Operation feature evaluation method and device based on high-dimensional spatial vector

Abstract

The invention discloses an operation feature evaluation method and device based on a high-dimensional spatial vector. The method comprises the steps of obtaining a plurality of operation feature data sets of a plurality of target companies; according to a preset reference scoring standard, scoring a plurality of operating features in the currently processed operating feature data set to obtain a plurality of basic scores; performing reference processing on the plurality of basic scores to obtain a plurality of target scores corresponding to the plurality of target companies; mapping the plurality of target scores in a preset score interval table to obtain a plurality of target mapping intervals; according to the multiple target mapping intervals, one-hot coding is carried out, and multiple target feature vectors are obtained; performing high-dimensional space clustering on the plurality of target feature vectors according to a density clustering algorithm to obtain a high branch company group and a low branch company group; and determining a plurality of high-score and low-score operation features according to the plurality of basic scores and the high-score and low-score company groups. According to the invention, the accuracy and comprehensiveness of company operation characteristic evaluation can be improved.

Description

Technical Field

[0001] The present application relates to the field of data processing technology, and in particular to a method and device for evaluating business characteristics based on high-dimensional space vectors. Background Art

[0002] The operating and financial conditions of a company are related to the survival and development of the company itself, and are more helpful in observing the development of the fields in which different companies are located. Therefore, regulators are in urgent need of in-depth understanding of the specific conditions of these companies. Regulators need to analyze the main characteristics that affect the company's operations, demonstrate and promote the characteristics that excellent companies can learn from, and play a positive role in the economic operation of various industries and regions; and analyze the main characteristics that bad companies can learn from, so as to be able to conduct a comprehensive analysis and evaluation of the company based on positive and negative characteristics.

[0003] However, the existing business characteristics evaluation system has many limitations and cannot meet the increasingly sophisticated and professional decision-making requirements of regulatory agencies. Whether in terms of the comprehensiveness of data, the accuracy of analysis, or the effectiveness of providing decision support, there is a large gap with actual needs. Therefore, it is urgent to develop an efficient, accurate and diversified business characteristics evaluation system that can meet the above-mentioned diverse needs. Summary of the invention

[0004] This application provides a method and device for evaluating business characteristics based on high-dimensional space vectors, which maps multiple business characteristics of listed companies to high-dimensional space vectors, uses machine learning to perform spatial density clustering, so that listed companies with similar feature vectors will form natural aggregation groups, and perform principal component analysis on the characteristics under the same aggregation group to obtain multiple main business characteristics of high-scoring and low-scoring company groups, which is conducive to analyzing the company's business conditions and providing decision-making for the company's operations.

[0005] In a first aspect, the present application provides a business characteristic evaluation method based on a high-dimensional space vector, which is applied to a server of a business characteristic evaluation system, and the method comprises: Acquire multiple operating characteristic data sets corresponding to multiple target companies, where a single operating characteristic data set includes multiple operating characteristics in multiple dimensions; The following processing is performed on each of the plurality of operating characteristic data sets to obtain a plurality of target scores corresponding to the plurality of target companies: scoring the plurality of operating characteristics in the currently processed operating characteristic data set according to a preset reference scoring standard to obtain a plurality of basic scores; and performing reference processing on the plurality of basic scores to obtain a plurality of target scores corresponding to the plurality of target companies, wherein the reference processing includes weighted summation and normalization; Mapping the multiple target scores in a preset score interval table to obtain multiple target mapping intervals; and performing one-hot encoding according to the multiple target mapping intervals to obtain multiple target feature vectors corresponding to the multiple target companies; Performing high-dimensional spatial clustering on the multiple target feature vectors according to a density clustering algorithm to obtain a target clustering result, wherein the target clustering result is used to indicate a high-score company group and a low-score company group; A plurality of high-scoring business characteristics and a plurality of low-scoring business characteristics are determined based on the plurality of basic scores, the high-scoring company group, and the low-scoring company group.

[0006] In a second aspect, an embodiment of the present application provides a business characteristic evaluation device based on a high-dimensional space vector, which is applied to a server of a business characteristic evaluation system, and the device includes: An acquisition unit, used to acquire multiple operating characteristic data sets corresponding to multiple target companies, wherein a single operating characteristic data set includes multiple operating characteristics under multiple dimensions; A processing unit, for performing the following processing on each of the multiple business feature data sets to obtain multiple target scores corresponding to the multiple target companies: scoring the multiple business features in the currently processed business feature data set according to a preset reference scoring standard to obtain multiple basic scores; and, performing reference processing on the multiple basic scores to obtain multiple target scores corresponding to the multiple target companies, the reference processing including weighted summation and normalization; and, for mapping the multiple target scores in a preset scoring interval table to obtain multiple target mapping intervals; and, performing one-hot encoding according to the multiple target mapping intervals to obtain multiple target feature vectors corresponding to the multiple target companies; and, for performing high-dimensional space clustering on the multiple target feature vectors according to a density clustering algorithm to obtain a target clustering result, the target clustering result being used to indicate a high-score company group and a low-score company group; A determination unit is used to determine a plurality of high-score business characteristics and a plurality of low-score business characteristics according to the plurality of basic scores, the high-score company group, and the low-score company group.

[0007] In a third aspect, an embodiment of the present application provides an electronic device, comprising a processor, a memory, and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the processor, and the program includes instructions for executing the steps in the first aspect of the embodiment of the present application.

[0008] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium having a computer program / instruction stored thereon, and the computer program / instruction is executed by a processor to implement the steps of the method described in the first aspect above.

[0009] It can be seen that in the embodiment of the present application, the server obtains multiple business feature data sets corresponding to multiple target companies, and a single business feature data set includes multiple business features under multiple dimensions; the following processing is performed on each business feature data set in the multiple business feature data sets to obtain multiple target scores corresponding to the multiple target companies: the multiple business features in the currently processed business feature data set are scored according to a preset reference scoring standard to obtain multiple basic scores; and, the multiple basic scores are reference processed to obtain multiple target scores corresponding to the multiple target companies, and the reference processing includes weighted summation and normalization; the multiple target scores are mapped in a preset scoring interval table to obtain multiple target mapping intervals; and, one-hot encoding is performed according to the multiple target mapping intervals to obtain multiple target feature vectors corresponding to the multiple target companies; the multiple target feature vectors are clustered in high-dimensional space according to a density clustering algorithm to obtain a target clustering result, and the target clustering result is used to indicate a high-score company group and a low-score company group; multiple high-score business features and multiple low-score business features are determined according to the multiple basic scores, the high-score company group and the low-score company group. In this way, compared with the existing operating characteristics evaluation schemes with one-sided data analysis and low accuracy, this application maps multiple operating characteristics of listed companies into high-dimensional space vectors, uses machine learning methods to perform spatial density clustering, and then obtains the main operating characteristics of high-quality and low-quality companies based on the clustering results, thereby improving the accuracy and comprehensiveness of the operating characteristics evaluation. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0011] Figure 1 It is a system architecture diagram of a business characteristic evaluation system provided in an embodiment of the present application; Figure 2 It is a structural schematic diagram of an electronic device provided in an embodiment of the present application; Figure 3 It is an overall flow chart of a business characteristic evaluation method based on high-dimensional space vectors provided in an embodiment of the present application; Figure 4 It is a flowchart of the steps of a method for evaluating business characteristics based on high-dimensional space vectors provided in an embodiment of the present application; Figure 5It is an overall flow chart of another business characteristic evaluation method based on high-dimensional space vectors provided in an embodiment of the present application; Figure 6 It is a schematic diagram of an interface of a business characteristics evaluation system provided in an embodiment of the present application; Figure 7 is a schematic diagram of an interface of another business characteristic evaluation system provided in an embodiment of the present application; Figure 8 It is a functional module diagram of a business characteristic evaluation device based on high-dimensional space vectors provided in an embodiment of the present application. DETAILED DESCRIPTION

