CN117557106A

CN117557106A - Enterprise operation health degree detection method, device, equipment and medium

Info

Publication number: CN117557106A
Application number: CN202410001776.9A
Authority: CN
Inventors: 陈声荣; 陈迪; 韩中元; 文享龙; 陈浩阳
Original assignee: Guangdong Haodi Zhiyun Technology Co ltd
Current assignee: Guangdong Haodi Zhiyun Technology Co ltd
Priority date: 2024-01-02
Filing date: 2024-01-02
Publication date: 2024-02-13

Abstract

The application relates to an enterprise operation health degree detection method, device, equipment and medium, wherein the method comprises the following steps: determining the business operation characteristic data of each enterprise in different time period ranges; according to the business operation characteristic data of each enterprise, adopting a logic function, a ReLU function and/or a maximum function to construct an enterprise operation health algorithm corresponding to the enterprise electricity consumption, the enterprise working time length and the newly-increased risk quantity of the enterprise; determining enterprise health corresponding to the enterprise power consumption, the enterprise working time length and the newly-increased risk number of the enterprise in different time periods based on an enterprise operation health algorithm corresponding to the enterprise power consumption, the enterprise working time length and the newly-increased risk number of the enterprise; and determining the comprehensive health degree corresponding to the enterprise according to the enterprise power consumption, the enterprise working time and the enterprise health degree corresponding to the newly-increased risk number of the enterprise in different time periods so as to finish the detection of the enterprise health degree. The method and the device can accurately and efficiently detect the business operation health degree corresponding to each business.

Description

Enterprise operation health degree detection method, device, equipment and medium

Technical Field

The present disclosure relates to the field of enterprise health detection, and in particular, to an enterprise business health detection method, a corresponding apparatus, an electronic device, and a computer readable storage medium.

Background

In the production and management process of small and medium enterprises, there is a general need to seek loans or financing from financial institutions. However, the capability of resisting market risks is weak in the development and management process of small and medium enterprises, the financial condition is opaque, the production and management state is not disclosed, the financial institution cannot quickly and accurately grasp the production and management state of the enterprises, the information grasped by the two parties is asymmetric, and the financial institution has a prominent pain point of not daring to credit or unwilling to credit for small and medium enterprises in the current stage.

At present, the financial institution obtains the channel of enterprise's operation information and the information existence information that obtains are objective, inaccurate, untimely, unable quantitative analysis, work efficiency low grade problem for the financial institution is comparatively difficult carries out timely, accurate, lasting authoritative aassessment to the production operation state of target loan enterprise, has increased financial institution's assessment of lending, the human cost and the risk cost of management.

In conclusion, the method is suitable for the problems that a channel for acquiring enterprise operation information by a financial institution in the prior art and information existence information acquired by the channel are not objective, inaccurate and untimely, cannot be quantitatively analyzed, have low working efficiency, and the financial institution difficultly carries out timely, accurate and continuous authoritative assessment on the production operation state of a target loan enterprise.

Disclosure of Invention

An object of the present application is to solve the above-mentioned problems and provide an enterprise operation health degree detection method, a corresponding apparatus, an electronic device and a computer readable storage medium.

In order to meet the purposes of the application, the application adopts the following technical scheme:

one of the purposes of the present application is to provide an enterprise operation health detection method, which includes:

determining enterprise operation characteristic data in different time period ranges, wherein the enterprise operation characteristic data comprises one or more of enterprise electricity consumption, enterprise working time length and enterprise newly-increased risk quantity;

according to the enterprise electricity consumption, the enterprise working time and the newly-increased risk quantity of the enterprise, adopting a logic function, a ReLU function and/or a maximum function to construct an enterprise operation health algorithm corresponding to the enterprise electricity consumption, the enterprise working time and the newly-increased risk quantity of each enterprise in the enterprise operation characteristic data;

determining enterprise health corresponding to the enterprise power consumption, the enterprise working time length and the newly-increased risk number of the enterprise in different time periods based on the enterprise operation health algorithm corresponding to the enterprise power consumption, the enterprise working time length and the newly-increased risk number of the enterprise;

And determining the corresponding comprehensive health degree of the enterprise according to the enterprise power consumption, the enterprise working time and the enterprise health degree corresponding to the newly-increased risk number of the enterprise in different time periods so as to finish the detection of the enterprise health degree.

Optionally, the step of constructing an enterprise operation health algorithm corresponding to the enterprise power consumption, the enterprise working time length and the enterprise newly-increased risk number in the enterprise operation feature data by adopting a logic function, a ReLU function and/or a maximum function according to the enterprise power consumption, the enterprise working time length and the enterprise newly-increased risk number comprises the following steps:

the enterprise operation health algorithm corresponding to the enterprise electricity consumption comprises one or more of the following formulas: logit (a+ (M-N)/N.times.k), wherein the Logit () isA and k are real numbers, M is the average value of the power consumption of M workdays, N is the average value of N workdays, and M is a positive integer smaller than N;

or max ((b-relu (-H)), c-relu (H-F)), wherein the relu () isB and c are positive real numbers, and H and F are the ratio of the loop ratio to the electricity consumption of a certain time period;

or logic (d+ (Y-R)/R j) g-R, wherein d, j, Y, R is a positive real number, Y is the current daily electricity consumption value or the electricity consumption average value of a certain time period, and R is the electricity consumption average value of a certain time period different from Y;

Or q-relu (-L), wherein q is a positive real number, and L is the same ratio of the electricity consumption of a certain time period;

or P/Q, wherein P is the daily electricity consumption of a certain time period, and Q is the daily electricity consumption of a certain time period different from P.

and the enterprise operation health degree algorithm corresponding to the enterprise working time is max (t, U/V)/d, wherein t and d are positive real numbers, U is the working time in the U time period, and V is the working time in the V time period.

the enterprise operation health degree algorithm corresponding to the newly-increased risk quantity of the enterprise is logic (W), wherein W is a certain newly-increased risk quantity of the enterprise in a certain time period, and the newly-increased risk quantity of the enterprise comprises one or more of the newly-increased legal risk quantity of the enterprise, the newly-increased operation risk quantity of the enterprise and the newly-increased supervision risk quantity of the enterprise.

