TWM633489U - Enterprise vision scoring system - Google Patents

Abstract

An enterprise vision scoring system, which is used to provide multiple external hosts to fill out the scoring questionnaire, and then perform the enterprise scoring operation according to the external data provided by the external database, including: a communication module for receiving the scoring questionnaire; a checking and scoring module, It is used to carry out the enterprise scoring operation with the scoring weight after receiving the scoring questionnaire, and generate the enterprise scoring; the machine learning module is used to perform the machine learning operation on the enterprise scoring of the external host, and obtain the corrected scoring weight after adjusting the scoring weight; and An internal database for storing enterprise ratings, rating weights and corrected weights.

Description

Corporate Vision Scoring System

This model discloses a corporate vision scoring system, especially a corporate vision scoring system that can adjust the scoring weight according to different situations.

When any company is operating, it will have different corporate visions. However, the corporate vision stated by many companies is often reduced to an expression, and there is no actual data that can be used as a reference. Therefore, many companies have begun to introduce a corporate vision scoring system.

Among them, ESG score is a common indicator of corporate vision score. ESG refers to the abbreviation of English words such as environment, society, and public governance. The details are Environmental: including energy use management, waste management, climate risk, scarcity of natural resources, pollution and waste, etc. Content; Social: including labor issues, information security, product liability, risk benefits, or employee health and safety management, marketing management, information security privacy protection, etc.; public governance (Governance): corporate governance and behavior Various topics, such as board quality and effectiveness, risk management, business ethics, supply chain management, etc.

The ESG score is a concept of corporate responsibility and sustainable investment. In recent years, more and more investors around the world use ESG scores to measure the social responsibility performance of a company. They believe that this score can measure the external risks of the company. See predict the future performance of a company. Moreover, ESG scoring is also similar to a risk assessment method, which is used to assess whether a company’s external risks are high or not. Although many institutions have established scoring systems to count a company’s ESG score, it may only be done once for a company. It is impossible to keep pace with the times, and the scoring methods of each institution are different, and the weight part will also vary depending on the industry and the location of the country.

Therefore, how to develop a corporate vision scoring system that can adjust the scoring weight according to the current situation, location and various factors has become a very important issue.

The purpose of this model is to provide an enterprise vision scoring system, which is used to provide multiple external hosts to fill out the scoring questionnaire, and then perform enterprise scoring operations according to the external data provided by the external database, including: communication module, checking and scoring module, Machine learning modules and internal databases.

As mentioned above, the communication module, the signal is connected to the external host, and is used to receive the scoring questionnaire; the checking and scoring module, the signal is connected to the communication module, and is used to perform the enterprise scoring operation with the scoring weight after receiving the scoring questionnaire, and generate Enterprise scoring; machine learning module, signal connection checking and scoring module, used to perform machine learning operations on the enterprise scoring of external hosts, to adjust the scoring weights to obtain revised scoring weights; and internal database, signal connection communication modules, The check scoring module and the machine learning module are used to store enterprise scores, scoring weights and corrected weights.

Optionally, the above enterprise rating operation is multiple regression analysis.

Optionally, the enterprise scoring job includes multiple scoring parameters, and the scoring parameters include environmental scoring, social scoring, and corporate governance scoring.

Optionally, the above-mentioned environmental score includes multiple scoring factors, and the above-mentioned factors include electricity consumption per million yuan of revenue per year, water consumption per million yuan of revenue per year, and water consumption per million yuan of revenue per year. Greenhouse gas emissions from revenue, and the total number of third-party certifications related to environmental protection obtained.

Optionally, the above-mentioned social score includes multiple scoring factors, and the above-mentioned factors include the number of violations of labor laws, public welfare expenses per million of revenue, turnover rate, and the number of major adverse events that violate social public welfare.

Optionally, the above-mentioned corporate governance score includes multiple scoring factors, and the above-mentioned factors include pre-tax net profit rate, shareholder return rate, and ratio of operating cash to after-tax profit.

Optionally, the above-mentioned enterprise scoring further includes risk levels, and the above-mentioned risk levels include high-risk levels, medium-risk levels, and low-risk levels.

Optionally, the above-mentioned enterprise scoring further includes a breach of contract, and when one of the above-mentioned external hosts has a breach of contract, the machine learning module adjusts the scoring weight according to the breach of contract.

To sum up, the new enterprise vision scoring system has the following functions: (1) It can adjust the scoring weight in a timely manner according to the scoring standards in external materials and the overall situation of the enterprise; (2) It can regularly update the scoring results of different enterprises, so that The scoring results maintain a certain degree of credibility; and (3) automatically remove unnecessary data through machine learning, making the scoring results closer to the actual operating conditions of the enterprise.

