TWM647924U - Intelligent financial information predicting and alarming system - Google Patents

Abstract

An intelligent financial information predicting and alarming system includes a trading device configured to generate a plurality of trading data, a data transmitting unit and a server. The data transmitting unit sends the trading data to the server. A database of the server stores a plurality of historical trading data of the trading device. A predicting and analyzing unit of the server analyzes the historical trading data by an algorithm to generate a predicting model and analyzes the trading data by the predicting model to generate a predicting result. When the difference between the trading mount and a predicting trading number of the predicting result is greater than a trading number threshold value, the comparing and alarming unit sends an alarm message.

Description

Intelligent financial communication prediction and alarm system

The present invention relates to a predictive warning system, and in particular, to an intelligent financial communication predictive warning system for predicting and monitoring financial communication security events.

Financial institutions mainly play the role of financial intermediaries. When customers need to conduct transactions such as using funds, applying funds, and buying and selling commodities, they usually conduct transactions through the trading tools of financial institutions or trading devices connected to financial institutions. With the rapid development of the Internet and electronic technology, financial institutions not only provide physical trading devices, but also provide virtual or online trading platforms for customers to conduct transactions to improve convenience.

With the diversity and diversification of trading tools, transaction data generated during transactions are transmitted through a variety of transmission methods, and financial institutions are paying more and more attention to information security. Therefore, the government has also formulated relevant information security regulations: When an information system information security incident occurs in a financial institution and causes damage to the rights and interests of customers or affects the operation of the financial institution, the financial institution shall notify the financial institution within a certain period of time (30 minutes) after confirmation. Report to the banking bureau internally and apply online. However, because financial institutions have a very large number of trading tools and devices, as well as transaction volumes, most of which are monitored and judged manually, which not only increases labor costs, but also results in extremely poor monitoring efficiency and accuracy. Moreover, when relevant personnel of financial institutions notice that there is an abnormality in capital security, they often If an IT security incident has already occurred, or even occurred for some time, not only cannot it be reported within the time required to comply with regulations, but it has also caused losses to customers and financial institutions.

In view of this, one of the scopes of this invention is to provide an intelligent financial communication prediction and alarm system to solve the problems of previous technologies. In a specific embodiment, the intelligent financial communication prediction and alarm system includes a transaction device, a data integration and transmission unit and a server. The trading device is used to generate a plurality of transaction data based on the transaction behavior, wherein the transaction data respectively include transaction time. The data integration and transmission unit is communicatively connected to the transaction device and is used to receive and send transaction data generated by the transaction device. The server connects to the data integration transmission unit and is used to receive transaction data. The server includes a database, a predictive analysis unit, and a comparison alarm unit. The database is used to store multiple historical transaction data of the transaction device. The prediction analysis unit uses an algorithm to analyze historical transaction data to establish a prediction model, and the prediction model analyzes the transaction data to generate prediction results corresponding to the transaction device. The prediction results include multiple predicted transaction times and their number of transactions. When the predicted transaction time is the same as the actual transaction time, and if the difference between the predicted number of transactions and the actual number of transactions is likely to be greater than the threshold, the comparison alarm unit will issue an alarm message.

Among them, the intelligent financial communication prediction and alarm system further includes a monitor connection server. The comparison alarm unit generates analysis reports based on transaction data, actual transaction numbers, and prediction results, and the display is used to receive and display the analysis reports.

Among them, the intelligent financial communication prediction and alarm system further includes a terminal device connected to the server. The comparison alarm unit generates analysis reports based on transaction data, actual transaction numbers and prediction results, and the server sends alarm messages and analysis reports to the terminal device.

Among them, the database is used to store transaction data and the actual number of transactions, and pre- The measurement and analysis unit uses algorithms to analyze historical transaction data, transaction data and actual transaction numbers to update the prediction model.

Among them, the intelligent financial communication prediction and alarm system further includes a plurality of transaction device connection data integration and transmission units, and these transaction devices include front-end transaction devices and back-end transaction devices. The front-end transaction device is used to send transaction requests according to transaction behavior, the data integration and transmission unit transmits the transaction requests to the back-end transaction device, and the back-end transaction device generates transaction data according to the transaction requests.