Optionally, the step of determining the comprehensive health degree corresponding to the enterprise according to the enterprise health degree corresponding to the enterprise electricity consumption, the enterprise working time length and the enterprise newly-increased risk number in the different time periods includes:

responding to an enterprise operation health degree detection event, and acquiring enterprise electricity consumption, enterprise working time length and enterprise newly-increased risk quantity in different time period ranges;

invoking an enterprise operation health degree algorithm corresponding to the enterprise electricity consumption, the enterprise working time length and the newly-increased risk quantity of each enterprise operation characteristic data, and determining enterprise operation health degrees corresponding to each enterprise operation characteristic data in different time periods;

and calling a preset linear model according to the enterprise operation health degree corresponding to the enterprise operation characteristic data to determine the enterprise operation comprehensive health degree corresponding to each enterprise so as to finish the detection of the enterprise health degree.

Optionally, after the step of determining the comprehensive health of enterprise operations corresponding to each enterprise, the method includes:

responding to the enterprise operation health grade assessment instruction, and determining the score of the enterprise operation comprehensive health degree corresponding to each enterprise;

and determining the corresponding enterprise operation health grade of each enterprise according to the score of the enterprise operation comprehensive health degree, wherein the enterprise operation health grade comprises one or more of excellent, good, qualified and unqualified.

Optionally, after the step of determining the comprehensive health of enterprise operations corresponding to each enterprise in different time periods, the method includes:

determining enterprise operation health degrees corresponding to the enterprise operation feature data, and determining an input sequence of an enterprise health degree detection model according to the enterprise operation health degrees corresponding to the enterprise operation feature data;

and inputting the input sequence into a pre-trained enterprise health degree detection model, and determining the enterprise operation comprehensive health degree corresponding to each enterprise according to the weight value between every two layers of neurons of the fully-connected neural network in the enterprise health degree detection model and a linear transformation prediction function.

An enterprise operation health degree detection apparatus adapted to another object of the present application includes:

the characteristic data determining module is used for determining enterprise operation characteristic data in different time period ranges, wherein the enterprise operation characteristic data comprises one or more of enterprise power consumption, enterprise working time length and enterprise newly-increased risk quantity;

the health degree algorithm construction module is arranged to construct an enterprise operation health degree algorithm corresponding to the enterprise power consumption, the enterprise working time length and the newly-increased risk quantity of the enterprise in the enterprise operation characteristic data by adopting a logic function, a ReLU function and/or a maximum function according to the enterprise power consumption, the enterprise working time length and the newly-increased risk quantity of the enterprise;

The single health degree determining module is configured to determine the enterprise health degree corresponding to the enterprise power consumption, the enterprise working time length and the enterprise newly-increased risk number in different time periods based on the enterprise operation health degree algorithm corresponding to the enterprise power consumption, the enterprise working time length and the enterprise newly-increased risk number;

the comprehensive health degree determining module is configured to determine the comprehensive health degree corresponding to the enterprise according to the enterprise power consumption, the enterprise working time length and the enterprise health degree corresponding to the newly-increased risk number of the enterprise in different time periods so as to complete detection of the enterprise health degree.

An electronic device adapted for another object of the present application comprises a central processor and a memory, said central processor being adapted to invoke the steps of executing a computer program stored in said memory for performing the method for detecting business health of an enterprise as described herein.

A computer readable storage medium adapted to another object of the present application stores a computer program implemented according to the enterprise business health detection method in the form of computer readable instructions, which when invoked by a computer, performs the steps comprised by the corresponding method.

Compared with the prior art, the method and the device for acquiring the enterprise operation information by the financial institutions are used for solving the problems that the channel for acquiring the enterprise operation information by the financial institutions in the prior art and the information acquired by the channel are not objective, inaccurate and timely, cannot be quantitatively analyzed, have low working efficiency, the financial institutions difficultly evaluate the production operation state of a target loan enterprise in time, accurately and continuously authoritative, and the like, the method and the device for acquiring the enterprise operation information by the financial institutions adopt a plurality of enterprise operation characteristic data closely related to the enterprise operation health degree, divide the enterprise operation characteristic data into sub-characteristic health degree calculation formulas under different periods, calculate the comprehensive operation health degree value of the enterprise by adopting a linear calculation model, and learn and train the characteristic weights of the health degrees of the enterprise operation characteristic data based on an enterprise operation health degree detection model, and the method and the device for acquiring the enterprise operation characteristic health degree comprises the following beneficial effects:

firstly, the method and the device are based on the fact that the enterprise production operation health degree is related to various factors, the influence of each enterprise operation characteristic data on the enterprise operation state health degree is integrated, a whole set of multi-parameter intelligent recognition enterprise operation health degree algorithm is determined according to the importance degree of each enterprise operation characteristic data in different time periods such as long, medium, short and ultra-short periods, so that the enterprise operation health degree corresponding to each enterprise operation characteristic data is obtained, a good data basis is laid for the follow-up accurate and authoritative calculation to judge the enterprise comprehensive health degree, and the enterprise operation health degree can be calculated quantitatively for enterprises of different types, industries and scales with high efficiency and high precision;

Secondly, the method and the device perform learning training on the characteristic weights of the health degrees of the enterprise operation characteristic data based on the enterprise operation health degree detection model, ensure the accuracy and the authority of the enterprise comprehensive operation health degree value, enable different types of enterprises to learn and train to have specific characteristic weight value combinations, enable the enterprises to accurately and quantitatively calculate the comprehensive operation health degree values of different enterprises, and provide authoritative decision basis for financial institutions to judge and analyze the operation state health degree of the enterprise and the operation trend of future enterprises;

thirdly, through fusing various characteristics of enterprise operation big data and based on an enterprise operation health degree detection model, the weight parameters of the model can be continuously self-learned and reinforced trained, the operation state health degree values of different enterprises can be obtained through more personalized, accurate and rapid calculation, the enterprise operation state can be known timely by data application parties such as financial institutions and the like, and continuous and efficient cooperation win-win of the financial institutions and enterprise clients is ensured;

further, the method and the device can accurately and efficiently detect the corresponding business operation health degree of each enterprise, greatly reduce the labor cost and the risk cost of the financial institutions, objectively, accurately and timely acquire the operation condition information of the enterprises, and greatly improve the working efficiency.

