Internet financial management system

Abstract

The invention relates to an internet financial management system, and the system comprises a transaction prediction mechanism which is used for intelligently predicting the total number of transactions, the number of peak transactions and the number of peak and valley transactions of a current security business point on the day through an AI transaction prediction model; and the machine distribution mechanism is used for determining the total number of the transaction servers needing to be distributed to the current security business point on the current day based on the peak transaction number of the current security business point on the current day. The internet financial management system is stable in operation and intelligent in control. The AI transaction prediction model designed in a customized structure can be adopted to intelligently predict the total number of transactions, the number of peak transactions and the number of peak and valley transactions of the current security business office on the day, so that the total number of transaction servers needing to be allocated to the current security business office on the day is determined; therefore, limited transaction server resources are fully utilized, and the transaction efficiency and the transaction effect of each security business office are ensured.

Description

Technical Field

[0001] The present invention relates to the field of Internet finance, and in particular to an Internet finance management system. Background Art

[0002] Financial management is an investment company that invests in equity in financial enterprises, exercises the rights and obligations of investors in financial enterprises in accordance with the law within the limit of the investment amount, and realizes the preservation and appreciation of financial assets. It can also be understood as a financial intermediary institution that pools the funds of individual investors and invests them in many securities or other assets. In terms of investment, it is necessary to choose a suitable investment portfolio based on one's own risk tolerance and financial goals. People with low risk preferences may prefer more stable investment methods such as savings and bonds; while people with high risk preferences may be involved in high-return but also high-risk investment areas such as stocks and funds. At the same time, regular evaluation and adjustment of the investment portfolio is also an indispensable link. Risk management occupies an important position in financial management. This includes the identification and prevention of market risk, credit risk, liquidity risk, etc. For example, companies may reduce risks by purchasing insurance and signing hedging contracts.

[0003] The number of trading servers of the same securities business company is limited, and some of them are even leased from third-party companies. The resources of trading servers are very precious. The same securities business company has multiple securities business outlets under its jurisdiction. Therefore, the total number of trading servers that need to be allocated to multiple securities business outlets needs to be determined according to the number of transactions of the multiple securities business outlets. The current difficulty is that the number of transactions of each securities business outlet in the same time period in the future is difficult to determine in advance, resulting in a lack of data basis for the allocation of trading servers, which can easily lead to insufficient or excessive number of trading servers. Summary of the invention

[0004] In order to solve the technical problems in the related fields, the present invention provides an Internet financial management system, which performs training actions on the Hoffitt neural network to obtain the Hoffitt neural network after completing each training action and outputs it as an AI transaction prediction model. The number of each training action performed by the Hoffitt neural network is positively correlated with the number of registered users at the current securities business point, so that AI transaction prediction models with different structures are designed for different securities business points in a targeted manner, and a reliable prediction mechanism is provided for the intelligent prediction of the number of transactions per day at each securities business point. The various configuration data of the current securities business point are obtained, namely, the number of blocks under the jurisdiction of the securities business point, the number of permanent residents in the blocks under the jurisdiction, the number of registered users, and the average number of transactions of registered users, and the number of transactions corresponding to the current securities business point in the past days is obtained. According to Yi Information, the single transaction information corresponding to the current securities business point every day is the total number of transactions, peak number of transactions and peak-valley number of transactions of the current securities business point on that day, thereby providing sufficient and comprehensive basic information for the intelligent prediction of the number of transactions of the current securities business point in a single day. The AI ​​transaction prediction model is also used to intelligently predict the total number of transactions, peak number of transactions and peak-valley number of transactions of the current securities business point on that day, and the total number of transaction servers that need to be allocated to the current securities business point on that day is determined based on the peak number of transactions of the current securities business point on that day. The total number of transaction servers that need to be allocated to the current securities business point on that day is positively correlated with the peak number of transactions of the current securities business point on that day, thereby making full use of limited transaction server resources and ensuring the transaction efficiency and transaction effect of each securities business point.