Among them, the data integration and transmission unit is the Enterprise Service Bus (ESB).

Among them, the algorithm is Long Short-Term Memory (LSTM).

Another aspect of this invention is to provide an intelligent financial communication prediction and alarm method to solve the problems of the previous technology.

In a specific embodiment, the intelligent financial communication prediction and alarm method includes the following steps: the transaction device generates a plurality of transaction data according to the transaction behavior, where the transaction data respectively include transaction time; the data integration and transmission unit receives the transaction data generated by the transaction device and Send transaction data to the server; the server's predictive analysis unit uses an algorithm to analyze historical transaction data to establish a prediction model. The prediction model analyzes the transaction data to generate prediction results corresponding to the trading device. The prediction results include a plurality of predicted transaction times and their transactions. The number of transactions; and when the predicted transaction time is the same as the actual transaction time and the difference between the predicted transaction number and the actual transaction number is greater than the transaction number threshold, the comparison alarm unit issues an alarm message.

Among them, the step of generating multiple transaction data based on transaction behavior on the transaction device , further including the following steps: the front-end transaction device sends a transaction request according to the transaction behavior; the data integration and transmission unit transmits the transaction request to the back-end transaction device; and the back-end transaction device generates transaction data according to the transaction request.

Among them, the intelligent financial communication prediction and alarm method further includes the following steps: the prediction analysis unit uses an algorithm to analyze historical transaction data, transaction data and actual transaction numbers to update the prediction model.

To sum up, the intelligent financial communication prediction and alarm system created by this invention can automatically predict, analyze and monitor the information service status of trading devices through prediction models without the need for manual inspection and judgment, thus reducing labor costs and time costs. Moreover, the intelligent financial communication prediction and alarm system created by this invention can detect abnormal information service status of the transaction device early and present analysis reports and alarm messages in real time, allowing relevant personnel to make judgments and follow-up processing based on the alarm messages at the first time, so as to Prevent customer rights and financial institutions' operations from being affected, thereby improving security and practicality. In addition, the intelligent financial communication prediction and alarm system created by this creation can continuously update and optimize the prediction model, thereby improving efficiency and the accuracy of abnormal detection.

1: Intelligent financial communication prediction and alarm system

11:Trading device

11A: Front desk transaction device

11B:Backend trading device

12: Data integration and transmission unit

13:Server

131:Database

132: Predictive analysis unit

1321: Predictive model

133: Compare alarm unit

14:Display

15:Terminal device

P: Predict trading curve

R: actual trading curve

T1: the first time

T2: The second time

T3: The third time

S1~S5, S11~S13: steps

Figure 1 is a functional block diagram of an intelligent financial financing prediction and alarm system according to a specific embodiment of the present invention.

Figure 2 is a schematic diagram illustrating an analysis report generated by the comparison alarm unit of the intelligent financial financing prediction and alarm system according to a specific embodiment of the present invention.

Figure 3 is a flowchart illustrating the steps of an intelligent financial financing prediction and alarm method according to a specific embodiment of the present invention.

Figure 4 is a flowchart illustrating the steps of an intelligent financial financing prediction and alarm method according to a specific embodiment of the present invention.

In order to make the advantages, spirit and characteristics of this invention more easily and clearly understood, specific embodiments will be described and discussed in detail with reference to the attached drawings. It is worth noting that these specific embodiments are only representative specific embodiments of the present invention, and the specific methods, devices, conditions, materials, etc. exemplified therein are not intended to limit the present invention or the corresponding specific embodiments. In addition, each device in the figure is only used to express its relative position and is not drawn according to its actual proportion, so it will be explained first.

In various embodiments of the present disclosure, the expression "or" includes any or all combinations of words listed simultaneously. For example, the expression "A or B" may include A, may include B, or may include both A and B. In addition, the indefinite articles "a", "an" and "an" before the device or component of this creation have no limit on the quantity requirements (ie, the number of occurrences) of the device or component. Therefore "a" should be read to include one or at least one, and the singular form of a device or element also includes the plural form, unless the number clearly refers to the singular form.

