TW202307759A - Automatic bookkeeping system and automatic bookkeeping method based on user feedback - Google Patents

Abstract

An automatic bookkeeping system and an automatic bookkeeping method are provided. The automatic bookkeeping system includes a server and a terminal device. The server is communicatively connected to the terminal device, wherein the server accesses a financial account and transmits a push message corresponding to a transaction event to the terminal device in response to detecting the transaction event corresponding to the financial account, wherein the terminal device transmits a feedback message corresponding to the push message to the server, wherein the server determines a classification of the transaction event according to the feedback message in response to receiving the feedback message, and updates an account report corresponding to the financial account according to the transaction event and the classification.

Description

Automatic accounting system and automatic accounting method based on user feedback

This disclosure relates to an automatic accounting system and automatic accounting method based on user feedback.

Currently, there are many online banking applications on the market that provide billing services. The online banking application can automatically judge the transaction classification when the transaction occurs, but the result of the system's automatic judgment is often not very accurate.

This disclosure provides an automatic accounting system and automatic accounting method based on user feedback, which can determine the classification of transaction events according to user feedback.

An automatic billing system based on user feedback disclosed in the present disclosure includes a terminal device and a server. The server is communicatively connected to the terminal device, wherein the server accesses the financial account corresponding to the terminal device, and transmits a push message corresponding to the transaction event to the terminal device in response to detecting a transaction event corresponding to the financial account, wherein the terminal The device sends a feedback message corresponding to the push message to the server, wherein the server determines the classification of the transaction event according to the feedback message in response to receiving the feedback message, and updates the account statement corresponding to the financial account according to the transaction event and the classification.

In an embodiment of the present disclosure, the above-mentioned server determines the classification of the transaction event according to a preset rule in response to not receiving the feedback message.

In an embodiment of the present disclosure, the above-mentioned server determines the suggested classification of the transaction event according to a preset rule, wherein the push message includes the suggested classification.

In an embodiment of the present disclosure, the above-mentioned feedback message includes an approval instruction, wherein the server determines the classification of the transaction event as a suggested classification according to the approval instruction.

In an embodiment of the present disclosure, the above-mentioned server obtains the historical transaction records corresponding to the financial account, and generates a forecast model corresponding to the financial account according to the historical transaction records, wherein the server responds to not receiving the feedback message according to the forecast The model determines the classification of transaction events.

In an embodiment of the present disclosure, the above-mentioned server obtains the historical transaction records corresponding to the financial account, and generates a forecast model corresponding to the financial account according to the historical transaction records, wherein the server determines the suggested classification of the transaction event according to the forecast model, The push message includes suggested categories.

In an embodiment of the present disclosure, the prediction model mentioned above is a machine learning model, wherein the machine learning model includes a long short-term memory model.

In an embodiment of the present disclosure, the aforementioned preset rule is associated with at least one of the following: transaction object, tax identification code of the transaction object, invoice number, transaction time, transaction amount, and transaction channel.

In an embodiment of the present disclosure, the above-mentioned server updates the default rules according to the feedback information.

In an embodiment of the present disclosure, the above-mentioned server updates the prediction model according to the feedback information.

In an embodiment of the present disclosure, the above server outputs the account statement to the electronic device in response to receiving the query request from the electronic device.

An automatic billing method based on user feedback of the present disclosure, comprising: accessing a financial account corresponding to a terminal device by a server, and sending a push message corresponding to a transaction event in response to detecting a transaction event corresponding to the financial account to the terminal device; the terminal device sends a feedback message corresponding to the push message to the server; and the server determines the classification of the transaction event according to the feedback message in response to receiving the feedback message, and updates the corresponding financial event according to the transaction event and classification. Account statement for the account.

Based on the above, the server of the present disclosure can send a push message to the terminal device when a transaction event occurs to request the terminal device to feedback the classification of the transaction event, and generate a corresponding report. The user of the terminal device can query the server for reports through the terminal device.

In order to make the content of the present disclosure easier to understand, the following specific embodiments are taken as examples in which the present disclosure can indeed be implemented. In addition, wherever possible, elements/components/steps using the same reference numerals in the drawings and embodiments represent the same or similar parts.

FIG. 1 is a schematic diagram of an automatic billing system 10 according to an embodiment of the disclosure. The automatic billing system 10 may include a server 100 and a terminal device 200 . The server 100 belongs to, for example, a bank or a financial institution, and the terminal device 200 belongs to, for example, a customer of the bank or financial institution. The server 100 is communicatively connected to the terminal device 200 .