Drawings

The foregoing and/or additional aspects and advantages of the present application will become apparent and readily appreciated from the following description of the embodiments, taken in conjunction with the accompanying drawings, in which:

FIG. 1 is a schematic flow chart of an enterprise operation health detection method in an embodiment of the present application;

FIG. 2 is a schematic flow chart of preprocessing enterprise operation characteristic data in an embodiment of the present application;

FIG. 3 is a flowchart illustrating a method for determining an overall health of an enterprise according to an embodiment of the present application;

FIG. 4 is a schematic flow chart of determining the business health level of an enterprise according to an embodiment of the present application;

FIG. 5 is a schematic flow chart of determining comprehensive health of an enterprise based on a fully connected neural network in an embodiment of the present application;

FIG. 6 is a schematic block diagram of an enterprise business health detection apparatus according to an embodiment of the present application;

fig. 7 is a schematic structural diagram of a computer device in an embodiment of the present application.

Detailed Description

Embodiments of the present application are described in detail below, examples of which are illustrated in the accompanying drawings, wherein the same or similar reference numerals refer to the same or similar elements or elements having the same or similar functions throughout. The embodiments described below by referring to the drawings are exemplary only for the purpose of illustrating the present application and are not to be construed as limiting the present application.

As used herein, the singular forms "a", "an", "the" and "the" are intended to include the plural forms as well, unless expressly stated otherwise, as understood by those skilled in the art. It will be further understood that the terms "comprises" and/or "comprising," when used in this specification, specify the presence of stated features, integers, steps, operations, elements, and/or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and/or groups thereof. It will be understood that when an element is referred to as being "connected" or "coupled" to another element, it can be directly connected or coupled to the other element or intervening elements may also be present. Further, "connected" or "coupled" as used herein may include wirelessly connected or wirelessly coupled. The term "and/or" as used herein includes all or any element and all combination of one or more of the associated listed items.

It will be understood by those skilled in the art that all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs unless defined otherwise. It will be further understood that terms, such as those defined in commonly used dictionaries, should be interpreted as having a meaning that is consistent with their meaning in the context of the prior art and will not be interpreted in an idealized or overly formal sense unless expressly so defined herein.

As used herein, "client," "terminal device," and "terminal device" are understood by those skilled in the art to include both devices that include only wireless signal receivers without transmitting capabilities and devices that include receiving and transmitting hardware capable of two-way communication over a two-way communication link. Such a device may include: a cellular or other communication device such as a personal computer, tablet, or the like, having a single-line display or a multi-line display or a cellular or other communication device without a multi-line display; a PCS (Personal Communications Service, personal communication system) that may combine voice, data processing, facsimile and/or data communication capabilities; a PDA (Personal Digital Assistant ) that can include a radio frequency receiver, pager, internet/intranet access, web browser, notepad, calendar and/or GPS (Global Positioning System ) receiver; a conventional laptop and/or palmtop computer or other appliance that has and/or includes a radio frequency receiver. As used herein, "client," "terminal device" may be portable, transportable, installed in a vehicle (aeronautical, maritime, and/or land-based), or adapted and/or configured to operate locally and/or in a distributed fashion, at any other location(s) on earth and/or in space. As used herein, a "client," "terminal device," or "terminal device" may also be a communication terminal, an internet terminal, or a music/video playing terminal, for example, a PDA, a MID (Mobile Internet Device ), and/or a mobile phone with music/video playing function, or may also be a device such as a smart tv, a set top box, or the like.

The hardware referred to by the names "server", "client", "service node" and the like in the present application is essentially an electronic device having the performance of a personal computer, and is a hardware device having necessary components disclosed by von neumann's principle, such as a central processing unit (including an arithmetic unit and a controller), a memory, an input device, and an output device, and a computer program is stored in the memory, and the central processing unit calls the program stored in the external memory to run in the memory, executes instructions in the program, and interacts with the input/output device, thereby completing a specific function.

It should be noted that the concept of "server" as referred to in this application is equally applicable to the case of a server farm. The servers should be logically partitioned, physically separate from each other but interface-callable, or integrated into a physical computer or group of computers, according to network deployment principles understood by those skilled in the art. Those skilled in the art will appreciate this variation and should not be construed as limiting the implementation of the network deployment approach of the present application.

One or several technical features of the present application, unless specified in the plain text, may be deployed either on a server to implement access by remotely invoking an online service interface provided by the acquisition server by a client, or directly deployed and run on the client to implement access.

The neural network model cited or possibly cited in the application can be deployed on a remote server and used for implementing remote call on a client, or can be deployed on a client with sufficient equipment capability for direct call unless specified in a clear text, and in some embodiments, when the neural network model runs on the client, the corresponding intelligence can be obtained through migration learning so as to reduce the requirement on the running resources of the hardware of the client and avoid excessively occupying the running resources of the hardware of the client.

The various data referred to in the present application, unless specified in the plain text, may be stored either remotely in a server or in a local terminal device, as long as it is suitable for being invoked by the technical solution of the present application.

Those skilled in the art will appreciate that: although the various methods of the present application are described based on the same concepts so as to be common to each other, the methods may be performed independently, unless otherwise indicated. Similarly, for each of the embodiments disclosed herein, the concepts presented are based on the same inventive concept, and thus, the concepts presented for the same description, and concepts that are merely convenient and appropriately altered although they are different, should be equally understood.

The various embodiments to be disclosed herein, unless the plain text indicates a mutually exclusive relationship with each other, the technical features related to the various embodiments may be cross-combined to flexibly construct a new embodiment, so long as such combination does not depart from the inventive spirit of the present application and can satisfy the needs in the art or solve the deficiencies in the prior art. This variant will be known to the person skilled in the art.