In the description of this specification, reference to the description of the terms "one embodiment," "another embodiment," or "part of the embodiment" means that the specific features, structures, materials, or characteristics described in connection with the embodiment include In at least one embodiment of the present invention. In this specification, schematic expressions of the above terms do not necessarily refer to the same embodiment. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments.

Please refer to Figure 1. Figure 1 is a functional block diagram of an intelligent financial communication prediction and alarm system 1 according to a specific embodiment of the present invention. As shown in Figure 1, in this specific embodiment, the smart The energy financial communication prediction and warning system 1 includes a plurality of transaction devices 11 , a data integration and transmission unit 12 and a server 13 . A plurality of transaction devices 11 are communicatively connected to the data integration and transmission unit 12 , and the data integration and transmission unit 12 is connected to the server 13 . In practice, the transaction device 11 can be wirelessly connected to the data integration and transmission unit 12 through the Internet, communication protocols, etc., and the data integration and transmission unit 12 can be connected to the server 13 in a wired or wireless manner. The server 13 may be a cloud server, a private cloud, or a host in a data center of a financial institution.

In this specific embodiment, the transaction device 11 is used to generate a plurality of transaction data based on transaction behavior. In practice, the transaction device 11 may be a cash machine (ATM), an online banking platform/server, an electronic payment card reader (POS machine), or other physical/virtual devices that can perform transactions. The transaction data can be transaction message information or transaction details, and the transaction data can include the name of the transaction device, the type of the transaction device, the number of the transaction device, the transaction amount, the transaction time, the message code, etc., where the message code can be the transaction method of the transaction device 11 code. For example, when the transaction device 11 is a cash machine, the cash withdrawal function, deposit function and transfer function of the cash machine can respectively correspond to different message codes. After the transaction device 11 generates transaction data, the transaction device 11 transmits the transaction data to the data integration and transmission unit 12 .

In addition to independently generating transaction data, the transaction device may also be in other forms. As shown in Figure 1, in this specific embodiment, the transaction device 11 may further include a front-end transaction device 11A and a back-end transaction device 11B. The front-end transaction device 11A is used to send a transaction request according to the transaction behavior, and the back-end transaction device 11B generates transaction data based on the transaction request. In practice, the front-end transaction device 11A may be a cash machine (ATM), and the back-end transaction device 11B may be an accounting host of a financial institution. When a user uses a cash machine to withdraw money, the cash machine will generate and send a transaction request based on the user's operation. At this time, the data integration and transmission unit 12 will receive and transmit the transaction request to Account host. Then, the accounting host will generate and send an authorization instruction to the data integration and transmission unit 12 according to the transaction request. Finally, the cash machine will receive the authorization instruction from the data integration and transmission unit 12 and provide banknotes according to the authorization instruction to complete the withdrawal transaction. . When the accounting host generates the authorization command, it will also record all the messages exchanged with the cash machine and generate transaction data, and transmit the authorization command and transaction data to the data integration and transmission unit 12 .

In this specific embodiment, the data integration and transmission unit 12 is used to receive transaction data generated by the transaction device 11 and to transmit all transaction data to the server 13 . In practice, the data integration and transmission unit is the Enterprise Service Bus (ESB) and can be integrated into the host or server. Since different types of transaction devices 11 have different communication methods, data transmission methods, communication protocols, transaction data formats or data encodings, the enterprise service bus can interface with multiple communication protocols to receive all types of transaction devices 11 The generated transaction data can also provide functions such as transcoding, format conversion, and logical judgment to integrate the transaction data into a unified format to serve as an intermediary device between the transaction device 11 and the server 13 .

In this specific embodiment, the server 13 includes a database 131, a prediction analysis unit 132, and a comparison alarm unit 133. The database 131 is connected to the prediction analysis unit 132 , and the prediction analysis unit 132 is connected to the comparison alarm unit 133 . In practice, the database 131 may be a hard disk. The prediction analysis unit 132 and the comparison alarm unit 133 may each be a single chip, but are not limited thereto. The prediction analysis unit 132 and the comparison alarm unit 133 can also be integrated into the same chip (such as a central processing unit).