[0012] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0013] The terms "first", "second", etc. in the specification and claims of this application and the above-mentioned drawings are used to distinguish different objects, rather than to describe a specific order. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes a series of steps or units is not limited to the listed steps or units, but optionally includes steps or units that are not listed, or optionally includes other steps or units inherent to these processes, methods, products or devices.

[0014] Reference to "embodiments" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present application. The appearance of the phrase in various locations in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0015] In the embodiments of the present application, "and / or" describes the association relationship of the associated objects, indicating that three relationships may exist. For example, A and / or B can represent the following three situations: A exists alone; A and B exist at the same time; B exists alone. Among them, A and B can be singular or plural.

[0016] In the embodiment of the present application, the symbol " / " can indicate that the objects associated with each other are in an "or" relationship. In addition, the symbol " / " can also indicate a division sign, that is, performing a division operation. For example, A / B can indicate A divided by B.

[0017] In the embodiments of the present application, "at least one item" or similar expressions refer to any combination of these items, including any combination of single items or plural items, and refer to one or more, and multiple refers to two or more. For example, at least one item of a, b, or c can represent the following seven situations: a, b, c, a and b, a and c, b and c, a, b, and c. Among them, each of a, b, and c can be an element or a set containing one or more elements.

[0018] In the embodiments of the present application, "equal to" can be used in conjunction with greater than, and is applicable to the technical solution adopted when greater than, and can also be used in conjunction with less than, and is applicable to the technical solution adopted when less than. When equal to is used in conjunction with greater than, it is not used in conjunction with less than; when equal to is used in conjunction with less than, it is not used in conjunction with greater than.

[0019] The existing business characteristics evaluation system has many limitations and cannot meet the increasingly sophisticated and professional decision-making requirements of regulatory agencies. Whether in terms of the comprehensiveness of data, the accuracy of analysis, or the effectiveness of decision support, there is a large gap with actual needs.

[0020] In order to solve the above problems, the present application provides a method and device for evaluating business characteristics based on high-dimensional space vectors. The embodiments of the present application are described in detail below in conjunction with the accompanying drawings.

[0021] See also Figure 1 , Figure 1 This is a system architecture diagram of a business characteristic evaluation system provided in an embodiment of the present application. The business characteristic evaluation system 100 includes a terminal device 110 and a server 120, wherein the server 120 includes a data acquisition module 121, a processing module 122 and a communication module 123.

[0022] The terminal device 110 is a device for users to interact with the business characteristic evaluation system 100, including but not limited to computers, smart phones, tablet computers, etc. The user sends operation instructions to the system through the terminal device 110, such as entering the name of the company to be evaluated, selecting the evaluation indicator dimension, setting the evaluation time range, etc. At the same time, the terminal device 110 receives and displays the results processed by the system, such as the company's business characteristic score, clustering results, effective characteristic analysis report, trend analysis chart, etc.

[0023] The data collection module 121 in the server 120 is used to collect data related to the company's operating characteristics from various data sources. Data sources include securities trading platforms, company official websites, financial information websites, and internal information systems of enterprises. This module can use web crawler technology to crawl data from public websites, connect with the internal system of the enterprise through a data interface to obtain internal operating data, and can also receive production and operation data collected by sensors and other equipment.

[0024] The processing module 122 in the server 120 is used to process the collected data according to preset algorithms and rules. Specifically, it includes scoring, weighted summation, normalization and other operations on the business characteristic data to calculate the target score of each company; using interval mapping and one-hot encoding technology to convert the score to generate a feature matrix; using DBSCAN and other density clustering algorithms to perform high-dimensional space clustering on the feature matrix to divide the high-score company group and the low-score company group; using variance calculation and other methods to screen out the effective features of each group, and analyze the trend of the features over time.

[0025] The communication module 123 in the server 120 is responsible for data transmission and communication between the server 120 and the terminal device 110, and between the server 120 and other external systems (such as data source systems). It uses standard network communication protocols to ensure the security and stability of data during transmission; at the same time, it performs packaging, unpacking and error checking on the transmitted data to ensure the integrity of the data.

[0026] Specifically, see Figure 2 , Figure 2 is a schematic diagram of the structure of an electronic device provided in an embodiment of the present application, for executing Figure 1 The operating characteristics evaluation system in Figure 2 As shown, the electronic device 20 may include one or more of the following components: a memory 23, a processor 21, a communication bus 30, a communication interface 22, and one or more programs 231. The one or more programs 231 are stored in the memory 23 and are configured to be executed by the processor 21. The one or more programs 231 include instructions for executing any step in the following method embodiments. In a specific implementation, the processor 21 is used to execute any step in the following method embodiments, and when performing data transmission such as sending, the communication interface 22 may be selectively called to complete the corresponding operation.

[0027] The electronic device 20 may be a mobile phone terminal, a tablet computer, a notebook computer, or a wearable smart device.

[0028] It is understandable that the electronic device 20 may include more or fewer structural elements than those in the above structural block diagram, for example, including a power module, physical buttons, a Wi-Fi module, a speaker, a Bluetooth module, a sensor, etc., which are not limited here.

[0029] See also Figure 3 , Figure 3 This is an overall flow chart of a business characteristic evaluation method based on high-dimensional space vectors provided in an embodiment of the present application, which is applied to Figure 1 The server 120 in Figure 3 As shown, the method comprises the following steps: Step S310, obtaining multiple operating characteristic data sets corresponding to multiple target companies, where a single operating characteristic data set includes multiple operating characteristics in multiple dimensions.

[0030] In a possible embodiment, the obtaining of a plurality of business characteristic data sets corresponding to a plurality of target companies includes: Acquire multiple characteristic data corresponding to the multiple target companies; Performing data cleaning processing on the plurality of feature data, wherein the data cleaning processing includes missing data processing and abnormal data processing; Screening the plurality of feature data after the data cleaning process according to the plurality of business features under the plurality of dimensions, so that only the plurality of feature data corresponding to the plurality of business features are retained; The screened plurality of characteristic data are stored in the operating characteristic data set of the corresponding target company, so as to obtain a plurality of operating characteristic data sets corresponding to the plurality of target companies.