With reference to the above exemplary scenario, referring to fig. 1, in one embodiment, the method for detecting enterprise operation health of the present application includes:

step S10, determining enterprise operation characteristic data in different time period ranges, wherein the enterprise operation characteristic data comprises one or more of enterprise power consumption, enterprise working time length and newly-increased risk quantity of an enterprise;

the computer terminal equipment of the financial institution can respond to the enterprise operation health degree detection event, the enterprise operation characteristic data acquisition device is adopted to acquire enterprise operation characteristic data in various different time periods, wherein the different time periods comprise a long period, a medium period, a short period, an ultra-short period and the like, the long period can be a time period taking a quarter and a year as time units, the medium period can be a time period taking a month and a quarter as time units, the short period can be a time period taking a week and a month as time units, the ultra-short period can be a time period taking a day and a week as time units, and a person skilled in the art can divide the different time periods as required, so that the method is not limited; the enterprise operation characteristic data can be one or any plurality of enterprise electricity consumption data, enterprise working time length or enterprise newly-increased risk quantity, wherein the enterprise newly-increased risk quantity comprises one or any plurality of enterprise newly-increased legal risk quantity, enterprise newly-increased operation risk quantity and enterprise newly-increased supervision risk quantity, and the enterprise newly-increased legal risk quantity, enterprise newly-increased operation risk quantity and enterprise newly-increased supervision risk quantity can be collected and acquired from the Internet or related information websites.

In some embodiments, after the enterprise operation feature data in each different time period range is obtained, the enterprise operation feature data is preprocessed, and the data, which obviously deviate from the average value or normal value, in the collected enterprise electricity data, the enterprise working time length, the enterprise newly-increased legal risk number, the enterprise newly-increased operation risk number, the enterprise newly-increased supervision risk number and the like in the enterprise operation feature data are checked, if the data are confirmed to be incorrect, the investigation is performed, and meanwhile, because each enterprise operation feature is divided into long, medium, short and ultra-short periods, the data, such as the enterprise electricity data, the enterprise working time length, the enterprise newly-increased legal risk number, the enterprise newly-increased operation risk number, the enterprise newly-increased supervision risk number and the like under different periods, should be calculated and counted respectively.

Step S20, adopting a logic function, a ReLU function and/or a maximum function according to the enterprise electricity consumption, the enterprise working time and the enterprise newly-increased risk quantity to construct an enterprise operation health degree algorithm corresponding to the enterprise electricity consumption, the enterprise working time and the enterprise newly-increased risk quantity in each enterprise operation characteristic data;

after determining the enterprise electricity consumption, the enterprise working time length and the enterprise newly-increased risk quantity in each enterprise operation feature data within different time periods, constructing an enterprise operation health algorithm corresponding to the enterprise electricity consumption, the enterprise working time length and the enterprise newly-increased risk quantity in each enterprise operation feature data by adopting a logic function, a ReLU function and/or a maximum function according to the enterprise electricity consumption, the enterprise working time length and the enterprise newly-increased risk quantity.

In some embodiments, the enterprise operation feature data are listed according to a long period, a medium period, a short period and an ultra-short period, and the enterprise operation health degree algorithm corresponding to each enterprise operation feature data is determined, and relevant parameter values in the enterprise operation health degree algorithm can be reasonably set according to the actual state of enterprise operation based on the enterprise operation health degree algorithm corresponding to each enterprise operation feature data, where the enterprise operation health degree algorithm corresponding to each enterprise operation feature data is expressed as follows:

Step S30, determining enterprise health degrees corresponding to the enterprise power consumption, the enterprise working time length and the newly-increased risk quantity of the enterprise in different time periods based on the enterprise operation health degree algorithm corresponding to the enterprise power consumption, the enterprise working time length and the newly-increased risk quantity of the enterprise;

after the enterprise operation health degree algorithm corresponding to the enterprise power consumption, the enterprise working time length and the enterprise newly-increased risk number is constructed, the enterprise operation health degree algorithm corresponding to the enterprise power consumption, the enterprise working time length and the enterprise newly-increased risk number can be called, and the enterprise power consumption, the enterprise working time length and the enterprise health degree corresponding to the enterprise newly-increased risk number in different time periods are determined according to the enterprise operation health degree algorithm.

Specifically, based on the enterprise operation health algorithm corresponding to the enterprise electricity consumption, the enterprise working time length and the enterprise newly-increased risk number, the enterprise operation health corresponding to the enterprise electricity consumption, the enterprise working time length and the enterprise newly-increased risk number in different time periods is determined.

Firstly, the enterprise operation health degree calculation method corresponding to each enterprise operation characteristic data under a long period is represented as follows:

for example, the business operation health corresponding to the business electricity consumption under a long period is expressed as follows:

1. logit (6+ (10 workday mean line-200 workday mean line value)/200 workday mean line value 10);

2. logit (6+ (20 workday mean line-200 workday mean line value)/200 workday mean line value 10);

3. logit (6+ (40 working day mean line-200 working day mean line value)/200 working day mean line value 10);

4. logit (6+ (80 working day mean line-200 working day mean line value)/200 working day mean line value 10);

5. the electricity consumption is equal on the same month/the electricity consumption is equal on the same month in the last year;

for example, the enterprise business health corresponding to the newly increased legal risk amount, the newly increased business risk amount, and the newly increased regulatory risk amount of the enterprise under a long period is expressed as follows:

6. logic (the newly increased legal risk amount of enterprises in one year);

7. logic (the number of new business risks increased by enterprises in a year);

8. logic (the number of newly added regulatory risks for enterprises in a year);

secondly, the enterprise operation health degree calculation method corresponding to the enterprise operation characteristic data in the middle period is represented as follows:

for example, the business operation health corresponding to the business operation duration in the middle period is expressed as follows:

9. max (2, the actual working time of the current month/the working time of the month)/2;

10. max (2, 10 day working hours/60 day average working hours)/2;

for example, the enterprise business health corresponding to the newly increased legal risk number, the newly increased business risk number, and the newly increased regulatory risk number of the enterprise in the middle period is expressed as follows:

11. logit (the amount of legal risk increased in one month);