[0005] According to the present invention, there is provided an Internet financial management system, the system comprising: An object training device, used to perform each training action on the Hoffet neural network to obtain the Hoffet neural network after each training action is completed and output as an AI transaction prediction model, wherein the number of each training action performed by the Hoffet neural network is positively correlated with the number of registered users of the current securities business point; The first capture device is used to obtain various configuration data of the current securities business point, wherein the various configuration data of the current securities business point are the number of blocks under the jurisdiction of the securities business point, the number of permanent residents in the blocks under the jurisdiction, the number of registered users, and the average number of transactions by registered users; The second capture device is used to obtain the transaction information corresponding to each day of the current securities business point. The single transaction information corresponding to each day of the current securities business point is the total number of transactions, the peak number of transactions, and the peak and valley number of transactions of the current securities business point on that day; A transaction prediction mechanism is connected to the object training device, the first capture device and the second capture device respectively, and is used to input various configuration data of the current securities business point, various transaction information corresponding to the previous days of the current securities business point and the duration of the time segment into the AI ​​transaction prediction model, and run the AI ​​transaction prediction model to obtain the total number of transactions, the peak number of transactions and the peak-valley number of transactions of the current securities business point on the day output by the AI ​​transaction prediction model; A machine allocation mechanism, connected to the transaction prediction mechanism, for determining the total number of transaction servers that need to be allocated to the current securities business point on the day based on the peak transaction number of the current securities business point on the day; Wherein, determining the total number of transaction servers that need to be allocated to the current securities business point on the day based on the peak number of transactions of the current securities business point on the day includes: the total number of transaction servers that need to be allocated to the current securities business point on the day is positively correlated with the peak number of transactions of the current securities business point on the day; Among them, various configuration data of the current securities business point, various transaction information corresponding to the current securities business point in the past days, and the duration length of the time segment are input into the AI ​​transaction prediction model together, and the AI ​​transaction prediction model is run to obtain the total number of transactions, the peak number of transactions, and the peak-valley number of transactions of the current securities business point on the day output by the AI ​​transaction prediction model, including: the day and the past days form a complete time interval and at zero o'clock on the day, various configuration data of the current securities business point, various transaction information corresponding to the current securities business point in the past days, and the duration length of the time segment are input into the AI ​​transaction prediction model together, and the AI ​​transaction prediction model is run to obtain the total number of transactions, the peak number of transactions, and the peak-valley number of transactions of the current securities business point on the day output by the AI ​​transaction prediction model.

[0006] Therefore, the present invention has the following three outstanding technical effects: Technical Effect 1: The Hoffitt neural network is trained each time to obtain the Hoffitt neural network after each training action and output as the AI ​​transaction prediction model. The number of each training action performed by the Hoffitt neural network is positively correlated with the number of registered users at the current securities business point, so that AI transaction prediction models with different structures are designed for different securities business points, providing a reliable prediction mechanism for the intelligent prediction of the number of transactions per day at each securities business point; Technical effect 2: The configuration data of the current securities business point are obtained, which are the number of blocks under the jurisdiction of the securities business point, the number of permanent residents in the blocks under the jurisdiction, the number of registered users, and the average number of transactions of registered users, and the transaction information corresponding to each day of the current securities business point is obtained. The single transaction information corresponding to each day of the current securities business point is the total number of transactions, the peak number of transactions, and the peak and valley number of transactions of the current securities business point on that day, thereby providing sufficient and comprehensive basic information for the intelligent prediction of the number of transactions in a single day of the current securities business point; Technical effect three: The AI ​​transaction prediction model is used to intelligently predict the total number of transactions, peak number of transactions and peak-valley number of transactions of the current securities business point on the day, and the total number of transaction servers that need to be allocated to the current securities business point on the day is determined based on the peak number of transactions of the current securities business point on the day. The total number of transaction servers that need to be allocated to the current securities business point on the day is positively correlated with the peak number of transactions of the current securities business point on the day, thereby making full use of limited transaction server resources and ensuring the transaction efficiency and transaction effect of each securities business point.