[0031] Data cleaning includes removing business characteristics with serious data missing and excluding them from subsequent evaluation strategies; and first trying to correct extreme outliers, and setting them to default values ​​if they cannot be corrected. For example, if the missing quarterly revenue data in a listed company's financial statements is too high, consider removing the dimension during analysis; for abnormal net profit data (such as sudden and significant growth or decline), first check whether it is a data entry error. If it cannot be corrected, it can be set to the default value such as the industry average.

[0032] Among them, the multiple operating characteristics under multiple dimensions include multiple first operating characteristics under the timing indicator dimension and multiple second operating characteristics under the value indicator dimension. The timing indicators are used to evaluate the operating status of the target company from a technical dimension, and the value indicators are used to evaluate the operating status of the target company from a financial dimension.

[0033] Specifically, the timing indicator evaluates the company's operating status from a technical perspective, reflecting the situation of the stock's technical indicators, including five indicators: capital, short-term indicators, price indicators, stock price forecasts, and company size. That is, it is the weighted score of multiple first operating characteristics. The higher the score, the better the company's securities market performance, and the stronger the support of the company's stock technical factors for the stock.

[0034] Specifically, the value index reflects the company's financial indicators and other fundamentals, including five indicators: financial capabilities, valuation indicators, institutional ratings, historical dividends, and risk control. That is, it is the weighted scores of multiple second operating characteristics. The higher the score, the better the company's operating performance and other aspects, and the more investment prospects the company has.

[0035] It can be understood that multiple dimensions can be adjusted in a timely manner based on the company's operating status and economic operation conditions, rather than just using the timing indicator dimension and value indicator dimension provided in this embodiment. In addition, the categories and quantities of multiple operating characteristics included in different dimensions can also be set and adjusted in advance based on objective and subjective conditions. The embodiment of this application only gives an example of the categories and quantities of multiple operating characteristics under multiple dimensions, and does not limit this.

[0036] Step S320, performing the following processing on each of the multiple business characteristic data sets to obtain multiple target scores corresponding to the multiple target companies: scoring the multiple business characteristics in the currently processed business characteristic data set according to a preset reference scoring standard to obtain multiple basic scores; and performing reference processing on the multiple basic scores to obtain multiple target scores corresponding to the multiple target companies, the reference processing including weighted summation and normalization.

[0037] In a possible embodiment, the multiple business characteristics in the currently processed business characteristic data set are scored according to a preset reference scoring standard to obtain multiple basic scores, including: Scoring a plurality of first business characteristics in the currently processed business characteristic data set according to a preset reference scoring standard to obtain a plurality of first basic scores; and, The plurality of second business characteristics in the currently processed business characteristic data set are scored according to a preset reference scoring standard to obtain a plurality of second basic scores.

[0038] Among them, the preset reference scoring standards can be shown in the following reference scoring standard table: Reference scoring standard table

[0039] Exemplarily, the target company includes target company A, target company B, and target company C. Among them, the multiple first operating characteristics and multiple first basic scores corresponding to target company A can be: capital side (good)-80 points, short-term indicators (qualified)-60 points, value indicators (good)-80 points, stock price forecast (good)-80 points, company size (large)-90 points; the multiple second operating characteristics and multiple second basic scores corresponding to target company A can be: financial ability (excellent)-90 points, valuation indicators (good)-80 points, institutional ratings (good)-80 points, historical dividends (medium)-79 points, risk control (good)-80 points. By analogy, based on the preset reference scoring criteria, multiple basic scores corresponding to multiple operating characteristics of target company B and target company C in multiple dimensions can be obtained. It should be noted that the basic scores of multiple operating characteristics corresponding to multiple target companies can be scored by the server according to the scoring criteria given in this example. In addition, it can also be imported while importing multiple feature data. The basic scores that have been determined in advance are imported. This application does not limit the source of the basic scores of operating characteristics.

[0040] In a possible embodiment, the target score includes a target timing score and a target value score; the reference processing of the multiple basic scores to obtain multiple target scores corresponding to the multiple target companies includes: The plurality of first basic scores and the plurality of second basic scores corresponding to the currently processed business characteristic data set are weighted and summed respectively to obtain a reference timing score and a reference value score of the target company corresponding to the currently processed business characteristic data set; Normalizing the reference timing score of each target company in the multiple target companies according to the multiple reference timing scores corresponding to the multiple target companies to obtain the multiple target timing scores corresponding to the multiple target companies; and The reference value score of each target company in the multiple target companies is normalized according to the multiple reference value scores corresponding to the multiple target companies to obtain the multiple target value scores corresponding to the multiple target companies.

[0041] In a possible embodiment, the weighted sum of the multiple first basic scores and the multiple second basic scores corresponding to the currently processed business characteristic data set to obtain a reference timing score and a reference value score of the target company corresponding to the currently processed business characteristic data set includes: Obtaining a plurality of first preset weights corresponding to the plurality of first business characteristics, and obtaining a plurality of second preset weights corresponding to the plurality of second business characteristics; The multiple first basic scores are weighted and summed according to the multiple first preset weights to obtain the reference timing score of the target company; and the multiple second basic scores are weighted and summed according to the multiple second preset weights to obtain the reference value score of the target company.

[0042] The formula for normalization is as follows: X1=

[0043] Among them, X1 is the target timing score corresponding to the target company, and x is the reference timing score corresponding to the target company. Represents the minimum value among multiple reference timing scores corresponding to all target companies. Represents the maximum value of multiple reference timing scores corresponding to all target companies.

[0044] For example, the timing index includes five sub-indicators: funding, short-term index, price index, stock price forecast, and company size, with weights of 0.2, 0.15, 0.2, 0.2, and 0.25 respectively; the value index includes five sub-indicators: financial capability, valuation index, institutional rating, historical dividend, and risk control, with weights of 0.25, 0.2, 0.15, 0.15, and 0.25 respectively. Then the reference timing score of target company A = 80×0.2+60×0.15+80×0.2+80×0.2+90×0.25=80.5 points, and the reference value score = 80×0.25+80×0.2+80×0.15+70×0.15+80×0.25=81 points. In this way, based on the same calculation method, the reference timing score of target company B is 50 points and the reference value score is 45 points, and the reference timing score of target company C is 85 points and the reference value score is 88 points.

[0045] Further, illustratively, according to the normalization formula, the target timing score of target company A can be obtained as follows: =87 points, Target Value Score of Target Company A= = 84 points. Thus, based on the same calculation method, the target timing score of target company B is 0 points, the target value score is 0 points, and the target timing score of target company C is 100 points, the target value score is 100 points.

[0046] It can be seen that in this embodiment, the server scores and calculates the operating characteristic data of each target company according to different dimensions, and obtains multiple target scores for multiple dimensions corresponding to each target company by weighted summing and normalizing the basic scores corresponding to multiple operating characteristics under each dimension. In this way, the pre-scoring operation is implemented, which is conducive to the subsequent high-dimensional space vector mapping based on multiple target scores to analyze the main operating characteristics of different companies and improve the accuracy of operating characteristic evaluation.

[0047] Step S330, mapping the multiple target scores in a preset score interval table to obtain multiple target mapping intervals; and performing one-hot encoding according to the multiple target mapping intervals to obtain multiple target feature vectors corresponding to the multiple target companies.