12. logic (increase of business risk number in one month);

13. logic (increase of the number of regulatory risks in one month);

thirdly, the enterprise operation health degree calculation method corresponding to the enterprise operation characteristic data under the short period is represented as follows:

for example, the business operation health corresponding to the business electricity consumption under a short period is expressed as follows:

14. max ((1-relu (-current cycle ratio)), 1-relu (current cycle ratio-last year contemporaneous cycle ratio);

15. Logit (6+ (5 weekday mean line-10 weekday mean line value)/10 weekday mean line value 10) 2-1;

16. logit (6+ (5 weekday mean line-30 weekday mean line value)/30 weekday mean line value 10) 2-1;

17. 1-relu (-Zhou Tongbi);

for example, the business operation health corresponding to the business operation duration in the short period is expressed as follows:

18. max (2, actual working time of the current week/working time of the week)/2;

for example, the enterprise business health corresponding to the newly increased legal risk number, the newly increased business risk number, and the newly increased regulatory risk number of the enterprise under the short period is expressed as follows:

19. logic (newly increased legal risk number in a week);

20. logic (newly increased number of business risks in a week);

21. logic (newly increased number of regulatory risks in one week);

22. logit (newly increased number of business risks in three days);

23. logit (the number of newly added regulatory risks in three days);

fourth, the enterprise operation health degree calculation method corresponding to the enterprise operation characteristic data under the ultra-short period is represented as follows:

24. max (2, working time of day/average working time of 10 days)/2;

For example, the business operation health corresponding to the business electricity consumption under the ultra-short period is expressed as follows:

25. logit (6+ (current value-5 working day mean line value)/5 working day mean line value 10) 2-1;

26. logit (6+ (2 weekday mean line-5 weekday mean line value)/5 weekday mean line value 10) 2-1;

27. logit (6+ (3 weekday mean line-5 weekday mean line value)/5 weekday mean line value 10) 2-1;

28. logit (6+ (current value-10 workday mean line value)/10 workday mean line value 10) 2-1;

29. logit (6+ (2 weekday mean line-10 weekday mean line value)/10 weekday mean line value 10) 2-1;

30. logit (6+ (3 weekday mean line-10 weekday mean line value)/10 weekday mean line value 10) 2-1;

31. logit (6+ (4 weekday mean line-10 weekday mean line value)/10 weekday mean line value 10) 2-1;

32. logit (6+ (current value-30 working day mean line value)/30 working day mean line value 10) 2-1;

33. logit (6+ (2 weekday mean line-30 weekday mean line value)/30 weekday mean line value 10) 2-1;

34. logit (6+ (3 weekday mean line-30 weekday mean line value)/30 weekday mean line value 10) 2-1;

35. logit (6+ (4 weekday average line-30 weekday average line value)/30 weekday average line value 10) 2-1.

In some embodiments, in order to exclude legal rest conditions such as holidays and emergency shutdown conditions caused by environmental protection, extreme weather, emergencies and the like in each region, a shutdown judgment check rule base such as holidays and the like for shutdown judgment needs to be set, and related rules are expressed as follows:

s- (b-D), wherein b is a positive integer, S is the number of days of continuous production stoppage, and can be one day, D is the legal holiday or the number of days of scheduled production stoppage in advance in the date of S production stoppage;

for example, 36, (whether or not it is out of production on the day) - (1-whether or not it is a legal holiday or a contracted out of production);

37. (whether or not production is stopped for 2 consecutive days) - (2-legal holidays or days with scheduled production stopping);

38. (whether or not production is stopped for 3 consecutive days) - (3-legal holidays or days with scheduled production stopping);

39. (whether or not there are 4 consecutive days of stalling) - (4-legal holidays or days of scheduled stalling);

40. (whether or not there is a 5-day outage) - (5-day legal holidays or days with scheduled outage) relu (1-if there is no holiday, the number of times a 5-day curve appears historically (historical data also excludes holidays) 0.3);

41. (stopping production in the same area, wherein the distance is within 500m, and whether two or more normal enterprises stop production at the same time or not is abnormal);

42. (whether there are multiple stalls in the same industry).

The method and formula for calculating the enterprise operation feature health degree corresponding to the enterprise electricity consumption, the enterprise working time, the enterprise newly-increased legal risk quantity, the enterprise newly-increased operation risk quantity, the enterprise newly-increased supervision risk quantity and the like are one of embodiments of the application, the application is not limited, a person skilled in the art can determine the enterprise operation health degree corresponding to each enterprise operation feature data according to actual needs as required, relevant parameters and formula construction of the formula can be dynamically adjusted according to the determination result of the enterprise operation state health degree, and the formula number and the collocation of calculation formulas of different time periods can be changed in various ways.

And S40, determining the comprehensive health degree corresponding to the enterprise according to the enterprise power consumption, the enterprise working time length and the enterprise health degree corresponding to the newly-increased risk number of the enterprise in different time periods so as to finish the detection of the enterprise health degree.

After determining the enterprise power consumption, the enterprise working time length and the enterprise operation health degree corresponding to the newly increased risk number of the enterprise in different time periods, invoking a preset linear model according to the enterprise operation health degree corresponding to each enterprise operation feature data, determining the enterprise operation comprehensive health degree corresponding to each enterprise operation feature data, constructing a linear model, and dividing the enterprise operation health degree algorithm corresponding to each enterprise operation feature data in the long period, the medium period, the short period and the ultra-short period, namely an enterprise operation sub-feature health degree calculation formula, inputting the enterprise operation health degree algorithm corresponding to each enterprise operation feature data in different periods into the enterprise operation health degree algorithm, setting an initial weight with the same value, dividing the enterprise comprehensive operation health degree standard into 100 points, dividing the enterprise operation comprehensive health degree standard higher than h (a) into normal operation enterprises, dividing the enterprise comprehensive health degree standard lower than h (a) into operation abnormal attention enterprises, and dividing the enterprise comprehensive health degree data lower than s (b) into operation abnormal alarm enterprises, wherein a and b are set thresholds.