[0007] The Internet financial management system of the present invention has stable operation and intelligent control. Since the AI ​​transaction prediction model with customized structure design can be used to intelligently predict the total number of transactions, peak transaction number and peak-valley transaction number of the current securities business point on the day, the total number of transaction servers that need to be allocated to the current securities business point on the day is determined, thereby making full use of limited transaction server resources and ensuring the transaction efficiency and transaction effect of each securities business point. BRIEF DESCRIPTION OF THE DRAWINGS

[0008] The embodiments of the present invention will be described below with reference to the accompanying drawings, wherein: Figure 1 The figure is a structural block diagram of an Internet financial management system according to the first embodiment of the present invention.

[0009] Figure 2 The figure is a structural block diagram of an Internet financial management system according to the second embodiment of the present invention.

[0010] Figure 3 The structure block diagram of the Internet financial management system according to the third embodiment of the present invention is shown. DETAILED DESCRIPTION

[0011] The embodiment of the Internet financial management system of the present invention will be described in detail below with reference to the accompanying drawings.

[0012] Figure 1 FIG. 1 is a structural block diagram of an Internet financial management system according to Embodiment 1 of the present invention. The system includes: An object training device, used to perform each training action on the Hoffet neural network to obtain the Hoffet neural network after each training action is completed and output as an AI transaction prediction model, wherein the number of each training action performed by the Hoffet neural network is positively correlated with the number of registered users of the current securities business point; Specifically, the object training device is used to perform each training action on the Hoffet neural network to obtain the Hoffet neural network after completing each training action and output it as an AI transaction prediction model, and the number of each training action performed by the Hoffet neural network is positively correlated with the number of registered users at the current securities business point, including: selecting a numerical conversion formula to represent the numerical conversion relationship of the number of each training action performed by the Hoffet neural network and the number of registered users at the current securities business point; The first capture device is used to obtain various configuration data of the current securities business point, wherein the various configuration data of the current securities business point are the number of blocks under the jurisdiction of the securities business point, the number of permanent residents in the blocks under the jurisdiction, the number of registered users, and the average number of transactions by registered users; The second capture device is used to obtain the transaction information corresponding to each day of the current securities business point. The single transaction information corresponding to each day of the current securities business point is the total number of transactions, the peak number of transactions, and the peak and valley number of transactions of the current securities business point on that day; A transaction prediction mechanism is connected to the object training device, the first capture device and the second capture device respectively, and is used to input various configuration data of the current securities business point, various transaction information corresponding to the previous days of the current securities business point and the duration of the time segment into the AI ​​transaction prediction model, and run the AI ​​transaction prediction model to obtain the total number of transactions, the peak number of transactions and the peak-valley number of transactions of the current securities business point on the day output by the AI ​​transaction prediction model; A machine allocation mechanism, connected to the transaction prediction mechanism, for determining the total number of transaction servers that need to be allocated to the current securities business point on the day based on the peak transaction number of the current securities business point on the day; Wherein, determining the total number of transaction servers that need to be allocated to the current securities business point on the day based on the peak number of transactions of the current securities business point on the day includes: the total number of transaction servers that need to be allocated to the current securities business point on the day is positively correlated with the peak number of transactions of the current securities business point on the day; Among them, inputting various configuration data of the current securities business point, various transaction information corresponding to the previous days of the current securities business point, and the duration length of the time segment into the AI ​​transaction prediction model, and running the AI ​​transaction prediction model to obtain the total number of transactions, the peak number of transactions, and the peak-valley number of transactions of the current securities business point on the day output by the AI ​​transaction prediction model includes: the current day and the previous days form a complete time interval and at zero o'clock on the current day, inputting various configuration data of the current securities business point, various transaction information corresponding to the previous days of the current securities business point, and the duration length of the time segment into the AI ​​transaction prediction model, and running the AI ​​transaction prediction model to obtain the total number of transactions, the peak number of transactions, and the peak-valley number of transactions of the current securities business point on the day output by the AI ​​transaction prediction model; Among them, obtaining each transaction information corresponding to each past day of the current securities business point, the single transaction information corresponding to each day of the current securities business point is the total number of transactions, the peak number of transactions and the valley number of transactions of the current securities business point on that day, including: the peak number of transactions of the current securities business point on that day is to divide the day into various time segments of equal duration, obtain each transaction number corresponding to each time segment, take the time segment corresponding to the maximum transaction number as the peak transaction time segment, and take the time segment corresponding to the minimum transaction number as the valley transaction time segment; Among them, various configuration data of the current securities business point are obtained, and the various configuration data of the current securities business point are the number of blocks under the jurisdiction of the securities business point, the number of permanent residents in the blocks under the jurisdiction, the number of registered users, and the average transaction frequency of registered users, including: the average number of transactions of registered users of the securities business point is the average of the number of transactions per day of each registered user of the securities business point.