[0048] In a possible embodiment, the target mapping interval includes a target timing interval corresponding to the target timing score and a target value interval corresponding to the target value score; performing one-hot encoding according to the multiple target mapping intervals to obtain multiple target feature vectors corresponding to the multiple target companies includes: Determine the target timing interval and the target value interval corresponding to each target company among the multiple target companies; The target timing interval and the target value interval corresponding to each target company are used to obtain the multiple target feature vectors: One-hot encoding is performed according to the target timing interval and the scoring interval table corresponding to the target company to obtain a first sub-feature vector; and one-hot encoding is performed according to the target value interval and the scoring interval table corresponding to the target company to obtain a second sub-feature vector; The target feature vector corresponding to the target company is obtained according to the first sub-feature vector and the second sub-feature vector.

[0049] Exemplarily, the server can set a score interval table based on the score size and score gap of multiple target timing scores and multiple target value scores corresponding to multiple target companies, for example, 0-100 points are divided into 4 intervals: 0-25 points are poor, 26-50 points are qualified, 51-75 points are good, and 76-100 points are excellent. Furthermore, the target timing interval corresponding to target company A is excellent and the target value interval is good, the target timing interval corresponding to target company B is poor and the target value interval is poor, and the target timing interval corresponding to target company C is excellent and the target value interval is excellent.

[0050] Furthermore, one-hot encoding is performed based on multiple target mapping intervals corresponding to multiple target companies to obtain multiple target feature vectors.

[0051] Among them, One-hot encoding, also known as one-bit effective encoding, is a technique commonly used to encode categorical data in data processing and machine learning. The core idea of ​​One-hot encoding is to convert categorical variables into a binary vector representation. For categorical variables with n different categories, One-hot encoding creates a binary vector of length n. In this vector, only the position corresponding to the corresponding category is 1, and the rest of the positions are 0. For example, in one-hot encoding, a vector of length 4 is used to represent these 4 intervals, because the target timing score of target company A belongs to the "excellent" interval, and the "excellent" interval corresponds to the first position of the vector in the setting (from left to right, index 0), so the encoding is [1,0,0,0].

[0052] Specifically, multiple target feature vectors corresponding to multiple target companies can be shown in the target feature vector table below: Target feature vector table

[0053] Step S340, performing high-dimensional space clustering on the multiple target feature vectors according to a density clustering algorithm to obtain a target clustering result, wherein the target clustering result is used to indicate a high-score company group and a low-score company group.

[0054] Among them, the density clustering algorithm usually uses the DBSCAN algorithm. The full name of the DBSCAN algorithm is: Density Based Spatial Clustering of Applications with Noise, which means a density-based clustering method with noise. It defines clusters by finding high-density areas separated by low-density areas, and regards these high-density areas as independent categories. The algorithm is determined by two parameters: radius (r) and point number threshold (MinPts). If the number of sample points in the neighborhood of a given object r is greater than or equal to MinPts, the object is called a core object; when the sample point q is in the neighborhood of the core object p, the sample point q is directly density-reachable from the core object p; if there is a string of sample points, the latter point is directly density-reachable relative to the previous point, then the first and last points are density-reachable; if there is a point o, point o is density-reachable to both the core object p and the sample point q, then the core object p and the sample point q are density-connected. All points that are density-reachable or density-connected to each other will eventually be merged into the same cluster because they form a connected high-density area through the transitivity of the core object.

[0055] For example, based on the target feature vectors corresponding to target company A, target company B, and target company C, for any two target feature vectors, their Euclidean distance is calculated, and the Euclidean distance between target company A and target company B is 2, the Euclidean distance between target company A and target company C is 0.4, and the Euclidean distance between target company B and target company C is 2. When the preset number of points is set to 2, if a certain point is centered and contains at least 2 points (including the point itself) within the distance threshold of 0.5, then this point is the core point. Since the distance between target company A and target company C is less than the threshold of 0.5, their mutual density is reachable, and they both meet the condition of the preset number of points being 2 in each other's density neighborhood, so they belong to density-connected points and can be clustered into one category. The distance between target company B and target company A and target company C exceeds the threshold of 0.5. Within the distance threshold of 0.5 centered on target company B, no other points that meet the preset number of points of 2 can be found, so target company B cannot be clustered into one category with target company A and target company C, and becomes a separate category. Finally, target company A and target company C are grouped into one category, and target company B is grouped into a separate category.

[0056] It can be understood that when the spatial position of the aggregated group is higher, it can be regarded as that the evaluation score of the aggregated company is higher, that is, the group is a high-quality (high-scoring) company group, and the effective characteristics of the group are positive (high-scoring) characteristics for the healthy operation of the company. Conversely, negative (low-scoring) characteristics that lead to poor operation of the company can also be picked out.

[0057] Step S350, determining a plurality of high-scoring business characteristics and a plurality of low-scoring business characteristics according to the plurality of basic scores, the high-scoring company group, and the low-scoring company group.

[0058] Specifically, Figure 4 As shown, Figure 4 Provide a step flow chart of a business characteristic evaluation method based on high-dimensional space vectors. Figure 4A flow chart of steps for determining multiple high-scoring business characteristics and multiple low-scoring business characteristics is shown, including: step S410, determining multiple first reference scores of multiple business characteristics corresponding to each target high-scoring company among multiple target high-scoring companies based on multiple basic scores, and determining multiple second reference scores of multiple business characteristics corresponding to each target low-scoring company among multiple target low-scoring companies; step S420, determining multiple first variances corresponding to multiple business characteristics based on multiple first reference scores, and determining multiple second variances corresponding to multiple business characteristics based on multiple second reference scores; step S430, determining multiple first variances that meet a first preset condition, and determining multiple second variances that meet a second preset condition; step S440, determining multiple business characteristics corresponding to multiple first variances that meet the first preset condition as multiple high-scoring business characteristics, and determining multiple business characteristics corresponding to multiple second variances that meet the second preset condition as multiple low-scoring business characteristics.

[0059] In a possible embodiment, the high-score company group includes a plurality of target high-score companies, and the low-score company group includes a plurality of target low-score companies; and determining a plurality of high-score business characteristics and a plurality of low-score business characteristics based on the plurality of basic scores, the high-score company group, and the low-score company group includes: Determining a plurality of first reference scores of the plurality of business characteristics corresponding to each of the plurality of target high-scoring companies according to the plurality of basic scores; and determining a plurality of second reference scores of the plurality of business characteristics corresponding to each of the plurality of target low-scoring companies according to the plurality of basic scores; Determine a plurality of first variances corresponding to the plurality of business characteristics according to a plurality of first reference scores of the plurality of business characteristics corresponding to each target high-scoring company; and determine a plurality of second variances corresponding to the plurality of business characteristics according to a plurality of second reference scores of the plurality of business characteristics corresponding to each target low-scoring company; The plurality of high-scoring business characteristics and the plurality of low-scoring business characteristics are determined according to the plurality of first variances, the plurality of second variances, the first preset condition, and the second preset condition.