The initial calculation formula of the linear model is as follows:

，

wherein Z is the comprehensive health value of enterprise operation, n is the total number of the characteristic health calculation formulas of enterprise operation,and (3) an enterprise operation health degree algorithm formula corresponding to the enterprise operation characteristic data under different periods.

Compared with the prior art, the method and the device for acquiring the enterprise operation information are not objective, inaccurate and untimely in terms of the channel for acquiring the enterprise operation information and the information existence information acquired by the channel, cannot be quantitatively analyzed, have low working efficiency, and are difficult for the financial institution to timely, accurately and continuously evaluate the production operation state of a target loan enterprise, and the method and the device for acquiring the enterprise operation information by the financial institution adopt a plurality of enterprise operation feature data closely related to the enterprise operation health degree, divide the enterprise operation feature data into sub-feature health degree calculation formulas under different periods, calculate the comprehensive operation health degree value of the enterprise by adopting a linear calculation model, and learn and train the feature weights of the health degrees of the enterprise operation feature data based on an enterprise operation health degree detection model, and have the following beneficial effects:

Referring to fig. 2, after the step of obtaining the business operation feature data of each enterprise in the different time period ranges, according to any embodiment of the present application, the method includes:

step S101, responding to an enterprise operation characteristic data preprocessing instruction, determining average values corresponding to the enterprise operation characteristic data in different time periods, and calculating and determining the difference value between the enterprise operation characteristic data and the average values;

after the enterprise operation feature data in each different time period range is obtained, the enterprise operation feature data is preprocessed, and the data which obviously deviate from the average value or normal value in the collected enterprise operation feature data such as enterprise electricity consumption data, enterprise working time, enterprise newly-increased legal risk quantity, enterprise newly-increased operation risk quantity, enterprise newly-increased supervision risk quantity and the like are removed.

In some embodiments, the calculation formula of the enterprise power consumption average value in the enterprise operation characteristic data of different time periods is:

，/>

wherein the saidCharacterization of the average power consumption value of K days under Z cycle, said +.>The daily electricity quantity on the K-Z+1 day is represented, the Z represents the period corresponding to the average value of the calculated electricity consumption, and the period is flexibly and variously set to be different recording periods such as a long period, a medium period, a short period, an ultra-short period and the like.

And determining an average value corresponding to the enterprise electricity consumption according to a calculation formula of the enterprise electricity consumption average value in the enterprise operation characteristic data, and calculating and determining a difference value between each enterprise electricity consumption and the average value.

Step S103, detecting whether the difference value between the enterprise operation characteristic data and the average value exceeds a preset threshold value, and if so, eliminating the enterprise operation characteristic data.

After the difference value between the power consumption of each enterprise and the average value is calculated and determined, whether the difference value between the power consumption of each enterprise and the average value exceeds a preset threshold value or not is detected, and if so, the power consumption data of the enterprise are removed.

According to the embodiment, the data which deviate from the average value or the normal value obviously in the collected enterprise operation characteristic data such as the enterprise electricity consumption data, the enterprise working time, the enterprise newly-increased legal risk quantity, the enterprise newly-increased operation risk quantity, the enterprise newly-increased supervision risk quantity and the like are removed, so that the enterprise operation characteristic data has greater flexibility and independence, the accuracy of each enterprise operation characteristic data can be ensured, and a solid data base is laid for efficiently and accurately detecting the enterprise operation health degree.

On the basis of any embodiment of the present application, referring to fig. 3, the step of determining the comprehensive health degree corresponding to the enterprise according to the enterprise health degree corresponding to the enterprise power consumption, the enterprise working time length and the number of newly increased risks in the different time periods includes:

step S401, responding to an enterprise operation health degree detection event, and acquiring enterprise electricity consumption, enterprise working time length and enterprise newly-increased risk quantity in different time period ranges;

step S403, an enterprise operation health degree algorithm corresponding to the enterprise electricity consumption, the enterprise working time length and the newly-increased risk number in the enterprise operation feature data is called, and the enterprise operation health degree corresponding to the enterprise operation feature data in different time periods is determined;

step 405, calling a preset linear model according to the enterprise operation health degree corresponding to the enterprise operation feature data, and determining the enterprise operation comprehensive health degree corresponding to each enterprise to complete the detection of the enterprise health degree.

As can be seen from the above embodiments, after determining the comprehensive business operation health corresponding to the business operation feature data, determining the production operation status corresponding to each business according to the comprehensive business operation health, determining the loan amount corresponding to each business based on the production operation status corresponding to each business, and determining the production operation status corresponding to each business according to the comprehensive business operation health, where the production operation status includes one or more of a normal operation status, an abnormal operation attention status, and an abnormal operation alarm status, and a person skilled in the art may determine the production operation status of the business according to the comprehensive business operation health of the business as required, which is not limited herein, and determine the loan amount corresponding to each business based on the production operation status corresponding to each business.

On the basis of any embodiment of the present application, referring to fig. 4, after the step of determining the comprehensive business health of each business corresponding to the business, the method includes:

step S407, responding to the enterprise operation health grade assessment instruction, and determining the score of the enterprise operation comprehensive health degree corresponding to each enterprise;

step S409, determining enterprise operation health grades corresponding to the enterprises according to the scores of the enterprise operation comprehensive health degrees, wherein the enterprise operation health grades comprise one or more of excellent, good, qualified and unqualified.

According to the embodiment, on the basis of detecting the business operation health degree, the business operation health grade corresponding to each business is determined, and each business is divided into the business operation health grades corresponding to each business, such as excellent, good, qualified or unqualified, so that the labor cost and the risk cost of a financial institution are greatly reduced, the business operation state of the business can be objectively, accurately and timely obtained, and the working efficiency is greatly improved.

In some embodiments, after the step of determining the comprehensive health of the enterprise operations corresponding to the respective enterprises, the steps of:

Step S4071, responding to an enterprise risk assessment instruction, detecting whether the enterprise operation health degree exceeds a first health degree threshold, if the enterprise operation health degree exceeds the first health degree threshold, determining the enterprise as a normal operation enterprise, and if the enterprise operation health degree is lower than the first health degree threshold, determining the enterprise as an abnormal operation attention enterprise;

and step S4073, determining the enterprise as an abnormal operation warning enterprise when the enterprise operation health degree is detected to be lower than a second health degree threshold.