The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of them. . Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without making progressive contributions all belong to the scope of protection of the present invention.

In view of the above problems, an embodiment of the present invention develops an enterprise vision scoring system, please refer to FIG. 1 , which is a functional block diagram of an enterprise vision scoring system according to an embodiment of the present invention. The enterprise vision scoring system 100 mentioned above is used to provide multiple external hosts 200 to fill out the scoring questionnaire, and then perform the enterprise scoring operation according to the external data provided by the external database 300, including: a communication module 110, a checking and scoring module 120, a machine Learning module 130 and internal database 140 .

As mentioned above, the communication module 110 is connected to the external host 200 for receiving scoring questionnaires; the checking and scoring module 120 is connected to the communication module 110 for receiving scoring questionnaires, and is used to score enterprises with scoring weights Operation, and generate enterprise rating; machine learning module 130, signal connection check scoring module 120, used to perform machine learning operation on the enterprise rating of external host 200, to obtain revised scoring weight after adjusting the scoring weight; and internal database 140 , signally connect the communication module 110 , the checking and scoring module 120 and the machine learning module 130 for storing the enterprise score, the scoring weight and the corrected weight.

Wherein, the above-mentioned external data is the standard for each scoring, and after the external data is captured, it is analyzed by the machine learning module 130 to eliminate useless parameters without manually inputting the scoring mechanism and its weight. It is worth noting that machine learning can update these data with similar neural network or AI artificial intelligence, and timely remove redundant information for correction, so that the result can be closer to the real situation.

According to one embodiment of the present invention, the above-mentioned business scoring operation is multiple regression analysis.

According to an embodiment of the present invention, the enterprise scoring job includes a plurality of scoring parameters, and the scoring parameters include environmental scoring, social scoring, and corporate governance scoring.

For example, the Environmental Score, Social Score, and Corporate Governance Score will each have a score (a ₁ , b ₁ , and c ₁ ) and their corresponding score weights (E _i , S _i , and G _i ), which are calculated to add up to The output value (assumed to be the y value) is the enterprise score produced by this enterprise scoring operation. The formula can for example be as follows: y ₁ = a ₁ E _i + b ₁ S _i + c ₁ G _i

Next, the above-mentioned environmental score includes a number of scoring factors. The above-mentioned factors include electricity consumption per million yuan of revenue per year, water consumption per million yuan of revenue per year, and The amount of greenhouse gas emissions and the total number of third-party certifications related to environmental protection obtained.

However, the present invention is not limited thereto, and the environmental scoring factors may also include climate change, natural capital, pollution and waste, and environmental opportunities.

Then, the above-mentioned social scoring also includes multiple scoring factors, including the number of violations of labor laws, public welfare expenses per million of revenue, turnover rate, and the number of major adverse events that violate social public welfare.

However, the present invention is not limited thereto, and the social scoring factors may also be human capital, product responsibility, interest relationship, and social opportunity.

Furthermore, the above-mentioned corporate governance score also includes multiple scoring factors. The above-mentioned factors include corporate governance and corporate behavior, such as pre-tax net interest rate, shareholder return rate, operating cash to after-tax net profit ratio, but this model is not limited to this.

In addition, according to an embodiment of the present invention, the enterprise rating further includes risk levels, and the risk levels include high risk levels, medium risk levels and low risk levels.

For example, the enterprise rating can be not just a number, but also grades, such as grades A to F. If the enterprise has a high score, it will be classified into grades A and B, which means that the enterprise has a low external risk (low risk level) ; If the enterprise score is medium, and classified in C, D level, representing the external risk of the enterprise is medium (medium risk level); and if the enterprise score is very low, classified in E, F level, representing the enterprise's Higher external risks (high risk rating). The above corporate ratings provide investors with a clear basis to judge whether they need to invest before rating different companies, and can also provide job seekers with a tool to evaluate whether the company has harsh treatment of employees before entering a company. The situation that caused the enterprise to score low.

Then, according to an embodiment of the present invention, the above-mentioned enterprise ratings further include breach of contract, when one of the above-mentioned external hosts 200 has a breach of contract, the machine learning module 130 adjusts the scoring weight according to the breach of contract, but the present invention does not take this as a basis Even if the company has not violated the contract, when collecting past data, it is found that some scoring standards are no longer applicable. For example, the carbon emission standard may be much stricter than ten years ago. Therefore, the machine learning module 130 is suitable for The scoring weight of this part will be adjusted, or a company has fully used green energy power generation from a certain year to achieve the goal of sustainable business operation, but there are indicators about non-green power generation in the scoring factor, so the machine The learning module 130 will also adjust the scoring weight of this part.