[0060] It can be understood that the multiple basic scores of the multiple operating characteristics corresponding to the target high-score company are the multiple first reference scores corresponding to the company, and the multiple basic scores of the multiple operating characteristics corresponding to the target low-score company are the multiple second reference scores corresponding to the company.

[0061] In a possible embodiment, before determining the multiple first variances corresponding to the multiple business characteristics according to the multiple first reference scores of the multiple business characteristics corresponding to each target high-scoring company, the method further includes: Determine a plurality of first reference null values ​​corresponding to the plurality of business characteristics corresponding to each target high-score company; and determine a plurality of second reference null values ​​corresponding to the plurality of business characteristics corresponding to each target low-score company; Determine at least one first reference null value and / or at least one second reference null value among the plurality of first reference null values ​​and the plurality of second reference null values ​​that is greater than a preset null value threshold; At least one operating characteristic corresponding to the at least one first reference null value is eliminated from the multiple operating characteristics corresponding to each target high-score company; and at least one operating characteristic corresponding to the at least one second reference null value is eliminated from the multiple operating characteristics corresponding to each target low-score company.

[0062] In a possible embodiment, determining the plurality of high-scoring business characteristics and the plurality of low-scoring business characteristics according to the plurality of first variances, the plurality of second variances, the first preset condition, and the second preset condition includes: Determine a plurality of first variances that meet the first preset condition, wherein the first preset condition is that the total number of the plurality of variances selected is less than a first preset number and the value of each variance is less than the first preset value in the order of selection from small to large variance values; and Determine a plurality of second variances that meet the second preset condition, wherein the second preset condition is that the total number of the plurality of variances selected is less than the second preset number and the value of each variance is greater than the second preset value in the order of selection from large to small variance values; The multiple high-scoring business characteristics are determined based on the multiple business characteristics corresponding to the multiple first variances that meet the first preset condition; and the multiple low-scoring business characteristics are determined based on the multiple business characteristics corresponding to the multiple second variances that meet the second preset condition.

[0063] For example, target company A and target company C are target high-scoring companies, and target company B is a target low-scoring company. The first reference score of target company A for the "financial capability" feature is 85 points, and the first reference score of target company C for the "financial capability" feature is 90 points; the first reference score of target company A for the "capital aspect" feature is 88 points, and the first reference score of target company C for the "capital aspect" feature is 92 points; then the variance of the "financial capability" feature is calculated according to the variance formula = =6.25. By analogy, we can obtain multiple first variances corresponding to high-scoring companies and multiple second variances corresponding to low-scoring companies.

[0064] It can be understood that the high-score company group corresponds to multiple operating characteristics, and a certain number of operating characteristics with smaller variance values ​​are selected as high-score operating characteristics; and the low-score company group corresponds to multiple operating characteristics, and a certain number of operating characteristics with larger variance values ​​are selected as low-score operating characteristics. This can effectively analyze the main characteristics of companies with different operating conditions and improve the accuracy and comprehensiveness of the operating characteristics evaluation.

[0065] In a possible embodiment, after determining a plurality of high-scoring business characteristics and a plurality of low-scoring business characteristics according to the plurality of basic scores, the high-scoring company group, and the low-scoring company group, the method further includes: Acquire a plurality of first target data of the target company to be evaluated under the dimensions of the plurality of high-scoring business characteristics, and acquire a plurality of second target data of the target company to be evaluated under the dimensions of the plurality of low-scoring business characteristics; Performing data preprocessing on the plurality of first target data and the plurality of second target data, wherein the data preprocessing includes missing value processing and outlier processing; A first trend graph is established based on the multiple first target data after the data preprocessing and the financial cycle of the target company to be evaluated, and the first trend graph is used to characterize the changes of the multiple high-scoring operating characteristics of the target company to be evaluated over time; and a second trend graph is established based on the multiple second target data after the data preprocessing and the financial cycle, and the second trend graph is used to characterize the changes of the multiple low-scoring operating characteristics of the target company to be evaluated over time.

[0066] It can be seen that in the embodiment of the present application, by mapping multiple operating characteristics of listed companies to high-dimensional space vectors, using machine learning methods to perform spatial density clustering, listed companies with similar feature vectors will form natural aggregation groups, and perform principal component analysis on the characteristics under the same aggregation group to obtain multiple main operating characteristics of high-scoring and low-scoring company groups, respectively. This improves the accuracy and comprehensiveness of operating characteristic evaluation.

[0067] See also Figure 5 , Figure 5 is an overall flow chart of another business characteristic evaluation method based on high-dimensional space vectors provided in an embodiment of the present application, which is applied to Figure 1 The server 120 in Figure 5 As shown, this method selected target company A, target company B and target company C for this business characteristics evaluation event.

[0068] The server evaluates each of target company A, target company B, and target company C from two dimensions: timing indicators and value indicators. First, it needs to collect multiple operating characteristic data of each company under these two dimensions, and determine multiple scores of multiple operating characteristics corresponding to each company. Among them, the timing indicators include five indicators: funding, short-term indicators, price indicators, stock price forecasts, and company size. The value indicators include five indicators: financial capabilities, valuation indicators, institutional ratings, historical dividends, and risk control.

[0069] Furthermore, the server generates corresponding target feature vectors based on the performance of each company in terms of timing indicators and value indicators. For example, the target feature vector of target company A is [1,0,0,0,0,0,1,0], the target feature vector of target company B is [0,0,0,1,0,0,0,1], and the target feature vector of target company C is [1,0,0,0,1,0,0,0]. This process is based on interval mapping and one-hot encoding of indicator scores, converting the company's operating feature information into a computer-processable vector form, providing a data basis for subsequent clustering analysis.

[0070] Furthermore, the server clusters the target feature vectors of each company in high-dimensional space density, and clusters the target companies according to the distance between feature vectors, preset distance thresholds, preset points, and other conditions based on density-based clustering algorithms such as DBSCAN. Clustering can be used to divide company groups with different operating characteristics, such as high-branch company groups and low-branch company groups, which helps to analyze the common characteristics of different company groups, screen positive and negative guiding features, and provide decision support for regulators or investors.

[0071] It can be seen that compared with the existing operating characteristics evaluation schemes with one-sided data analysis and low accuracy, this application maps multiple operating characteristics of listed companies into high-dimensional space vectors, uses machine learning methods to perform spatial density clustering, and then obtains the main operating characteristics of high-quality companies and low-quality companies based on the clustering results, thereby improving the accuracy and comprehensiveness of the operating characteristics evaluation.

[0072] See also Figure 6 , Figure 6 is a schematic diagram of an interface of a business characteristics evaluation system provided in an embodiment of the present application, such as Figure 6 As shown, this interface diagram is a visualization interface of the comprehensive evaluation results in the business characteristics evaluation system.

[0073] The display interface of the business characteristics evaluation system mainly includes three parts. The top part is the title part, and the middle main part is divided into the left navigation bar part and the right display part. Among them, the right side of the title part has the text and icons corresponding to the home page, history, report generation and user management arranged from left to right. Among them, the left navigation bar part has multiple navigation bars arranged from top to bottom, including management of objects to be evaluated, management of characteristic indicators, evaluation results and analysis, evaluation suggestions, notifications and announcements, and user feedback and evaluation.