Specifically, the comprehensive operation health degree standard of the enterprise can be set to be 100 points, the enterprise is higher than h (a) and is divided into normal operation enterprises, the enterprise is lower than h (a) and is divided into operation abnormality attention enterprises, and the enterprise is lower than s (b) and is divided into operation abnormality alarm enterprises, wherein a and b are set thresholds, the first health degree threshold is h (a), and the second health degree threshold is s (b); the terminal equipment can respond to an enterprise risk assessment instruction, detect whether the enterprise operation health degree exceeds a first health degree threshold h (a), if the enterprise operation health degree exceeds the first health degree threshold, determine the enterprise as a normal operation enterprise, and if the enterprise operation health degree is lower than the first health degree threshold h (a), determine the enterprise as an operation abnormality attention enterprise; and when the business operation health degree of the enterprise is detected to be lower than a second health degree threshold value s (b), determining the enterprise as an abnormal operation alarm enterprise.

According to the embodiment, on the basis of detecting the business operation health degree, the preset threshold value of the business operation health degree is detected to divide each business into corresponding business operation states, such as operation abnormality concerns the business, operation abnormality alarms and the like, so that the business operation health degree corresponding to each business can be accurately and efficiently detected, the labor cost and the risk cost of a financial institution are greatly reduced, the business operation state information of the business can be objectively, accurately and timely obtained, and the working efficiency is greatly improved.

Referring to fig. 5, after determining the comprehensive business health of each business in different time periods, the method includes:

step S4075, determining enterprise operation health degrees corresponding to the enterprise operation feature data, and determining an input sequence of an enterprise health degree detection model according to the enterprise operation health degrees corresponding to the enterprise operation feature data;

step S4077, inputting the input sequence into a pre-trained enterprise health degree detection model, and determining the enterprise operation comprehensive health degree corresponding to each enterprise according to the weight value between every two layers of neurons of the fully-connected neural network in the enterprise health degree detection model and the linear transformation prediction function.

An enterprise health degree detection model capable of autonomously optimizing weight parameters can be constructed, the enterprise health degree detection model is combined with the basic framework of the multi-parameter enterprise comprehensive operation health degree calculation linear model, the sub-feature weights of the linear model are trained and learned by adopting a preset artificial intelligent algorithm, a reasonable loss function is determined for the enterprise health degree detection model, index weights of different sub-features are constantly and autonomously optimized, an enterprise operation health degree value is obtained through comprehensive calculation, and a quantifiable authoritative decision basis is provided for a financial institution to continuously judge enterprise operation state information.

The training of the enterprise health detection model comprises the following steps: determining a data set of a preset enterprise health detection model, and dividing the data set of the enterprise health detection model into a training set and a verification set; according to the training set and the verification set, forward propagation iterative training is carried out on the updated enterprise health detection model, so that in each forward propagation iterative training, an input sequence of the enterprise health detection model is input into the updated enterprise health detection model, errors of actual results of model data samples and calculation results of model training are determined, and a target loss function is calculated; updating model parameters according to the target loss function until the change value of the target loss function is smaller than a preset value or the training times are larger than the preset value, storing the model parameters and completing training of the enterprise health detection model.

The enterprise is healthyThe fitness detection model can be constructed based on a fully connected neural network, and the input-output relationship is thatWherein->Is the output signal of the neuron, called the activation function, which can generally use a linear function, or a nonlinear function can be adopted, and the nonlinear function comprises Sigmoid function and ReLU function; the input signals of the enterprise health degree detection model are transmitted forward, training samples are transmitted from an input layer and are transmitted from an output layer after being processed by all hidden layers, and the calculation process of each layer is as follows: />。

Determining an error between the actual results of the model data samples and the calculated results of the model training when the actual output is not equal to the calculated expected output, and determining an objective loss function of the enterprise health detection model, which may beUpdating model parameters according to the target loss function, optimizing the parameters W and b, and obtaining +.>Substituting the target loss function->In (1) to obtainWherein e is the error between the actual result of the model data sample and the calculated result of the model training,/and->For the actual result of the model data sample, +.>The result of the calculation is trained for the model.

As can be seen from the above equation, the error is reduced by adjusting the weight value of the model, the weight value adjustment amount and the error negative gradient are in a direct proportion relationship, and the calculation formula is as follows Wherein->Adjusting the variation for the weight value, +.>For a given learning rate, e is the model error of the above, +.>And training the enterprise health degree detection model according to the formula as an error until the calculated health degree error meets the actual requirement.

Further, determining index weights corresponding to the enterprise operation feature data in different time periods in the linear model, after training the enterprise health detection model, inputting the enterprise operation health degree corresponding to the enterprise operation feature data into the trained enterprise health detection model, and determining the enterprise operation comprehensive health degree corresponding to each enterprise to finish detection of the enterprise operation health degree.

Specifically, the values of the business health feature data calculated in the above steps are used as inputs to a trained business health detection model. According to the determined model parameters of the enterprise health detection model, an input sequence corresponding to the model parameters can be defined asThe threshold value of the enterprise health degree detection model is b, and the weight value between every two layers of neurons is +.>The prediction function of the linear transformation is +. >。

According to the embodiment, the preset fully-connected neural network is adopted as the basic network architecture of the enterprise health degree detection model, the sub-characteristic weights of the learning linear model are trained, the reasonable loss function is determined for the enterprise health degree detection model, the index weights of different sub-characteristics are continuously and autonomously optimized, the business health degree value of an enterprise is comprehensively calculated, the detection precision of the enterprise health degree can be remarkably improved, manpower and material resources are greatly saved, the working efficiency of a financial institution is remarkably improved, the customer satisfaction degree is greatly improved, and a quantifiable authoritative decision basis is provided for the financial institution to continuously judge the business state information of the enterprise.