In this specific embodiment, the database 131 is used to store a plurality of historical transaction data of all transaction devices 11 . In practice, historical transaction data may include in addition to the data content of the aforementioned transaction data items, it may also include integrated historical transaction data. For example, at a certain time in the past, the transaction device 11 completed a total of N transactions. The number of historical transactions, where the time can be one A specific time, date, or period of time. In addition, the historical transaction data can also be divided into a plurality of historical transaction data groups according to the types of different transaction devices 11 and/or the functions of the transaction devices 11 . In practical applications, the historical transaction data can record the service status of the transaction device 11, and can also reflect the distribution relationship of the transaction volume of the transaction device 11 at different time points.

In this specific embodiment, the prediction analysis unit 132 uses an algorithm to analyze historical transaction data stored in the database 131 to establish a prediction model 1321. In practice, the algorithm may be a long short-term memory algorithm (Long Short-Term Memory, LSTM). The prediction analysis unit 132 can use the aforementioned historical transaction data of all transaction devices 11 as a training set, and then establish a prediction model 1321 for predicting the service status of the transaction device 11 through the LSTM algorithm. It is worth noting that, in addition to constructing a prediction model through the LSTM algorithm, the prediction analysis unit 132 can also be constructed through supervised learning, unsupervised learning, semi-supervised learning, deep learning or other prediction algorithms. Predictive model. Then, the prediction analysis unit 132 uses the prediction model 1321 to analyze the transaction data transmitted by the data integration transmission unit 12 to generate prediction results corresponding to the transaction device 11 . In practical applications, since the historical transaction data includes the number of past historical transactions of the transaction device 11, after the server 13 receives the transaction data sent by the data integration and transmission unit 12, the prediction analysis unit 132 can input the transaction data into In the prediction model 1321, the prediction results of the distribution relationship between the transaction device 11 and the transaction volume at future time points are analyzed to predict the future service status of the transaction device 11. The prediction result may include a plurality of predicted transaction numbers, and each predicted transaction number includes a predicted transaction time. Furthermore, the predicted trading time can be a period of time, and each predicted trading time can correspond to a predicted number of transactions. It is worth noting that the prediction analysis unit 132 can generate a plurality of prediction results based on the transaction data of all transaction devices 11 , and different types of transaction devices 11 can respectively correspond to different prediction results.

In this specific embodiment, the comparison alarm unit 133 is used to calculate and generate an actual number of transactions corresponding to the transaction time based on the transaction data. In practice, when the transaction data received by the server 13 is used as the input data of the prediction model 1321 of the prediction analysis unit 132, the comparison and alarm unit 133 will also count the transaction time of each transaction device 11 and its corresponding actual transactions. Number of pens. In practical applications, since the data integration and transmission unit 12 will continue to receive and transmit the transaction data of all transaction devices 11 at different times, the comparison and alarm unit 133 will also keep recording the current service status of all transaction devices 11 .

Please refer to Figure 1 and Figure 2 together. FIG. 2 is a schematic diagram illustrating an analysis report generated by the comparison alarm unit 133 of the intelligent financial financing prediction and alarm system 1 according to a specific embodiment of the present invention. In this specific embodiment, the comparison and alarm unit 133 further creates an analysis report based on the transaction data and actual number of transactions of all transaction devices 11 and the prediction results generated by the prediction analysis unit 132 . As shown in Figure 2, in practice, the analysis report can be presented in the form of a chart, but is not limited to this. The comparison and alarm unit 133 can draw a predicted transaction curve P based on the plurality of predicted transaction times and the number of predicted transactions in the prediction results, and the comparison and alarm unit 133 can also draw a predicted transaction curve P based on the transaction time in the transaction data and the actual transactions corresponding to the transaction time. The number of transactions is plotted as the actual transaction curve R. It is worth noting that Figure 2 only presents a chart of one type of trading device. In practice, the analysis report may include charts of multiple different trading devices.

In addition, in this specific embodiment, the comparison alarm unit 133 further pre-stores a transaction number threshold. When the transaction time matches the predicted transaction time and the difference between the actual number of transactions and the number of predicted transactions is greater than the transaction number threshold, the comparison alarm unit 133 sends an alarm message. In practice, the transaction number threshold can be a numerical value or a standard deviation value, and the transaction number threshold can be determined by professionals and historical experience. The alarm message can be text, sound, image, Illustrations, markings, etc. In practical applications, when the difference between the actual number of transactions and the number of predicted transactions is greater than the transaction number threshold, it means that the actual number of transactions of the trading device 11 suddenly increases sharply or is much lower than the predicted number of transactions. When, the comparison alarm unit 133 sends an alarm message.