The server 100 may include a processing unit (such as a processor but not limited thereto), a communication unit (such as a mobile communication chip, a Bluetooth chip or a WiFi chip, etc. but not limited thereto) and a storage unit (such as a removable random access memory, flash memory or hard disk, etc., but not limited to) and other necessary components for running the server 100 . The terminal device 200 may include but not limited to client, desktop computer, notebook computer, network computer, workstation, personal digital assistant (personal digital assistant, PDA), personal computer (personal computer, PC), tablet computer or Telephone equipment, etc. The terminal device 200 may include transceiver circuits, analog-to-digital (A/D)/digital-to-analog (D/A) converters, processing circuits, memory circuits, and one or more antenna units, which are necessary components for operating the terminal device 200 .

The server 100 can access the financial account corresponding to the user of the terminal device 200 . For example, the server 100 can communicate with the banking system to access information related to the user's financial account. When a transaction event occurs in the user's financial account, the server 100 can detect the transaction event from the banking system. For example, a banking system may record credit card swipes associated with a user's financial account. After the user conducts the credit card transaction, the server 100 can obtain information related to the credit card transaction by accessing the financial account. As another example, a banking system may record securities transactions associated with a user's financial account. After the user conducts the securities transaction, the server 100 can obtain information related to the securities transaction by accessing the financial account.

The server 100 may transmit a push message corresponding to the transaction event to the terminal device 200 in response to detecting the transaction event corresponding to the financial account. The push message can instruct the user of the terminal device 200 to judge the classification of the transaction event, and send a feedback message corresponding to the push message to the server 100 through the terminal device 200, wherein the feedback message can include the user's response to the transaction event. classification results. For example, the user can operate the terminal device 200 according to the instruction of the push message to classify the transaction event as "restaurant consumption" and generate a classification result corresponding to "restaurant consumption". The user can operate the terminal device 200 to send the feedback message including the classification result to the server 100 .

After receiving the feedback information from the terminal device 200, the server 100 can determine the classification of the transaction event according to the feedback information in response to receiving the feedback information, and generate or update the account statement of the financial account according to the transaction event and the classification of the transaction event (account report), where the account report may include information such as transaction classification, transaction object, transaction amount, transaction method (such as: financial account debit, credit card, or remittance) or transaction date. For example, if the feedback message instructs the user to classify the transaction event as "restaurant consumption", the server 100 may classify the transaction event as "restaurant consumption" according to the feedback message. Next, the server 100 may generate or update an account report corresponding to the user's financial account according to the transaction event and the classification of the transaction event, wherein the account report may indicate that the classification of the transaction event is "restaurant consumption".

If the server 100 does not receive the feedback message from the terminal device 200, it means that the user has not manually classified the transaction event, or the communication between the terminal device 200 and the server 100 is terminated. In response to not receiving the feedback message, the server 100 may determine the classification of the transaction event according to a preset rule or prediction model pre-stored in the server 100, and then generate or update an account statement.

Preset rules can be associated with the transaction object, the transaction object's tax identification number (tax identification number, TIN), invoice number (invoice number), transaction time, transaction amount or transaction channel (for example: credit card machine, automatic teller machine or network banks, etc.), etc., but the present disclosure is not limited thereto. For example, if the transaction event corresponds to the tax identification code of a specific restaurant, the server 100 may classify the transaction event as "restaurant consumption" according to the tax identification code. Predictive models may include machine learning models such as long short-term memory (LSTM) models. The server 100 can input the relevant information of the transaction event into the prediction model to generate the classification result of the transaction event. In one embodiment, the server 100 can access the banking system to obtain historical transaction records corresponding to financial accounts, and train a prediction model according to the historical transaction records.

After generating or updating the account statement of the user's financial account, the server 100 may store the account statement. If the user wants to inquire about the account statement, the user can operate an electronic device (for example: the terminal device 200 ) to send an inquiry request to the server 100 . The server 100 may output the account statement to the electronic device in response to receiving the query request from the electronic device.

In one embodiment, the push message may include suggested categories of transaction events provided by the server 100 . The user can refer to the suggested classification to determine the classification of the transaction event. Specifically, the server 100 can determine the classification of the transaction event according to a preset rule or a prediction model to generate a suggested classification, and transmit the suggested classification to the terminal device 200 through a push message. If the user agrees with the suggested category provided by the server 100 , the user can operate the terminal device 200 to send a feedback message including an approval instruction to the server 100 . The server 100 may set the classification of the transaction event as a suggested classification according to the approval instruction.

