifically included: S21. Obtain periodic reports disclosed by listed companies; S22. Obtaining an operating income statement from a periodic report according to a preset keyword; S23. Obtain the corresponding turnover ratio of the total operating income from the operating income table according to the preset revenue product type.

3. The classification method according to claim 2, characterized in that: In step S23, the following steps are specifically included: S231. Establish a two-dimensional coordinate system using any horizontal border and vertical border of the operating income table; S232, obtaining the center coordinates of a preset keyword according to a two-dimensional coordinate system; S233, extending two horizontal and vertical pointing lines from the central coordinates of the keyword and penetrating the table area of ​​the operating income table; S234, determining whether the pointing line passes through the center coordinates of the preset keyword; If so, delete the pointing line; If not, keep the pointing line; S235, obtaining the coordinates of the first intersection point where all the pointing lines intersect; S236, extracting the content in the cell containing the first intersection from the operating income table; S237. Obtain the proportion of the turnover of the revenue product type to the total operating income based on the extracted content and the two keywords corresponding to the coordinates of the first intersection point.

4. The classification method according to claim 3, characterized in that: In step S236, the following steps are specifically included: S2361, obtaining the coordinates of all second intersections of the two pointing lines corresponding to the first intersection and the table border; S2362, obtaining coordinates of four second intersection points closest to the first intersection point; S2363. Establish an extraction rectangle according to the coordinates of the four second intersection points, and obtain the extraction content within the extraction rectangle.

5. The classification method according to claim 1, characterized in that: In step S61, the calculation formula for the purity difference of each adjacent level classification industry is: In the above formula, It represents the purity difference between two adjacent level classification industries, that is, The first level of the classification industry Classification Industry and The first level of the classification industry The purity of each industry is poor. Indicates The first level of the classification industry The purity of each industry category, Indicates The first level of the classification industry The purity of a classified industry.

Images