Deep learning-based definite increase type equity incentive method, system and device, and medium
Through deep learning-based methods, the problems of high cost of formulating equity incentive plans for private placement in the existing technology and uneven distribution of benefits are solved, and the rapid and efficient acquisition of equity incentive plans for private placement is achieved, saving labor costs, and providing enterprises with a basis for judgment and reference for laws and regulations.
Patent Information
- Application Number
- CN202410360654.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-03-27
- Publication Date
- 2025-05-27
AI Technical Summary
When formulating fixed-increase equity incentive plans, the existing technology has problems such as high labor costs, biased selection of issuance objects, uneven distribution of interests and imperfect supervision, which harms the legitimate rights and interests of small and medium-sized investors.
Using a deep learning-based method, we can obtain the basic data of equity incentives for private placement, build a database, and use deep learning units such as recursive neural networks to learn, and output the mapping relationship between the main body, object and conditions of private placement equity incentives for professionals to adjust and optimize.
It has achieved rapid and efficient acquisition of private placement equity incentive schemes, which are suitable for different enterprise scales, greatly saving labor costs, and providing enterprises with a basis for judgment and reference for laws and regulations.
Smart Images
Figure CN120047239A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of information processing, and in particular, to a private placement equity incentive method, system, computer device, and storage medium based on deep learning. Background Art
[0002] The equity incentive system refers to an incentive mechanism commonly adopted by modern enterprises, which provides long-term incentives for the directors, senior executives, and other employees of the enterprise, and is beneficial to the long-term development of the enterprise. With the continuous advancement of the equity branch reform, more and more listed companies adopt the method of private placement of securities (referred to as "private placement") for equity refinancing. Generally speaking, the objects of private placement are the major shareholders of the company and institutional investors with strength. Although the advantages of private placement are obvious, such as low issuance costs, flexible issuance conditions, and the ability to inject high-quality assets, introduce advanced core technologies, and attract outstanding talents, which helps to continuously improve the management level of listed companies and enhance their endogenous competitiveness; however, with the continuous deepening of private placement practices, a series of problems have also emerged one after another. For example, there is a specific bias in the selection of issuance objects, and there are disadvantages in the uneven distribution of interests. Coupled with the need to improve relevant supervision, major shareholders often use resource control rights, and management uses information advantage rights to conduct interest transfers through methods such as manipulating the issuance price, reducing holdings and cashing out, and injecting assets with overvalued prices, damaging the legitimate rights and interests of small and medium investors, and so on.
[0003] With the rapid development of artificial intelligence technology in recent years, some jobs that could only be completed by humans in the past have gradually begun to be replaced by intelligent machines. Among them, equity incentives, especially more complex private placement incentive programs, are usually jointly formulated by various professionals such as accountants, lawyers, investors, and senior corporate managers. Coupled with the different situations of each enterprise, it is not only difficult to have a relatively standard incentive program, but also the labor cost of formulating the incentive program is very high. Summary of the Invention
[0004] To overcome the deficiencies of the prior art, embodiments of the present invention provide a private placement equity incentive method, computer device, and storage medium based on deep learning. Based on the deep learning model, a private placement equity incentive program can be quickly obtained, which is efficient and fast, applicable to different enterprise scales, and greatly saves labor costs.
[0005] A private placement equity incentive method based on deep learning includes:
[0006] Obtain the basic data of private placement equity incentive, where the basic data at least includes private placement equity incentive subject information, private placement equity incentive object information, private placement equity incentive condition information, and evaluation information on the performance of the private placement equity incentive subject;
[0007] Construct a private placement-based equity incentive database according to the basic data, where the private placement-based equity incentive database includes a data scraping interface;
[0008] The preset deep learning unit obtains the basic data from the private placement-based equity incentive database through the data scraping interface and conducts learning for a preset number of rounds; the preset deep learning unit at least includes a recurrent neural network, and during the learning process of the preset deep learning unit, the evaluation information regarding the performance of the private placement-based equity incentive entity is used as an input parameter for the backpropagation algorithm;
[0009] After the iteration of the preset number of rounds, the preset deep learning unit outputs a first mapping relationship that at least includes the private placement-based equity incentive entity, the private placement-based equity incentive object, and the private placement-based equity incentive conditions.