[0074] Specifically, the display part on the right is the display interface corresponding to the comprehensive evaluation result navigation bar under the evaluation result and analysis navigation bar. Among them, the upper part of the display part on the right presents the scores, target timing scores and target value scores of multiple operating characteristics corresponding to target company A, target company B and target company C in the form of a table. Among them, the lower part of the display part on the right represents the results corresponding to the statistics and analysis of target companies and the statistics and analysis of high / low-scoring operating characteristics in the form of pie charts and bar charts, respectively. The pie chart is specifically used to present the proportion of high-scoring companies and low-scoring companies, and the bar chart is specifically used to present the variance corresponding to multiple high-scoring / low-scoring operating characteristics. The high-scoring / low-scoring operating characteristics statistics are switched by the switch button in the upper right corner.

[0075] It can be seen that in the embodiment of the present application, the operating characteristic evaluation results of multiple target companies are visualized through the operating characteristic evaluation system. Users can intuitively and clearly see the scoring of the operating characteristics of multiple companies in multiple different dimensions, and can clearly identify the high-scoring operating characteristics corresponding to high-scoring companies and the low-scoring operating characteristics corresponding to low-scoring companies. This is conducive to analyzing the company's operating conditions, providing decision-making for the company's operations, and improving the accuracy and comprehensiveness of the operating characteristic evaluation.

[0076] See also Figure 7 , Figure 7 is a schematic diagram of an interface of another business characteristic evaluation system provided in an embodiment of the present application, such as Figure 7 As shown, the interface diagram is a visualization interface for business characteristic trend analysis in the business characteristic evaluation system.

[0077] The middle area on the right side of the asset valuation interface is the output area for the analysis results of the operating characteristics and trends. The upper part of the middle area includes the prompt text "Please confirm the object to be analyzed" and the corresponding icon, which is used to determine at least one target company from multiple target companies for characteristic trend analysis. Figure 7 Target Company A and Target Company C were selected.

[0078] Specifically, the lower half of the middle area includes two line charts on the left and right, which are used to respectively present the changes of the three high-scoring operating characteristics corresponding to target company A with the company's financial cycle and the changes of the three low-scoring operating characteristics corresponding to target company C with the company's financial cycle. The analysis chart of target company A involves the trends of the three characteristics of value indicators, stock price forecasts, and institutional ratings; the analysis chart of target company C involves the trends of the three characteristics of company size, risk control, and financial capabilities. Among them, each line chart includes three lines corresponding to the three operating characteristics, the horizontal axis of the statistical chart includes the financial cycle Q1-Q4, the vertical axis is the characteristic score, and different lines are displayed in different colors.

[0079] It can be seen that in the embodiment of the present application, the line graph intuitively displays the score change trend of different operating characteristics of each company along with the financial cycle, making the abstract data intuitive and easy to understand. Users can quickly understand whether the company's performance in each characteristic dimension is rising, falling or stable, and timely discover potential problems or advantages. At the same time, it is convenient for users to compare the development trends of different companies in the same or different characteristic dimensions, which helps to discover the differences and commonalities between companies and provide references for investment decisions, competition analysis, etc.

[0080] See also Figure 8 , Figure 8 A functional module diagram of a business characteristic evaluation device 800 based on a high-dimensional space vector provided in an embodiment of the present application is shown in FIG. Figure 8 As shown, the business characteristic evaluation device 800 based on high-dimensional space vectors includes the following units: An acquisition unit 810 is used to acquire multiple business characteristic data sets corresponding to multiple target companies, where a single business characteristic data set includes multiple business characteristics in multiple dimensions; The processing unit 820 is used to perform the following processing on each of the multiple business feature data sets to obtain multiple target scores corresponding to the multiple target companies: scoring the multiple business features in the currently processed business feature data set according to a preset reference scoring standard to obtain multiple basic scores; and, performing reference processing on the multiple basic scores to obtain multiple target scores corresponding to the multiple target companies, the reference processing including weighted summation and normalization; and, mapping the multiple target scores in a preset scoring interval table to obtain multiple target mapping intervals; and, performing one-hot encoding according to the multiple target mapping intervals to obtain multiple target feature vectors corresponding to the multiple target companies; and, performing high-dimensional space clustering on the multiple target feature vectors according to a density clustering algorithm to obtain a target clustering result, the target clustering result is used to indicate a high-score company group and a low-score company group; The determination unit 830 is used to determine a plurality of high-score business characteristics and a plurality of low-score business characteristics according to the plurality of basic scores, the high-score company group, and the low-score company group.

[0081] It can be seen that in the present application, the server obtains multiple business feature data sets corresponding to multiple target companies, and a single business feature data set includes multiple business features under multiple dimensions; the following processing is performed on each business feature data set in the multiple business feature data sets to obtain multiple target scores corresponding to the multiple target companies: the multiple business features in the currently processed business feature data set are scored according to a preset reference scoring standard to obtain multiple basic scores; and, the multiple basic scores are reference processed to obtain multiple target scores corresponding to the multiple target companies, and the reference processing includes weighted summation and normalization; the multiple target scores are mapped in a preset scoring interval table to obtain multiple target mapping intervals; and, one-hot encoding is performed according to the multiple target mapping intervals to obtain multiple target feature vectors corresponding to the multiple target companies; the multiple target feature vectors are clustered in high-dimensional space according to a density clustering algorithm to obtain a target clustering result, and the target clustering result is used to indicate a high-score company group and a low-score company group; multiple high-score business features and multiple low-score business features are determined according to multiple basic scores, high-score company groups and low-score company groups. In this way, compared with the existing operating characteristics evaluation schemes with one-sided data analysis and low accuracy, this application maps multiple operating characteristics of listed companies into high-dimensional space vectors, uses machine learning methods to perform spatial density clustering, and then obtains the main operating characteristics of high-quality and low-quality companies based on the clustering results, thereby improving the accuracy and comprehensiveness of the operating characteristics evaluation.

[0082] In one embodiment, the high-score company group includes multiple target high-score companies, and the low-score company group includes multiple target low-score companies; the determining of multiple high-score business characteristics and multiple low-score business characteristics based on the multiple basic scores, the high-score company group and the low-score company group includes: determining multiple first reference scores of the multiple business characteristics corresponding to each of the multiple target high-score companies based on the multiple basic scores; and, determining multiple second reference scores of the multiple business characteristics corresponding to each of the multiple target low-score companies based on the multiple basic scores; determining multiple first variances corresponding to the multiple business characteristics based on the multiple first reference scores of the multiple business characteristics corresponding to each target high-score company; and, determining multiple second variances corresponding to the multiple business characteristics based on the multiple second reference scores of the multiple business characteristics corresponding to each target low-score company; determining the multiple high-score business characteristics and multiple low-score business characteristics based on the multiple first variances, the multiple second variances, the first preset condition and the second preset condition.