Referring to fig. 6, an enterprise business health detection apparatus, which is suitable for one of the purposes of the present application, includes a feature data determining module 1100, a health algorithm constructing module 1200, a single health determining module 1300, and a comprehensive health determining module 1400. The feature data determining module 1100 is configured to determine each enterprise operation feature data within a range of different time periods, where the enterprise operation feature data includes one or more of an enterprise power consumption, an enterprise working time length, and an increased risk number of the enterprise; the health algorithm construction module 1200 is configured to construct an enterprise operation health algorithm corresponding to the enterprise power consumption, the enterprise working time length and the number of newly increased risks of the enterprise in the enterprise operation feature data by adopting a logic function, a ReLU function and/or a maximum function according to the enterprise power consumption, the enterprise working time length and the number of newly increased risks of the enterprise; the single health degree determining module 1300 is configured to determine the health degree of the enterprise corresponding to the enterprise power consumption, the enterprise working time length and the newly increased risk number of the enterprise in different time periods based on the enterprise operation health degree algorithm corresponding to the enterprise power consumption, the enterprise working time length and the newly increased risk number of the enterprise; the comprehensive health determination module 1400 is configured to determine the comprehensive health corresponding to the enterprise according to the enterprise power consumption, the enterprise working time length and the enterprise health corresponding to the number of newly added risks of the enterprise in the different time periods, so as to complete the detection of the enterprise health.

On the basis of any embodiment of the present application, please refer to fig. 7, another embodiment of the present application further provides an electronic device, where the electronic device may be implemented by a computer device, and as shown in fig. 7, the internal structure of the computer device is schematically shown. The computer device includes a processor, a computer readable storage medium, a memory, and a network interface connected by a system bus. The computer readable storage medium of the computer device stores an operating system, a database and computer readable instructions, the database can store a control information sequence, and when the computer readable instructions are executed by a processor, the processor can realize an enterprise operation health degree detection method. The processor of the computer device is used to provide computing and control capabilities, supporting the operation of the entire computer device. The memory of the computer device may store computer readable instructions that, when executed by the processor, cause the processor to perform the enterprise business health detection method of the present application. The network interface of the computer device is for communicating with a terminal connection. It will be appreciated by those skilled in the art that the structure shown in fig. 7 is merely a block diagram of some of the structures associated with the present application and is not limiting of the computer device to which the present application may be applied, and that a particular computer device may include more or fewer components than shown, or may combine certain components, or have a different arrangement of components.

The processor in this embodiment is configured to execute specific functions of each module and its sub-module in fig. 6, and the memory stores program codes and various data required for executing the above modules or sub-modules. The network interface is used for data transmission between the user terminal or the server. The memory in this embodiment stores program codes and data required for executing all modules/sub-modules in the enterprise business health detection apparatus of the present application, and the server can call the program codes and data of the server to execute the functions of all sub-modules.

The present application also provides a storage medium storing computer readable instructions that, when executed by one or more processors, cause the one or more processors to perform the steps of the method for detecting business health of an enterprise according to any embodiment of the present application.

The present application also provides a computer program product comprising computer programs/instructions which, when executed by one or more processors, implement the steps of the method for detecting business health of an enterprise according to any embodiment of the present application.

Those skilled in the art will appreciate that implementing all or part of the above-described methods of embodiments of the present application may be accomplished by way of a computer program stored on a computer readable storage medium, which when executed, may comprise the steps of embodiments of the methods described above. The storage medium may be a computer readable storage medium such as a magnetic disk, an optical disk, a Read-Only Memory (ROM), or a random access Memory (RandomAccess Memory, RAM).

The foregoing is only a partial embodiment of the present application, and it should be noted that, for a person skilled in the art, several improvements and modifications can be made without departing from the principle of the present application, and these improvements and modifications should also be considered as the protection scope of the present application.

In summary, the method and the device can accurately and efficiently detect the business operation health degree of each enterprise, greatly reduce the labor cost and the risk cost of the financial institutions, objectively, accurately and timely acquire the business operation condition information of the enterprises, and greatly improve the working efficiency.

Claims

1. A method for detecting business health of an enterprise, comprising:

2. The method for detecting the business operation health degree according to claim 1, wherein the step of constructing the business operation health degree algorithm corresponding to the business electricity consumption, the business operation time length and the newly added risk number of the business in the business operation feature data by adopting a logic function, a ReLU function and/or a maximum function according to the business electricity consumption, the business operation time length and the newly added risk number of the business comprises the following steps:

3. The method for detecting the business operation health degree according to claim 1, wherein the step of constructing the business operation health degree algorithm corresponding to the business electricity consumption, the business operation time length and the newly added risk number of the business in the business operation feature data by adopting a logic function, a ReLU function and/or a maximum function according to the business electricity consumption, the business operation time length and the newly added risk number of the business comprises the following steps:

4. The method for detecting the business operation health degree according to claim 1, wherein the step of constructing the business operation health degree algorithm corresponding to the business electricity consumption, the business operation time length and the newly added risk number of the business in the business operation feature data by adopting a logic function, a ReLU function and/or a maximum function according to the business electricity consumption, the business operation time length and the newly added risk number of the business comprises the following steps:

5. The method for detecting the business operation health degree of an enterprise according to claim 1, wherein the step of determining the comprehensive health degree corresponding to the enterprise according to the enterprise health degree corresponding to the enterprise power consumption, the enterprise working time length and the number of newly increased risks of the enterprise in different time periods comprises the following steps:

6. The method for detecting the health of business operations according to claim 5, wherein after the step of determining the comprehensive health of business operations corresponding to each business, the method comprises:

7. The method for detecting the business operation health degree according to any one of claims 1 to 5, wherein after the step of determining the business operation integrated health degree corresponding to each business in different time periods, it comprises:

8. An enterprise business health degree detection device, characterized by comprising:

9. An electronic device comprising a central processor and a memory, characterized in that the central processor is arranged to invoke a computer program stored in the memory for performing the steps of the method according to any of claims 1 to 7.

10. A computer-readable storage medium, characterized in that it stores in the form of computer-readable instructions a computer program implemented according to the method of any one of claims 1 to 7, which, when invoked by a computer, performs the steps comprised by the corresponding method.