[0010] A private placement-based equity incentive system based on deep learning, comprising:
[0011] A basic data acquisition unit, which is used to acquire the basic data of private placement-based equity incentives. The basic data at least includes private placement-based equity incentive entity information, private placement-based equity incentive object information, private placement-based equity incentive condition information, and evaluation information regarding the performance of the private placement-based equity incentive entity;
[0012] A database construction unit, which is used to construct a private placement-based equity incentive database according to the basic data. The private placement-based equity incentive database includes a data scraping interface;
[0013] A preset deep learning unit, which is used to obtain the basic data from the private placement-based equity incentive database through the data scraping interface and conduct learning for a preset number of rounds; the preset deep learning unit at least includes a recurrent neural network, and during the learning process of the preset deep learning unit, the evaluation information regarding the performance of the private placement-based equity incentive entity is used as an input parameter for the backpropagation algorithm;
[0014] After the iteration of the preset number of rounds, the preset deep learning unit outputs a first mapping relationship that at least includes the private placement-based equity incentive entity, the private placement-based equity incentive object, and the private placement-based equity incentive conditions.
[0015] A computer device includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the steps of the above-mentioned private placement-based equity incentive method based on deep learning
[0016] A computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the above-mentioned additional issuance-based equity incentive method based on deep learning are implemented.
[0017] The above-mentioned additional issuance-based equity incentive method, system, computer device and storage medium based on deep learning use a deep learning module to learn existing cases of additional issuance-based equity incentives, so as to obtain a relatively general additional issuance-based equity incentive plan to balance the interests of all parties, facilitate professionals to adjust and optimize based on the plan output by the deep learning module, and greatly reduce labor costs. Compared with the prior art, through this solution, it can provide a judgment basis for enterprises implementing equity incentive plans or objects receiving additional issuance-based equity incentives, provide constructive references for enterprises to formulate and improve laws and regulations on equity incentives, and also provide specific data references for the implementation of the "additional issuance-based" equity incentive model in the future. Description of the Drawings
[0018] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments of the present invention. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.
[0019] Figure 1 is a flowchart of an additional issuance-based equity incentive method based on deep learning in an embodiment of the present invention;
[0020] Figure 2 is a framework diagram of an additional issuance-based equity incentive system based on deep learning in an embodiment of the present invention. Detailed Embodiments
[0021] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the drawings in the embodiments of the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art without creative efforts based on the embodiments of the present invention belong to the protection scope of the present invention.
[0022] In one embodiment, as Figure 1 shown, an additional issuance-based equity incentive method based on deep learning is provided, which mainly includes the following steps:
[0023] S1, obtaining the basic data of the additional issuance-based equity incentive, and the basic data at least includes the information of the additional issuance-based equity incentive subject, the information of the additional issuance-based equity incentive object, the information of the additional issuance-based equity incentive conditions, and the evaluation information of the performance of the additional issuance-based equity incentive subject.
[0024] Among them, the subject of the private placement-based equity incentive, that is, the subject of the private placement, and the information of this subject includes but is not limited to the industry field to which the subject of the private placement-based equity incentive belongs, the market value before the private placement-based equity incentive, the share capital scale, and so on. The information of the private placement-based equity incentive object includes at least the type of the private placement-based equity incentive object, that is, the object is a natural person, a legal person, or the like. The information of the private placement-based equity incentive conditions includes but is not limited to the number of shares of the private placement-based equity incentive, the incentive time of the private placement-based equity, the unlocking time, the exercisable period, etc.; the evaluation information of the performance of the subject of the private placement-based equity incentive includes at least any one or more of the financial indicators, market indicators, and operation indicators after the private placement-based equity incentive. Financial indicators such as revenue, profit, net profit, gross profit margin, net profit margin, rate of return, etc. are data reflecting the profitability and financial status of the private placement subject. Market indicators such as market share, market value, stock price, etc. are data reflecting the position and performance of the private placement subject in the market. Operation indicators such as product quality, production efficiency, supply chain information, etc.
[0025] Specifically, the above basic data can be entered through a computer device or obtained through relevant channels.
[0026] S2. According to the basic data, construct a private placement-based equity incentive database, and the private placement-based equity incentive database includes a data scraping interface.
[0027] The obtained basic data is stored in a database as a private placement-based equity incentive database to facilitate data scraping, data mining, and learning by the deep learning module. Specifically, the private placement-based equity incentive database is not limited to existing commercial or non-commercial database systems, including but not limited to oracle databases, MySQL databases, etc. Among them, the data scraping interface is used to provide an interface at the API application layer, facilitating various application programs to obtain the basic data of the private placement-based equity incentive through the data scraping interface. For example, for the private placement overview of a certain private placement subject, the capital scale of the corresponding private placement object, the stock price changes before and after the private placement, etc. can be obtained.