[0083] In one embodiment, the determining of the multiple high-scoring business characteristics and the multiple low-scoring business characteristics based on the multiple first variances, the multiple second variances, the first preset condition and the second preset condition includes: determining the multiple first variances that meet the first preset condition, the first preset condition being that the total number of the multiple variances selected is less than the first preset number and the value of each variance is less than the first preset value in the order of selection from small to large variance values; and, determining the multiple second variances that meet the second preset condition, the second preset condition being that the total number of the multiple variances selected is less than the second preset number and the value of each variance is greater than the second preset value in the order of selection from large to small variance values; determining the multiple high-scoring business characteristics based on the multiple business characteristics corresponding to the multiple first variances that meet the first preset condition; and, determining the multiple low-scoring business characteristics based on the multiple business characteristics corresponding to the multiple second variances that meet the second preset condition.

[0084] In one embodiment, after determining a plurality of high-scoring business characteristics and a plurality of low-scoring business characteristics based on the plurality of basic scores, the high-scoring company group, and the low-scoring company group, the method further includes: obtaining a plurality of first target data of the target company to be evaluated under the dimensions of the plurality of high-scoring business characteristics, and obtaining a plurality of second target data of the target company to be evaluated under the dimensions of the plurality of low-scoring business characteristics; performing data preprocessing on the plurality of first target data and the plurality of second target data, the data preprocessing including missing value processing and outlier processing; establishing a first trend graph based on the plurality of first target data after the data preprocessing and the financial cycle of the target company to be evaluated, the first trend graph being used to characterize how the plurality of high-scoring business characteristics of the target company to be evaluated change over time; and establishing a second trend graph based on the plurality of second target data after the data preprocessing and the financial cycle, the second trend graph being used to characterize how the plurality of low-scoring business characteristics of the target company to be evaluated change over time.

[0085] In one embodiment, the multiple operating characteristics under the multiple dimensions include multiple first operating characteristics under the timing indicator dimension and multiple second operating characteristics under the value indicator dimension, the timing indicator is used to evaluate the operating status of the target company from a technical dimension, and the value indicator is used to evaluate the operating status of the target company from a financial dimension; the scoring of the multiple operating characteristics in the currently processed operating characteristic data set according to the preset reference scoring standard to obtain multiple basic scores includes: scoring the multiple first operating characteristics in the currently processed operating characteristic data set according to the preset reference scoring standard to obtain multiple first basic scores; and, scoring the multiple second operating characteristics in the currently processed operating characteristic data set according to the preset reference scoring standard to obtain multiple second basic scores.

[0086] In one embodiment, the target score includes a target timing score and a target value score; the reference processing of the multiple basic scores to obtain multiple target scores corresponding to the multiple target companies includes: weighted summing up the multiple first basic scores and the multiple second basic scores corresponding to the currently processed operating characteristic data set to obtain a reference timing score and a reference value score of the target company corresponding to the currently processed operating characteristic data set; normalizing the reference timing score of each of the multiple target companies according to the multiple reference timing scores corresponding to the multiple target companies to obtain the multiple target timing scores corresponding to the multiple target companies; and normalizing the reference value score of each of the multiple target companies according to the multiple reference value scores corresponding to the multiple target companies to obtain the multiple target value scores corresponding to the multiple target companies.

[0087] In one embodiment, the target mapping interval includes a target timing interval corresponding to the target timing score and a target value interval corresponding to the target value score; performing one-hot encoding according to the multiple target mapping intervals to obtain multiple target feature vectors corresponding to the multiple target companies includes: determining the target timing interval and the target value interval corresponding to each target company among the multiple target companies; for the target timing interval and the target value interval corresponding to each target company, to obtain the multiple target feature vectors: performing one-hot encoding according to the target timing interval and the scoring interval table corresponding to the target company to obtain a first sub-feature vector; and performing one-hot encoding according to the target value interval and the scoring interval table corresponding to the target company to obtain a second sub-feature vector; obtaining the target feature vector corresponding to the target company according to the first sub-feature vector and the second sub-feature vector.

[0088] It can be understood that since the method embodiment and the device embodiment are different presentation forms of the same technical concept, the content of the method embodiment part in this application should be synchronously adapted to the device embodiment part and will not be repeated here.

[0089] In addition, an embodiment of the present application also provides a computer storage medium, which stores a computer program that can be loaded by a processor and executed such as the above-mentioned method for evaluating business characteristics based on high-dimensional space vectors. The computer-readable storage medium includes, for example: a U disk, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, and other media that can store program codes.

[0090] It should be noted that, for the aforementioned method embodiments, for the sake of simplicity, they are all expressed as a series of action combinations, but those skilled in the art should be aware that the present application is not limited by the described order of actions, because according to the present application, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required by the present application.

[0091] In the several embodiments provided in the present application, it should be understood that the disclosed methods, devices and systems can be implemented in other ways. For example, the device embodiments described above are only schematic; for example, the division of the units is only a logical function division, and there may be other division methods in actual implementation; for example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms. The units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the scheme of this embodiment. In addition, the functional units in various embodiments of the present invention can be integrated into a processing unit, or each unit can be physically included separately, or two or more units can be integrated into one unit. The above-mentioned integrated units can be implemented in the form of hardware or in the form of hardware plus software functional units.

[0092] The above-mentioned integrated unit implemented in the form of a software functional unit can be stored in a computer-readable storage medium. The above-mentioned software functional unit is stored in a storage medium, including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform some steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: a USB flash drive, a mobile hard disk, a magnetic disk, an optical disk, a volatile memory or a non-volatile memory. Among them, the non-volatile memory can be a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM) or a flash memory. The volatile memory can be a random access memory (RAM), which is used as an external cache. By way of example but not limitation, many forms of random access memory (RAM) are available, such as static random access memory (SRAM), dynamic random access memory (DRAM), synchronous dynamic random access memory (SDRAM), double data rate synchronous dynamic random access memory (DDRSDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous link dynamic random access memory (SLDRAM) and direct memory bus random access memory (DRRAM), etc., various media that can store program code.

[0093] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0094] The embodiments of the present application are introduced in detail above. Specific examples are used in this article to illustrate the principles and implementation methods of the present application. The description of the above embodiments is only used to help understand the method and core idea of ​​the present application. At the same time, for general technical personnel in this field, according to the idea of ​​the present application, there will be changes in the specific implementation method and application scope. In summary, the content of this specification should not be understood as a limitation on the present application.

Claims

1. A method for evaluating business characteristics based on high-dimensional space vectors, characterized in that: The method applied to a server of a business characteristic evaluation system comprises: Acquire multiple operating characteristic data sets corresponding to multiple target companies, where a single operating characteristic data set includes multiple operating characteristics in multiple dimensions; The following processing is performed on each of the plurality of operating characteristic data sets to obtain a plurality of target scores corresponding to the plurality of target companies: scoring the plurality of operating characteristics in the currently processed operating characteristic data set according to a preset reference scoring standard to obtain a plurality of basic scores; and performing reference processing on the plurality of basic scores to obtain a plurality of target scores corresponding to the plurality of target companies, wherein the reference processing includes weighted summation and normalization; Mapping the multiple target scores in a preset score interval table to obtain multiple target mapping intervals; and performing one-hot encoding according to the multiple target mapping intervals to obtain multiple target feature vectors corresponding to the multiple target companies; Performing high-dimensional spatial clustering on the multiple target feature vectors according to a density clustering algorithm to obtain a target clustering result, wherein the target clustering result is used to indicate a high-score company group and a low-score company group; A plurality of high-scoring business characteristics and a plurality of low-scoring business characteristics are determined based on the plurality of basic scores, the high-scoring company group, and the low-scoring company group.