[0013] Figure 2 The figure is a structural block diagram of an Internet financial management system according to the second embodiment of the present invention.

[0014] Compared to Figure 1 , Figure 2 The Internet financial management system may also include: a feature extraction component, connected to the transaction prediction mechanism, the object training device, the first capture device, and the second capture device, respectively, and used to measure the current power usage of each of the transaction prediction mechanism, the object training device, the first capture device, and the second capture device; Wherein, the feature extraction component is respectively connected to the transaction prediction mechanism, the object training device, the first capture device and the second capture device, and is used to respectively measure the current power usage of the transaction prediction mechanism, the object training device, the first capture device and the second capture device, including: the feature extraction component includes a plurality of power measurement units, which are respectively connected to the transaction prediction mechanism, the object training device, the first capture device and the second capture device, so as to complete the respective measurement of the current power usage of the transaction prediction mechanism, the object training device, the first capture device and the second capture device; Wherein, the feature extraction component includes a plurality of power measurement units, which are used to be connected to the transaction prediction mechanism, the object training device, the first capture device and the second capture device respectively, so as to complete the respective measurement of the current power usage of the transaction prediction mechanism, the object training device, the first capture device and the second capture device respectively, including: the plurality of power measurement units are a plurality of power sensing circuits, which are used to be connected to the transaction prediction mechanism, the object training device, the first capture device and the second capture device respectively, so as to complete the respective measurement of the current power usage of the transaction prediction mechanism, the object training device, the first capture device and the second capture device respectively; Wherein, the plurality of power measurement units are a plurality of power sensing circuits, which are used to be respectively connected to the transaction prediction mechanism, the object training device, the first capture device and the second capture device to respectively measure the current power usage of the transaction prediction mechanism, the object training device, the first capture device and the second capture device, including: the structures of the plurality of power sensing circuits are the same; Among them, the multiple power measurement units are multiple power sensing circuits, which are used to be connected to the transaction prediction mechanism, the object training device, the first capture device and the second capture device respectively, so as to complete the separate measurement of the current power usage of the transaction prediction mechanism, the object training device, the first capture device and the second capture device respectively, and also include: the multiple power sensing circuits have the same power measurement upper limit value and power measurement lower limit value.

[0015] Figure 3 The structure block diagram of the Internet financial management system according to the third embodiment of the present invention is shown.

[0016] Compared to Figure 1 , Figure 3 The Internet financial management system may also include: a response processing device, arranged near the transaction prediction mechanism, the object training device, the first capture device and the second capture device, and used to provide the transaction prediction mechanism, the object training device, the first capture device and the second capture device with heat dissipation services required by each of them; Wherein, the response processing device is arranged near the transaction prediction mechanism, the object training device, the first capture device and the second capture device, and is used to provide the transaction prediction mechanism, the object training device, the first capture device and the second capture device with the heat dissipation services required by each of them, including: the response processing device includes a plurality of fan heat dissipation units, which are respectively arranged near the transaction prediction mechanism, the object training device, the first capture device and the second capture device to provide the transaction prediction mechanism, the object training device, the first capture device and the second capture device with the heat dissipation services required by each of them; And wherein, the response processing device includes a plurality of fan cooling units, which are respectively arranged near the transaction prediction mechanism, the object training device, the first capture device and the second capture device to provide the transaction prediction mechanism, the object training device, the first capture device and the second capture device with their respective required cooling services, including: the rated output power of the plurality of fan cooling units is equal.