[0028] S3. The preset deep learning unit obtains the basic data from the private placement-based equity incentive database through the data scraping interface and conducts learning for a preset number of rounds; the preset deep learning unit includes at least a recurrent neural network, and during the learning process of the preset deep learning unit, the evaluation information of the performance of the subject of the private placement-based equity incentive is used as the input parameter of the backpropagation algorithm.
[0029] The preset deep learning unit mainly includes a recurrent neural network. Specifically, the number of learning rounds of the preset deep learning unit can be preset, and the evaluation information of the performance of the subject of the private placement-based equity incentive is used as the input parameter of the backpropagation algorithm (Back-probagation, BP), and the backpropagation algorithm is used to update the various parameters of the private placement-based equity incentive conditions, such as the optimal range of the incentive duration, the unlocking time, etc.
[0030] S4. After iterations for a preset number of rounds, the preset deep learning unit outputs a first mapping relationship that at least includes a seasoned equity offering-based equity incentive subject, a seasoned equity offering-based equity incentive object, and seasoned equity offering-based equity incentive conditions.
[0031] After undergoing multiple rounds of learning iterations, the preset deep learning unit outputs a learning result that at least includes a first mapping relationship among a seasoned equity offering-based equity incentive subject, a seasoned equity offering-based equity incentive object, and seasoned equity offering-based equity incentive conditions. That is, according to this first mapping relationship, after a user inputs any two of the data items, the third data item can be obtained. For example, by inputting seasoned offering subject information and seasoned offering object information, corresponding seasoned equity incentive conditions can be matched, which can be used as a reference for seasoned equity incentives.
[0032] Further, the preset deep learning unit further includes a convolutional neural network and a recurrent neural network. That is, the three neural networks process data in parallel at different times. Each of the three neural networks includes at least one input layer and one output layer. The input layer is docked with the data scraping interface to provide raw data to the deep learning unit; the output layer outputs the first mapping relationship, the second mapping relationship, and the third mapping relationship that include the relationship among a seasoned equity offering-based equity incentive subject, a seasoned equity offering-based equity incentive object, and seasoned equity offering-based equity incentive conditions. Among them, the three groups of mapping relationships are obtained based on the learning of different neural networks, which is convenient for users to consider comprehensively as a reference.
[0033] Further preferably, at least one hidden layer is included between the input layer and the output layer, and the hidden layer is associated with a preset expected value; the preset expected value includes the expectations of a seasoned equity offering-based equity incentive subject and a seasoned equity offering-based equity incentive object. That is, before the seasoned equity offering, the expectations of some parameters such as market value and unit price of a seasoned equity offering-based equity incentive subject and a seasoned equity offering-based equity incentive object. The hidden layer and the expected value can be used as flexible variables in the learning process of the deep learning unit, which can affect the three groups of mapping relationships output, providing more choice possibilities for users.
[0034] In one embodiment, a seasoned equity offering-based equity incentive system based on deep learning is provided, and its framework diagram is as Figure 2 shown, and mainly includes the following modules:
[0035] A basic data acquisition unit, which is used to acquire the basic data of the seasoned equity offering-based equity incentive. The basic data at least includes seasoned equity offering-based equity incentive subject information, seasoned equity offering-based equity incentive object information, seasoned equity offering-based equity incentive condition information, and evaluation information on the performance of the seasoned equity offering-based equity incentive subject;
[0036] A database construction unit, which is used to construct a seasoned equity offering-based equity incentive database according to the basic data. The seasoned equity offering-based equity incentive database includes a data scraping interface;
[0037] A preset deep learning unit is configured to obtain basic data from a private placement equity incentive database through a data scraping interface and perform learning for a preset number of rounds; the preset deep learning unit includes at least a recurrent neural network, and during the learning process of the preset deep learning unit, the evaluation information for the performance of the equity incentive entity for private placement is used as an input parameter for the backpropagation algorithm;
[0038] After iteration for a preset number of rounds, the preset deep learning unit outputs a first mapping relationship including at least the equity incentive entity for private placement, the equity incentive object for private placement, and the equity incentive conditions for private placement.
[0039] This deep learning-based private placement equity incentive system relies on a specific computer device or system, and its working mode corresponds to the above-mentioned deep learning-based private placement equity incentive method. To avoid repetition, it will not be elaborated here.
[0040] In one embodiment, a computer device is provided, which includes a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the computer program, it implements the steps of the deep learning-based private placement equity incentive method in the above-mentioned embodiment. To avoid repetition, it will not be elaborated here.
