Order creation method, order creation device and computer readable storage medium

Abstract

The invention provides an order creation method, an order creation device and a computer readable storage medium, and the method comprises the steps: obtaining the mail information of a new mail after the new mail is received; performing order word information extraction on the mail information by using a first language model, and judging whether the mail information contains the order word information or not; if yes, using a second language model to extract order information in the mail information, the order information comprising at least one of the following: a customer name, a commodity name, an address, a commodity number, and a delivery date; and connecting with each production system through a communication interface, verifying the order information by using an auditing model, judging whether the order information meets the production requirements of each production system, if so, creating an order, and sending an order processing result to a customer. According to the invention, manual auditing steps can be reduced, the work handling capacity of auditing personnel is reduced, and the order creating time is shortened.

Description

Technical Field

[0001] The present invention relates to the field of communication technology, and in particular to an order creation method, an order creation device and a computer-readable storage medium. Background Art

[0002] Today, the process of creating production orders in enterprises involves multiple manual processing steps. For example, customers send order requests via email, customer service staff read the email, query and verify inventory information, and then manually enter the order information into the order system. Finally, manual review and confirmation are required. When manually processing a large number of orders, it is easy to cause problems such as omissions of order information or input errors in the order, which affects the efficiency of order production and customer satisfaction.

[0003] An existing method for generating feedback work orders is to generate a feedback email by receiving feedback information sent by customers, and send the feedback email to a target feedback mailbox; when the target feedback mailbox receives the feedback email, the feedback email is verified and cleaned using a pre-built keyword feature extraction algorithm to obtain the target field of the cleaned feedback email; the target field of the feedback email is extracted, and the target field is remotely sent and entered into the work order system. This method does not connect to each production system to verify the order, and manual verification is required during subsequent production, which increases the verification time and is prone to verification errors. Summary of the invention

[0004] The first object of the present invention is to provide an order creation method that reduces manual review steps.

[0005] The second object of the present invention is to provide an order creation device for implementing the above-mentioned order creation method.

[0006] A third object of the present invention is to provide a computer-readable storage medium to which the above-mentioned order creation method is applied.

[0007] To achieve the first purpose of the present invention, the order creation method provided by the present invention includes: monitoring a mailbox, and after receiving a new email, obtaining email information of the new email; using a first language model to extract order word information from the email information, and determining whether the email information contains order word information; if so, using a second language model to extract order information from the email information, the order information including at least one of the following: customer name, product name, address, product quantity, and delivery date; connecting with each production system through a communication interface, using an audit model to verify the order information, determining whether the order information meets the production requirements of each production system, and if so, creating an order and sending the order processing result to the customer.

[0008] As can be seen from the above scheme, after extracting the order information from the order, the various production systems are connected and the order information is verified through the audit model. For example, the address is verified to prevent delivery to restricted areas; the inventory of goods is verified to determine whether delivery can be completed on the order delivery date based on the production cycle; and the order information is checked for high-risk keyword information, etc. The order is verified through the audit model, thereby reducing the steps of manual review, while ensuring the accuracy, reducing the time for verification, increasing the speed of verification, and shortening the time to create an order.

[0009] In a further solution, the production system includes at least one of the following: an inventory management system, a production scheduling system, and an address verification system.

[0010] In a further solution, the step of using the audit model to verify the order information includes: obtaining production information related to the order information from each production system, inputting the production information and the order information into the audit model, and outputting verification result information.

[0011] It can be seen that the inventory information of the inventory management system is obtained through the inventory management system, so as to verify the inventory of the goods, and judge whether the delivery can be completed on the order delivery date according to the production cycle; the order scheduling time is obtained through the production dispatch system to judge whether the delivery can be completed on the order delivery date; the address is verified through the address verification system to prevent delivery to restricted areas. The production information is obtained from each production system to obtain the verification result, so as to generate the verified order, reduce the manual review steps, and reduce the workload of the auditors.

[0012] In a further solution, after extracting the order information from the email message using the second language model, the following is further performed: integrating the order information into a preset template.

[0013] It can be seen that by setting up a preset template, order information can be integrated, and when the audit model is audited later, the audit model can quickly obtain comparison information.

[0014] In a further solution, after receiving a new email, the following steps are performed: determining whether the sender information of the new email is in the whitelist; if so, using the language model to extract order term information of potential orders from the email information; if the sender information of the new email is not in the whitelist, marking the email as a manually reviewed email.

[0015] It can be seen that by setting up a whitelist, interference from other spam emails can be prevented.