[0017] In addition, in the Internet financial management system, each training action is performed on the Hoffet neural network to obtain the Hoffet neural network after each training action is completed and output as the AI ​​transaction prediction model, and the number of each training action performed by the Hoffet neural network is positively correlated with the number of registered users of the current securities business point, including: using an information conversion formula to represent the information conversion relationship of the positive correlation between the number of each training action performed by the Hoffet neural network and the number of registered users of the current securities business point; And wherein, the information conversion relationship that uses the information conversion formula to represent the positive correlation between the number of training actions performed by the Hoffet neural network and the number of registered users of the current securities business outlet includes: in the information conversion formula, the number of registered users of the current securities business outlet is the input data, and the number of training actions performed by the Hoffet neural network is the output data.

[0018] Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art will appreciate that various modifications and substitutions may be made thereto without departing from the spirit and scope of the invention as set forth in the appended claims.

Claims

1. An Internet financial management system, characterized in that: The system comprises: An object training device, used to perform each training action on the Hoffet neural network to obtain the Hoffet neural network after each training action is completed and output as an AI transaction prediction model, wherein the number of each training action performed by the Hoffet neural network is positively correlated with the number of registered users of the current securities business point; The first capture device is used to obtain various configuration data of the current securities business point, wherein the various configuration data of the current securities business point are the number of blocks under the jurisdiction of the securities business point, the number of permanent residents in the blocks under the jurisdiction, the number of registered users, and the average number of transactions by registered users; The second capture device is used to obtain the transaction information corresponding to each day of the current securities business point in the past. The single transaction information corresponding to each day of the current securities business point is the total number of transactions, the peak number of transactions, and the peak and valley number of transactions of the current securities business point on that day; A transaction prediction mechanism is connected to the object training device, the first capture device and the second capture device respectively, and is used to input various configuration data of the current securities business point, various transaction information corresponding to the previous days of the current securities business point and the duration of the time segment into the AI ​​transaction prediction model, and run the AI ​​transaction prediction model to obtain the total number of transactions, the peak number of transactions and the peak-valley number of transactions of the current securities business point on the day output by the AI ​​transaction prediction model; A machine allocation mechanism, connected to the transaction prediction mechanism, for determining the total number of transaction servers that need to be allocated to the current securities business point on the day based on the peak transaction number of the current securities business point on the day; Wherein, determining the total number of transaction servers that need to be allocated to the current securities business point on the day based on the peak number of transactions of the current securities business point on the day includes: the total number of transaction servers that need to be allocated to the current securities business point on the day is positively correlated with the peak number of transactions of the current securities business point on the day; Among them, various configuration data of the current securities business point, various transaction information corresponding to the current securities business point in the past days, and the duration length of the time segment are input into the AI ​​transaction prediction model together, and the AI ​​transaction prediction model is run to obtain the total number of transactions, the peak number of transactions, and the peak-valley number of transactions of the current securities business point on the day output by the AI ​​transaction prediction model, including: the day and the past days form a complete time interval and at zero o'clock on the day, various configuration data of the current securities business point, various transaction information corresponding to the current securities business point in the past days, and the duration length of the time segment are input into the AI ​​transaction prediction model together, and the AI ​​transaction prediction model is run to obtain the total number of transactions, the peak number of transactions, and the peak-valley number of transactions of the current securities business point on the day output by the AI ​​transaction prediction model.