[0041] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, it implements the deep learning-based private placement equity incentive method in the above-mentioned method embodiment. To avoid repetition, it will not be elaborated here.
[0042] The above-described embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that: they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements for some of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included in the protection scope of the present invention.
Claims
1. A method of equity incentive based on deep learning, characterized in that: include: Obtaining basic data of fixed-increase equity incentives, wherein the basic data at least includes information on the fixed-increase equity incentive subject, information on the fixed-increase equity incentive object, information on the fixed-increase equity incentive conditions, and evaluation information on the performance of the fixed-increase equity incentive subject; According to the basic data, a fixed-increase equity incentive database is constructed, wherein the fixed-increase equity incentive database includes a data capture interface; The preset deep learning unit obtains the basic data from the fixed-increase equity incentive database through the data capture interface and performs a preset number of rounds of learning; the preset deep learning unit at least includes a recursive neural network, and in the learning process of the preset deep learning unit, the evaluation information on the performance of the incremental equity incentive subject is used as an input parameter of the back propagation algorithm; After the preset number of iterations, the preset deep learning unit outputs a first mapping relationship including at least the fixed-increase equity incentive subject, the fixed-increase equity incentive object and the fixed-increase equity incentive conditions.
2. The method for equity incentive based on deep learning as claimed in claim 1, characterized in that: The information on the subject of the fixed-increase equity incentive at least includes the industry field to which the subject of the fixed-increase equity incentive belongs, the market value and equity scale before the fixed-increase equity incentive; the information on the objects of the fixed-increase equity incentive at least includes the type of the objects of the fixed-increase equity incentive; the information on the conditions of the fixed-increase equity incentive at least includes any one or more of the number of shares of the fixed-increase equity incentive, the incentive time of the fixed-increase equity, the unlocking time, and the exercise period; the evaluation information on the performance of the subject of the fixed-increase equity incentive at least includes any one or more of the financial indicators, market indicators, and operating indicators after the fixed-increase equity incentive.
3. The method for fixed increase equity incentive based on deep learning as claimed in claim 1 or 2, characterized in that: The preset deep learning unit also includes a convolutional neural network and a recurrent neural network; based on the convolutional neural network and the recurrent neural network, after a preset number of iterations, the preset deep learning unit respectively outputs a second mapping relationship and a third mapping relationship including the fixed-increase equity incentive subject, the fixed-increase equity incentive object and the fixed-increase equity incentive conditions.
4. The method for equity incentive based on deep learning as claimed in claim 3, characterized in that: The recursive neural network, convolutional neural network and recurrent neural network respectively include an input layer and an output layer, and include at least one hidden layer between the input layer and the output layer, and the hidden layer is associated with a preset expected value; the preset expected value includes the expectations of the fixed-increase equity incentive subject and the fixed-increase equity incentive object.
5. A fixed increase equity incentive system based on deep learning, characterized in that: include: A basic data acquisition unit, the basic data acquisition unit is used to acquire basic data of the fixed-increase equity incentive, the basic data at least including fixed-increase equity incentive subject information, fixed-increase equity incentive object information, fixed-increase equity incentive condition information, and evaluation information on the performance of the fixed-increase equity incentive subject; A database construction unit, the database construction unit is used to construct a fixed-increase equity incentive database based on the basic data, and the fixed-increase equity incentive database includes a data capture interface; A preset deep learning unit, the preset deep learning unit is used to obtain the basic data from the fixed-increase equity incentive database through the data capture interface and perform a preset number of rounds of learning; the preset deep learning unit at least includes a recursive neural network, and in the learning process of the preset deep learning unit, the evaluation information on the performance of the incremental equity incentive subject is used as an input parameter of the back propagation algorithm; After the preset number of iterations, the preset deep learning unit outputs a first mapping relationship including at least the fixed-increase equity incentive subject, the fixed-increase equity incentive object and the fixed-increase equity incentive conditions.
6. The deep learning-based equity incentive system according to claim 5, characterized in that: The preset deep learning unit also includes a convolutional neural network and a recurrent neural network; based on the convolutional neural network and the recurrent neural network, after a preset number of iterations, the preset deep learning unit respectively outputs a second mapping relationship and a third mapping relationship including the fixed-increase equity incentive subject, the fixed-increase equity incentive object and the fixed-increase equity incentive conditions.
7. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the steps of the fixed-increase equity incentive method based on deep learning as described in any one of claims 1 to 4 are implemented.
8. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the steps of the fixed-increase equity incentive method based on deep learning as described in any one of claims 1 to 4 are implemented.