2. The method according to claim 1, characterized in that The high-score company group includes a plurality of target high-score companies, and the low-score company group includes a plurality of target low-score companies; and determining a plurality of high-score business characteristics and a plurality of low-score business characteristics according to the plurality of basic scores, the high-score company group, and the low-score company group includes: Determine a plurality of first reference scores of the plurality of business characteristics corresponding to each of the plurality of target high-scoring companies according to the plurality of basic scores; and determine a plurality of second reference scores of the plurality of business characteristics corresponding to each of the plurality of target low-scoring companies according to the plurality of basic scores; Determine a plurality of first variances corresponding to the plurality of business characteristics according to a plurality of first reference scores of the plurality of business characteristics corresponding to each target high-scoring company; and determine a plurality of second variances corresponding to the plurality of business characteristics according to a plurality of second reference scores of the plurality of business characteristics corresponding to each target low-scoring company; The plurality of high-scoring business characteristics and the plurality of low-scoring business characteristics are determined according to the plurality of first variances, the plurality of second variances, the first preset condition, and the second preset condition.

3. The method according to claim 2, characterized in that The determining of the plurality of high-scoring business characteristics and the plurality of low-scoring business characteristics according to the plurality of first variances, the plurality of second variances, the first preset condition and the second preset condition comprises: Determine a plurality of first variances that meet the first preset condition, wherein the first preset condition is that the total number of the plurality of variances selected is less than a first preset number and the value of each variance is less than the first preset value in the order of selection from small to large variance values; and Determine a plurality of second variances that meet the second preset condition, wherein the second preset condition is that the total number of the plurality of variances selected is less than the second preset number and the value of each variance is greater than the second preset value in the order of selection from large to small variance values; The multiple high-scoring business characteristics are determined based on the multiple business characteristics corresponding to the multiple first variances that meet the first preset condition; and the multiple low-scoring business characteristics are determined based on the multiple business characteristics corresponding to the multiple second variances that meet the second preset condition.

4. The method according to any one of claims 1 to 3, characterized in that: After determining a plurality of high-scoring business characteristics and a plurality of low-scoring business characteristics according to the plurality of basic scores, the high-scoring company group, and the low-scoring company group, the method further includes: Acquire a plurality of first target data of the target company to be evaluated under the dimensions of the plurality of high-scoring business characteristics, and acquire a plurality of second target data of the target company to be evaluated under the dimensions of the plurality of low-scoring business characteristics; Performing data preprocessing on the plurality of first target data and the plurality of second target data, wherein the data preprocessing includes missing value processing and outlier processing; A first trend graph is established based on the multiple first target data after the data preprocessing and the financial cycle of the target company to be evaluated, and the first trend graph is used to characterize the changes of the multiple high-scoring operating characteristics of the target company to be evaluated over time; and a second trend graph is established based on the multiple second target data after the data preprocessing and the financial cycle, and the second trend graph is used to characterize the changes of the multiple low-scoring operating characteristics of the target company to be evaluated over time.

5. The method according to claim 4, characterized in that The multiple operating characteristics under the multiple dimensions include multiple first operating characteristics under the timing indicator dimension and multiple second operating characteristics under the value indicator dimension, the timing indicator is used to evaluate the operating status of the target company from a technical dimension, and the value indicator is used to evaluate the operating status of the target company from a financial dimension; The multiple business characteristics in the currently processed business characteristic data set are scored according to the preset reference scoring standard to obtain multiple basic scores, including: Scoring a plurality of first operating characteristics in the currently processed operating characteristic data set according to a preset reference scoring standard to obtain a plurality of first basic scores; as well as, The plurality of second business characteristics in the currently processed business characteristic data set are scored according to a preset reference scoring standard to obtain a plurality of second basic scores.

6. The method according to claim 5, characterized in that The target score includes a target timing score and a target value score; the reference processing of the multiple basic scores to obtain multiple target scores corresponding to the multiple target companies includes: The plurality of first basic scores and the plurality of second basic scores corresponding to the currently processed business characteristic data set are weighted and summed respectively to obtain a reference timing score and a reference value score of the target company corresponding to the currently processed business characteristic data set; Normalizing the reference timing score of each target company in the multiple target companies according to the multiple reference timing scores corresponding to the multiple target companies, to obtain the multiple target timing scores corresponding to the multiple target companies; and The reference value score of each target company in the multiple target companies is normalized according to the multiple reference value scores corresponding to the multiple target companies to obtain the multiple target value scores corresponding to the multiple target companies.

7. The method according to claim 6, characterized in that The target mapping interval includes the target timing interval corresponding to the target timing score and the target value interval corresponding to the target value score; The one-hot encoding is performed according to the multiple target mapping intervals to obtain multiple target feature vectors corresponding to the multiple target companies, including: Determine the target timing interval and the target value interval corresponding to each target company among the multiple target companies; The target timing interval and the target value interval corresponding to each target company are used to obtain the multiple target feature vectors: One-hot encoding is performed according to the target timing interval and the scoring interval table corresponding to the target company to obtain a first sub-feature vector; and one-hot encoding is performed according to the target value interval and the scoring interval table corresponding to the target company to obtain a second sub-feature vector; The target feature vector corresponding to the target company is obtained according to the first sub-feature vector and the second sub-feature vector.

8. A business characteristic evaluation device based on high-dimensional space vectors, characterized in that: A server applied to a business characteristics evaluation system, the device comprising: An acquisition unit, used to acquire multiple operating characteristic data sets corresponding to multiple target companies, wherein a single operating characteristic data set includes multiple operating characteristics under multiple dimensions; A processing unit, for performing the following processing on each of the multiple business feature data sets to obtain multiple target scores corresponding to the multiple target companies: scoring the multiple business features in the currently processed business feature data set according to a preset reference scoring standard to obtain multiple basic scores; and, performing reference processing on the multiple basic scores to obtain multiple target scores corresponding to the multiple target companies, the reference processing including weighted summation and normalization; and, for mapping the multiple target scores in a preset scoring interval table to obtain multiple target mapping intervals; and, performing one-hot encoding according to the multiple target mapping intervals to obtain multiple target feature vectors corresponding to the multiple target companies; and, for performing high-dimensional space clustering on the multiple target feature vectors according to a density clustering algorithm to obtain a target clustering result, the target clustering result being used to indicate a high-score company group and a low-score company group; A determination unit is used to determine a plurality of high-score business characteristics and a plurality of low-score business characteristics according to the plurality of basic scores, the high-score company group, and the low-score company group.

9. An electronic device, characterized in that: The method comprises a processor, a memory, a communication interface and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the processor, and the programs include instructions for executing the steps in the method according to any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program / instruction stored thereon, characterized in that: The computer program / instructions are executed by a processor to implement the steps of the method according to any one of claims 1 to 7.

Images