That is to say, the new enterprise vision scoring system 100 is different from other scoring systems in the past, which only do one-time scoring, and can also be obtained from enterprises (external hosts 200) that participate in the enterprise scoring operation of the enterprise vision scoring system 100 on a regular basis. The information is constantly updated, and the itinerary is closest to the scoring weight of each enterprise type, so that the corporate vision score will not be reduced to a formality or out of date.

Next, another embodiment is given, and please refer to FIG. 2 . FIG. 2 is a flow chart of an enterprise vision scoring system according to an embodiment of the present invention. In step S102 of FIG. 2, the enterprise (external host 200) uses the new type of enterprise vision scoring system 100 to perform ESG scoring within the company. The module 110 sends back to the enterprise vision scoring system 100 .

Then, in steps S104-S106 of FIG. 2 , after receiving the above-mentioned scoring questionnaire delivered by the communication module 110 , the checking and scoring module 120 conducts the enterprise scoring operation according to the scoring weights, and generates enterprise scoring.

For example, the above-mentioned enterprise scoring operation may be multiple regression analysis (as mentioned above), but the present invention is not limited thereto.

Next, in step S108 of FIG. 2 , the checking and scoring module 120 transmits the above-mentioned scoring questionnaire and enterprise scoring to the internal database 140 for storage.

It is worth noting that when performing multiple regression analysis, the weights of each value can be adjusted according to the external data sent by the external database 300. For example, the ESG scoring standards of countries in this year have been greatly revised compared with the past, and the scoring weights will be adjusted. , or automatically adjust the scoring weight according to the location of the participating enterprises or the business category of the enterprises.

In addition, in this embodiment, it is assumed that the scoring questionnaire of this year is the Nth year, and there is no other scoring questionnaire of the enterprise before the Nth year (that is, the scoring weight can be regarded as the initial scoring weight).

Then, in step S110 of FIG. 2 , in the next year, the enterprise also needs to fill in the scoring questionnaire and send it back to the enterprise vision scoring system 100 through the communication module 110. The scoring questionnaire at this time is for the N+1 year. questionnaire.

Next, in step S112 of FIG. 2 , it is determined whether the scoring weight needs to be adjusted according to whether the enterprise has historical data or breach of contract (or as mentioned above, according to whether there is a major change in the scoring criteria in the external data). If there is historical data of the enterprise or there is a breach of contract, proceed to steps S114 and S116A; otherwise, if the enterprise has no historical data or there is no breach of contract, proceed to step S116B.

In step S114 of Figure 2, when the enterprise has historical data or a breach of contract occurs, for example, the enterprise has filled out the rating questionnaire last year and generated a corresponding enterprise rating, which can be used as historical data and Adjust the weights through the machine learning module 130 (ie, machine learning operations); or, if the company has breached the contract during the year, such as a violation of the Labor Standards Act, the weights will be adjusted accordingly through the machine learning module 130 , making the enterprise score lower.

Then, in step S116A of FIG. 2 , when the machine learning module 130 adjusts the above-mentioned scoring weights and obtains the revised scoring weights, it sends back to the checking and scoring module 120, and uses the revised scoring weights to perform the enterprise scoring operation, and then performs Step S118, generating enterprise ratings. It is worth noting that the enterprise in the N+1th year is scored here.

And when the company has no historical data or no breach of contract, and there is no special change in the ESG scoring standards in recent years, proceed to step S116B in Figure 2, use the same scoring weights as before to perform the corporate scoring operation, and then proceed to step S118 , yielding a firm score. The enterprise score obtained here is also the enterprise score of the N+1 year.

Finally, go to step S120 in FIG. 2 , store the corrected or uncorrected rating weights and the company rating for the N+1 year in the internal database 140 , and complete the business rating operation for the N+1 year.