As shown in Figure 2, when the difference between the actual number of transactions of the actual transaction curve R and the predicted number of transactions of the predicted transaction curve P is greater than the transaction number threshold at the first time T1, the comparison alarm unit 133 can Mark (dot) corresponding to the first time T1 on the predicted trading curve P and the actual trading curve R as an alarm message. Moreover, when the difference between the actual number of transactions of the actual transaction curve R and the predicted number of transactions of the predicted transaction curve P is greater than the transaction number threshold in the interval from the second time T2 to the third time T3, then The warning unit 133 may mark lines of different colors/widths on the predicted trading curve P and the actual trading curve R in the interval corresponding to the second time T2 to the third time T3 as warning messages.

As shown in FIG. 1 , in this specific embodiment, the intelligent financial communication prediction and alarm system 1 further includes a display 14 connected to a server 13 . After the comparison alarm unit 133 generates the analysis report, the server 13 can transmit the analysis report to the display 14, and the display 14 receives and displays the analysis report. In practice, the display 14 may be an electronic bulletin board or a transaction monitoring screen/dashboard of a financial institution. Furthermore, the display 14 can also display the alarm message sent by the comparison alarm unit 133 . It is worth noting that the display 14 can display analysis reports of multiple trading devices 11 at the same time.

When the intelligent financial communication prediction and alarm system 1 of the present invention is in operation, the data integration and transmission unit 12 will continue to receive transaction data from all transaction devices 11 and transmit the transaction data to the server 13 . Then, the prediction analysis unit 132 of the server 13 analyzes the transaction data through the prediction model 1321 to generate the prediction results of the transaction device 11. At the same time, the comparison and alarm unit 133 of the server 13 records and integrates the transaction data, and integrates the transaction data according to the prediction results and integration Establish corresponding transactions based on subsequent transaction data Analysis report of easy device 11. Then, the comparison and alarm unit 133 compares the actual number of transactions and the predicted number of transactions of the transaction device 11 based on the transaction number threshold to see whether there are any abnormalities, and generates alarm messages for abnormal data. Finally, the server 13 can transmit the analysis report to the display 14, and the display 14 receives and displays the analysis report for viewing by relevant personnel of the financial institution, and the relevant personnel can perform subsequent processing based on the alarm message.

In practical applications, when the transaction device 11 is a cash machine (ATM) and the actual number of transactions of the transaction device 11 is much lower than the predicted number of transactions, it means that the ATM may be down due to a machine crash, severe system delays, front and backend failures. If the service status is abnormal due to transaction abnormality or connection function abnormality, at this time, relevant personnel of the financial institution can immediately confirm the actual cause of the service status abnormality through the alarm message of the analysis report on the display 14 . If the actual reason will cause damage to customers' rights and interests or a financial security incident that affects the operation of financial institutions, relevant personnel can report to the banking bureau or online. When the transaction device 11 is an online bank and the actual number of transactions of the transaction device 11 is much higher than the predicted number of transactions, it may be that the financial institution temporarily holds activities or irregular promotions/businesses, resulting in a sudden increase in transaction volume. At this time, relevant personnel of the financial institution can immediately confirm the actual cause with the business unit through the alarm message of the analysis report on the display 14 . If the actual reason is to hold an event, the relevant personnel can ignore the alarm message and not report it.

Therefore, the intelligent financial communication prediction and alarm system created by this invention can automatically predict, analyze and monitor the information service status of the trading device through the prediction model without the need for manual inspection and judgment, thereby reducing labor costs and time costs. Moreover, the intelligent financial communication prediction and alarm system created by this invention can detect abnormal information service status of the transaction device early and present analysis reports and alarm messages in real time, allowing relevant personnel to make judgments and follow-up processing based on the alarm messages at the first time, so as to Prevent customers’ rights and financial institutions’ operations from being affected, thereby improving security and practicality.