2. The Internet financial management system according to claim 1, characterized in that: Obtaining each transaction information corresponding to each past day of the current securities business point, the single transaction information corresponding to each day of the current securities business point is the total number of transactions, the peak number of transactions and the valley number of transactions of the current securities business point on that day, including: the peak number of transactions of the current securities business point on that day is to divide the day into various time segments of equal duration, obtain each transaction number corresponding to each time segment, take the time segment corresponding to the maximum transaction number as the peak transaction time segment, and take the time segment corresponding to the minimum transaction number as the valley transaction time segment; Among them, various configuration data of the current securities business point are obtained, and the various configuration data of the current securities business point are the number of blocks under the jurisdiction of the securities business point, the number of permanent residents in the blocks under the jurisdiction, the number of registered users, and the average transaction frequency of registered users, including: the average number of transactions of registered users of the securities business point is the average of the number of transactions per day of each registered user of the securities business point.

3. The Internet financial management system according to claim 2, characterized in that: The system further comprises: a feature extraction component, connected to the transaction prediction mechanism, the object training device, the first capture device, and the second capture device, respectively, and used to measure the current power usage of each of the transaction prediction mechanism, the object training device, the first capture device, and the second capture device; Among them, the feature extraction component is respectively connected to the transaction prediction mechanism, the object training device, the first capture device and the second capture device, and is used to respectively measure the current power usage of the transaction prediction mechanism, the object training device, the first capture device and the second capture device. It includes: the feature extraction component includes multiple power measurement units, which are used to be respectively connected to the transaction prediction mechanism, the object training device, the first capture device and the second capture device to complete the respective measurements of the current power usage of the transaction prediction mechanism, the object training device, the first capture device and the second capture device.

4. The Internet financial management system according to claim 3, characterized in that: The feature extraction component includes a plurality of power measurement units, which are used to be respectively connected to the transaction prediction mechanism, the object training device, the first capture device and the second capture device to complete the respective measurements of the current power usage of the transaction prediction mechanism, the object training device, the first capture device and the second capture device. The plurality of power measurement units are a plurality of power sensing circuits, which are used to be respectively connected to the transaction prediction mechanism, the object training device, the first capture device and the second capture device to complete the respective measurements of the current power usage of the transaction prediction mechanism, the object training device, the first capture device and the second capture device.

5. The Internet financial management system according to claim 4, characterized in that: The multiple power measurement units are multiple power sensing circuits, which are used to be connected to the transaction prediction mechanism, the object training device, the first capture device and the second capture device respectively, so as to complete the respective measurements of the current power usage of the transaction prediction mechanism, the object training device, the first capture device and the second capture device, including: the structures of the multiple power sensing circuits are the same.

6. The Internet financial management system according to claim 5, characterized in that: The multiple power measurement units are multiple power sensing circuits, which are used to be connected to the transaction prediction mechanism, the object training device, the first capture device and the second capture device respectively, so as to complete the separate measurement of the current power usage of the transaction prediction mechanism, the object training device, the first capture device and the second capture device respectively, and also include: the multiple power sensing circuits have the same power measurement upper limit value and power measurement lower limit value.

7. The Internet financial management system according to claim 3, characterized in that: The system further comprises: The response processing device is arranged near the transaction prediction mechanism, the object training device, the first capture device and the second capture device, and is used to provide the transaction prediction mechanism, the object training device, the first capture device and the second capture device with the heat dissipation services they need respectively.

8. The Internet financial management system according to claim 7, characterized in that: A response processing device is arranged near the transaction prediction mechanism, the object training device, the first capture device and the second capture device, and is used to provide the transaction prediction mechanism, the object training device, the first capture device and the second capture device with the heat dissipation services they need respectively. The response processing device includes multiple fan heat dissipation units, which are respectively arranged near the transaction prediction mechanism, the object training device, the first capture device and the second capture device to provide the transaction prediction mechanism, the object training device, the first capture device and the second capture device with the heat dissipation services they need respectively.

9. The Internet financial management system according to claim 8, characterized in that: The response processing device includes multiple fan cooling units, which are respectively arranged near the transaction prediction mechanism, the object training device, the first capture device and the second capture device to provide the transaction prediction mechanism, the object training device, the first capture device and the second capture device with their respective required cooling services, including: the rated output power of the multiple fan cooling units is equal.

Images