Y _n= a _nE _i+(n-1)+ b _nS _i+(n-1)+ c _nG _i+(n-1) That is to say, when it comes to the N+2 year, N+3 year... and other years, the machine learning module 130 will regularly adjust the scoring weights according to the current situation, the default status of the enterprise, and so on. In addition, when there are many pieces of data, there will be many pieces of multiple regression analysis data (as shown below), and then, when referring to these historical data, the machine learning module 130 will also remove out-of-time peaks according to the situation, so that ESG scores can maintain a high degree of reliability, and can be used as a reference for others or the company's internal corporate policy. y ₁ = a ₁ E _i + b ₁ S _i + c ₁ G _i y ₂ = a ₂ E _i+1 + b ₂ S _i+1 + c ₂ G _i+1 y ₃ = a ₃ E _i+2 + b ₃ S _i+2 + c ₃ G _i+2

Y _n = a _n E _i+(n-1) + b _n S _i+(n-1) + c _n G _i+(n-1)

It is worth noting that this is just an example of the need to reassess the company’s ESG score every year, but this model is not limited to this. The frequency of updating the company’s score can vary, for example: once every three months, every Once every six months, once every two years, or when there is a major change in the scoring standard (learned from external sources), a notification is sent to each external host 200 for re-scoring, so as to keep the ESG score of each enterprise at a very high level. High accuracy without distortion.

To sum up, the new synchronization system for fund rights and interests has the following effects: (1) The scoring weight can be adjusted in a timely manner according to the scoring standards in external materials and the overall situation of the enterprise; (2) The scoring results of different companies can be updated regularly, so that the scoring results maintain a certain degree of credibility; and (3) Automatically remove unnecessary data through machine learning, making the scoring results closer to the actual operating conditions of the enterprise.

Although the embodiments of the present invention have been shown and described, those skilled in the art can understand that various changes, modifications, substitutions and variants, the scope of the present invention is defined by the appended patent scope and its equivalents.

100: Corporate Vision Scoring System 110: Communication module 120: Check and score module 130:Machine Learning Module 140: Internal database 200: external host 300: External database

The accompanying drawings are provided to enable those having ordinary knowledge in the technical field of the present invention to further understand the present invention, and are incorporated and constitute a part of the description of the present invention. The drawings illustrate exemplary embodiments of the invention and, together with the description of the invention, serve to explain the principle of the invention. The following is a brief description of each figure of the present invention: FIG. 1 is a functional block diagram of an enterprise vision scoring system according to an embodiment of the present invention. Fig. 2 is a flow chart of the operation of the enterprise vision scoring system according to an embodiment of the present invention.

100: Corporate Vision Scoring System

110: Communication module

120: Check and score module

130:Machine Learning Module

140: Internal database

200: external host

300: External database

Claims (8)

An enterprise vision scoring system is used to provide a plurality of external hosts to fill out a scoring questionnaire, and then perform an enterprise scoring operation according to an external data provided by an external database, including: A communication module, connected to the external hosts for receiving the scoring questionnaire; A checking and scoring module, which is connected to the communication module with a signal, and is used to perform the scoring operation of the enterprise with a scoring weight after receiving the scoring questionnaire, and generate a corporate scoring; A machine learning module, which is signal-connected to the checking and scoring module, and is used to perform a machine learning operation on the enterprise scoring of the external hosts, so as to obtain a revised scoring weight after adjusting the scoring weight; and An internal database, signally connected to the communication module, the review scoring module and the machine learning module, for storing the enterprise score, the score weight and the corrected weight. The enterprise vision scoring system as described in Claim 1, wherein the enterprise scoring operation is a multiple regression analysis. The enterprise vision scoring system as described in Claim 1, wherein the enterprise scoring operation includes a plurality of scoring parameters, and the scoring parameters include an environmental score, a social score, and a corporate governance score. The enterprise vision scoring system as described in claim 3, wherein the environmental score includes multiple scoring factors, and these scoring factors include electricity consumption per million yuan of revenue per year, and revenue per million yuan of revenue per year The amount of water consumed, the amount of greenhouse gas emissions per million yuan of revenue per year, and the total number of third-party certifications related to environmental protection obtained. The corporate vision scoring system as described in claim 3, wherein the social scoring includes multiple scoring factors, and these scoring factors include the number of violations of labor laws, public welfare expenses per million of revenue, turnover rate, major violations of social welfare number of events. The corporate vision scoring system as described in Claim 3, wherein the corporate governance score includes multiple scoring factors, and these scoring factors include pre-tax net profit rate, shareholder return rate, and ratio of operating cash to after-tax net profit. The enterprise vision scoring system as described in Claim 1, wherein the enterprise scoring further includes a risk level, and the risk level includes a high risk level, a medium risk level and a low risk level. The enterprise vision scoring system as described in Claim 1, wherein the enterprise score further includes a breach condition, and when one of the external hosts has the breach condition, the machine learning module adjusts the scoring weight according to the breach condition .

Images