In addition to being installed in the server, the aforementioned database can also be in other forms. In a specific embodiment, the database can be installed outside the server and connected to the server in a wired or wireless manner. In practice, the database can be a cloud database or a Network Attached Storage (NAS). The function of the database in this specific embodiment is substantially the same as the function of the database in the previous specific embodiment, and will not be described again here.

See Figure 1 again. In this specific embodiment, the intelligent financial communication prediction and alarm system 1 further includes a terminal device 15 connected to the server 13 . In practice, the terminal device 15 can be a smart phone, a notebook computer, a tablet computer, etc., and the terminal device 15 can be connected to the server 13 in a wired or wireless manner. When the comparison alarm unit 133 of the server 13 generates the analysis report and the alarm message, the server 13 can transmit the analysis report and the alarm message to the terminal device 15 for viewing by relevant personnel. It is worth noting that the server 13 can also send analysis reports and alarm messages to the display 14 and the terminal device 15 at the same time.

In this specific embodiment, the database 131 is used to store transaction data and the actual number of transactions, and the prediction analysis unit 132 uses an algorithm to analyze the historical transaction data, transaction data and the actual number of transactions stored in the database 131 to update Predictive Models1321. In practice, after the server 13 receives the transaction data of all transaction devices sent by the data integration and transmission unit 12, in addition to analyzing the transaction data, the predictive analysis unit 132 can also copy the transaction data and store it in the database 131 middle. Moreover, after the comparison and alarm unit 133 calculates and generates the actual number of transactions based on the transaction data, the comparison and alarm unit 133 can also store the actual number of transactions of all transaction devices in the database 131 . At this time, the transaction data and the actual number of transactions stored in the database 131 can be used as historical data, and the predictive analysis unit 132 can further analyze the historical transaction data to determine and historical data to update the prediction model 1321. Therefore, the intelligent financial communication prediction and alarm system created by this invention can continuously update and optimize the prediction model, thereby improving efficiency and the accuracy of abnormal detection.

Please refer to Figure 1 and Figure 3 together. Figure 3 is a flowchart illustrating the steps of an intelligent financial financing prediction and alarm method according to a specific embodiment of the present invention. The steps in Figure 3 can be executed by the intelligent financial communication prediction and alarm system of Figure 1 . As shown in Figure 3, in this specific embodiment, the intelligent financial financing prediction and alarm method includes the following steps: Step S1: The transaction device 11 generates a plurality of transaction data according to the transaction behavior, where the transaction data respectively include transaction time; Step S2: The data integration and transmission unit 12 receives the transaction data generated by the transaction device 11 and sends the transaction data to the server 13; Step S3: The prediction analysis unit 132 of the server 13 analyzes a plurality of historical transaction data stored in the database 131 using an algorithm. Establish a prediction model 1321, and use the prediction model 1321 to analyze the transaction data to generate a prediction result corresponding to the transaction device 11, where the prediction result includes a plurality of predicted transaction numbers, and each predicted transaction number includes a predicted transaction time; and step S4: The comparison alarm unit 133 of the server 13 calculates and generates the actual number of transactions corresponding to the transaction time based on the transaction data. When the transaction time matches the predicted transaction time and the difference between the actual number of transactions and the predicted number of transactions is greater than the transaction number threshold When the value is reached, the comparison alarm unit 133 sends an alarm message.

Please refer to Figure 1 and Figure 4 together. Figure 4 is a flowchart illustrating the steps of an intelligent financial financing prediction and alarm method according to a specific embodiment of the present invention. The steps of Figure 4 can be executed by the intelligent financial communication prediction and alarm system of Figure 1, and Figure 4 is a further step process of Figure 3. As shown in Figure 4, in this specific embodiment, the intelligent financial financing prediction and alarm method includes the following steps: Step S11: The front-end transaction device 11A sends a transaction request according to the transaction behavior; Step S12: The data integration and transmission unit 12 sends the transaction request to The backend transaction device 11B; and step S13: the backend transaction device 11B generates transaction data according to the transaction request. Moreover, the intelligent financial communication prediction and alarm method further improves the The steps include the following steps: Step S5: The prediction analysis unit 132 uses an algorithm to analyze historical transaction data, transaction data and the actual number of transactions to update the prediction model 1321.