In one embodiment, the server 100 can update the preset rules according to the feedback information from the terminal device 200 . For example, if the feedback message indicates that the transaction event performed at a specific merchant is classified as "restaurant consumption", the server 100 may update the default rule. The updated default rule may classify any transaction event with the specific merchant as a "restaurant purchase". In one embodiment, the server 100 can update the prediction model according to the feedback information from the terminal device 200 . Specifically, the server 100 can use the feedback information as new training data to retrain the prediction model.

FIG. 2 shows a flowchart of an automatic billing method according to an embodiment of the present disclosure, wherein the automatic billing method can be implemented by the automatic billing system 100 shown in FIG. 1 . In step S201, the server accesses the financial account corresponding to the terminal device, and sends a push message corresponding to the transaction event to the terminal device in response to detecting the transaction event corresponding to the financial account. In step S202, the terminal device sends a feedback message corresponding to the push message to the server. In step S203, the server determines the classification of the transaction event according to the feedback message in response to receiving the feedback message, and updates the account statement corresponding to the financial account according to the transaction event and the classification.

To sum up, the server of the present disclosure can send a push message to the terminal device when a transaction event occurs to request the terminal device to report back the classification of the transaction event, and generate a corresponding report. If the terminal device does not send a feedback message to the server, the server can determine the classification of the transaction event according to a preset rule or a prediction model. The server can update the default rules or prediction models according to the feedback information. Therefore, the accuracy of the server's self-determined classification of transaction events will increase with the increase in the number of feedback messages from the terminal device. The push message may include suggested classifications generated according to the judgment of the server. After the terminal device receives the push message, the user of the terminal device can refer to the suggested classification to determine the classification of the transaction event, so as to feed back more accurate feedback information through the terminal device. Users can query the server for reports through terminal devices.

10: Automatic accounting system 100: server 200: terminal device S201, S202, S203: steps

FIG. 1 is a schematic diagram of an automatic billing system according to an embodiment of the disclosure. FIG. 2 is a flow chart of an automatic billing method according to an embodiment of the disclosure.

10: Automatic accounting system

100: server

200: terminal device

Claims (12)

An automatic billing system based on user feedback, including: terminal device; and a server, communicatively connected to said terminal device, wherein The server accesses a financial account corresponding to the terminal device, and in response to detecting a transaction event corresponding to the financial account, transmits a push message corresponding to the transaction event to the terminal device, wherein The terminal device sends a feedback message corresponding to the push message to the server, wherein The server determines the classification of the transaction event according to the feedback message in response to receiving the feedback message, and updates an account statement corresponding to the financial account according to the transaction event and the classification. The automatic billing system according to claim 1, wherein the server determines the classification of the transaction event according to a preset rule in response to not receiving the feedback message. The automatic billing system according to claim 1, wherein the server determines the suggested category of the transaction event according to a preset rule, wherein the push message includes the suggested category. The automatic accounting system according to claim 3, wherein the feedback message includes an approval instruction, wherein the server determines the classification of the transaction event as the suggested classification according to the approval instruction. The automatic accounting system as described in claim 1, wherein the server obtains historical transaction records corresponding to the financial account, and generates a forecast model corresponding to the financial account based on the historical transaction records, wherein the server The controller determines the classification of the transaction event according to the predictive model in response to not receiving the feedback message. The automatic accounting system as described in claim 1, wherein the server obtains historical transaction records corresponding to the financial account, and generates a forecast model corresponding to the financial account based on the historical transaction records, wherein the server A device determines a suggested classification of the transaction event according to the predictive model, wherein the push message includes the suggested classification. The automatic billing system according to claim 5, wherein the predictive model is a machine learning model, wherein the machine learning model includes a long short-term memory model. The automatic accounting system according to Claim 2, wherein the preset rule is associated with at least one of the following: transaction object, tax identification code of the transaction object, invoice number, transaction time, transaction amount, and transaction channel. The automatic billing system as claimed in claim 2, wherein the server updates the preset rule according to the feedback message. The automatic billing system according to claim 5, wherein the server updates the prediction model according to the feedback information. The automatic billing system according to claim 1, wherein the server outputs the account statement to the electronic device in response to receiving an inquiry request from the electronic device. An automatic billing method based on user feedback, comprising: accessing, by the server, a financial account corresponding to the terminal device, and in response to detecting a transaction event corresponding to the financial account, sending a push message corresponding to the transaction event to the terminal device; sending a feedback message corresponding to the push message to the server by the terminal device; and The server determines the classification of the transaction event according to the feedback message in response to receiving the feedback message, and updates an account statement corresponding to the financial account according to the transaction event and the classification.

Images