To sum up, the intelligent financial communication prediction and alarm system created by this invention can automatically predict, analyze and monitor the information service status of trading devices through prediction models without the need for manual inspection and judgment, thus reducing labor costs and time costs. Moreover, the intelligent financial communication prediction and alarm system created by this invention can detect abnormal information service status of the transaction device early and present analysis reports and alarm messages in real time, allowing relevant personnel to make judgments and follow-up processing based on the alarm messages at the first time, so as to Prevent customer rights and financial institutions' operations from being affected, thereby improving security and practicality. In addition, the intelligent financial communication prediction and alarm system created by this creation can continuously update and optimize the prediction model, thereby improving efficiency and the accuracy of abnormal detection.

Through the above detailed description of the preferred embodiments, it is hoped that the characteristics and spirit of the present invention can be more clearly described, but the scope of the present invention is not limited by the above-mentioned preferred embodiments. On the contrary, the purpose is to cover various modifications and equivalent arrangements within the scope of the patent application for this invention. Therefore, the scope of the patentable scope of this invention should be interpreted in the broadest manner according to the above description, so as to cover all possible changes and equivalent arrangements.

1: Intelligent financial communication prediction and alarm system

11:Trading device

11A: Front desk transaction device

11B:Backend trading device

12: Data integration and transmission unit

13:Server

131:Database

132: Predictive analysis unit

1321: Predictive model

133: Compare alarm unit

14:Display

15:Terminal device

Claims (7)

An intelligent financial circulation prediction and alarm system, including: A trading device used to generate a plurality of transaction data based on a transaction behavior, wherein each of the transaction data includes a transaction time; A data integration and transmission unit, communication connected to the transaction device, the data integration and transmission unit is used to receive and send the transaction data generated by the transaction device; and A server connected to the data integration transmission unit and used to receive the transaction data. The server further includes: A database used to store multiple historical transaction data of the transaction device; A prediction analysis unit uses an algorithm to analyze the historical transaction data to establish a prediction model, and uses the prediction model to analyze the transaction data to generate a prediction result corresponding to the trading device, wherein the prediction result includes a plurality of The number of predicted transactions, and each of the number of predicted transactions includes a predicted transaction time; and A comparison alarm unit pre-stores a transaction number threshold, and is used to calculate and generate an actual transaction number corresponding to the transaction time based on the transaction data. When the transaction time meets the predicted transaction time and the actual transaction number When the difference between the number and the predicted number of transactions is greater than the threshold of the number of transactions, the comparison alarm unit sends an alarm message. The intelligent financial communication prediction and alarm system described in item 1 of the patent application scope further includes a display connected to the server, and the comparison alarm unit generates an analysis based on the transaction data, the actual number of transactions and the prediction result. report, and the display is used to receive and display the analysis Report. The intelligent financial communication prediction and alarm system described in Item 1 of the patent application scope further includes a terminal device connected to the server, and the comparison alarm unit generates an alarm based on the transaction data, the actual number of transactions and the prediction result. Analyze the report, and the server sends the alarm message and the analysis report to the terminal device. For example, the intelligent financial communication prediction and alarm system described in item 1 of the patent application scope, wherein the database is used to store the transaction data and the actual number of transactions, and the prediction analysis unit analyzes the historical transactions using the algorithm The prediction model is updated based on the data, the transaction data and the actual number of transactions. The intelligent financial communication prediction and alarm system described in item 1 of the patent application scope further includes a plurality of the transaction devices connected to the data integration transmission unit, and the transaction devices include a front-end transaction device and a back-end transaction device. The front-end transaction device The transaction device is used to send a transaction request according to the transaction behavior, the data integration and transmission unit transmits the transaction request to the backend transaction device, and the backend transaction device generates the transaction data according to the transaction request. For example, in the intelligent financial communication prediction and alarm system described in Item 1 of the patent application, the data integration and transmission unit is an Enterprise Service Bus (ESB). For example, in the intelligent financial communication prediction and alarm system described in item 1 of the patent application, the algorithm is a long short-term memory algorithm (Long Short-Term Memory, LSTM).

Images