[0016] In a further solution, the first language model is used to extract order word information from the email information, and the email information is cleaned.

[0017] It can be seen that cleaning email information and removing useless information in email information can make the email information more in line with the needs of the language model and be more conducive to the language model's extraction of order features.

[0018] In a further solution, after outputting the verification result information, the following is further performed: training the audit model using the order information, production information and verification result information.

[0019] This shows that by training the audit model with new data, the audit model's predictions become more accurate.

[0020] In order to achieve the second purpose, the order creation device provided by the present invention includes a processing module and a memory, the memory stores a computer program, and the processing module implements the above-mentioned order creation method when executing the computer program.

[0021] In a further solution, the processing module includes an email receiving module, an information extraction module, an audit module and an order generation module, the information extraction module is connected to the email receiving module, the audit module is connected to the information extraction module, and the order generation module is connected to the audit module.

[0022] It can be seen that new emails are received through the email receiving module to obtain the email information of the new email; the language model can be used to extract order word information from the email information through the information extraction module, and the order information of the email information can be extracted; the order information can be reviewed through the review module; the order can be created through the order generation module, connected to each production system, and the order can be sent to each production system.

[0023] In order to achieve the third purpose, the present invention provides a computer-readable storage medium, on which a computer program is stored, characterized in that: when the computer program is executed by the test system, the above-mentioned order creation method is implemented. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 It is a system structure block diagram of an embodiment of an order creation device of the present invention.

[0025] Figure 2 It is a flow chart of an embodiment of the order creation method of the present invention.

[0026] The present invention is further described below in conjunction with the accompanying drawings and embodiments. DETAILED DESCRIPTION

[0027] The present invention extracts order information from the email information of new emails through a language model, then connects to each production system, verifies the order information through an audit model, and determines whether the order information meets the production requirements of each production system. If so, an order is created. The present invention performs auditing through an audit model, reducing the steps of manual auditing.

[0028] Order creation device embodiment: The order creation device in this embodiment includes a processing module 1 and a memory 2. The memory 2 stores a computer program. The processing module can execute the computer program obtained by the memory 2 to implement the order creation method. Among them, the processing module 1 includes an email receiving module 11, an information extraction module 12, an audit module 13 and an order generation module 14. The email receiving module 11 is used to receive new emails and obtain email information of new emails; the information extraction module 12 can use a language model to extract order word information from the email information and extract the order information from the email information; the audit module 13 can audit the order information; the order generation module 14 can create an order, connect to each production system, and send the order to each production system.

[0029] Order creation method implementation example: The order creation method of the present invention applies the order creation device of the above-mentioned order creation device embodiment. The mail receiving module 11 of the processing module first executes step S1 to monitor the mailbox to determine whether a new mail is received. After receiving the new mail, the mail receiving module also determines whether the sender information of the mail information is in the white list. The white list is a customer list. By setting the white list, the rest of the mails are filtered out to prevent interference from other spam mails. If the sender information of the new mail is not in the white list, it proves that the mail is not an order mail sent by the customer, and the mail is marked as a mail that needs manual review.

[0030] If the sender information of the new email is in the whitelist, the information extraction module 12 needs to clean the email information of the new email. Before the email information is input into the language model, it needs to be cleaned so that the email information is more in line with the requirements of the language model and is more convenient for the language model to extract features in the email information. Email information cleaning can remove useless information in the email information, such as punctuation marks, extra spaces, line breaks, etc., to improve data quality and availability and reduce the noise of the email information.

[0031] After cleaning the email information of the new email, the information extraction module 12 executes step S2 to extract order word information from the email information using the first language model. The new email sent by the customer may not be an order email, so before creating an order, it is necessary to extract order word information from the email information, and the order word information includes at least one of the following: "order", "purchase", "place an order". In this embodiment, the first language model is a natural language processing model.

[0032] After the information extraction module 12 uses the first language model to extract the order word information from the email information, step S3 is further executed to determine whether the email information contains the order word information. If the email information contains the order word information, it proves that the email information is the customer's order email, and then step S4 is executed to extract the order information in the email information using the second language model. The order information includes at least one of the following: customer name, product name, address, product quantity, and delivery date. In this embodiment, the second language model is a natural language processing model.

[0033] If the email information does not contain order word information, it proves that the email information is not the customer's order email, then step S8 is executed to mark the new email and mark the email as a manual review email, and then manually review the email.

[0034] After extracting the order information from the email information, the order information is integrated into the preset template. The preset template is a structured data template, so that the order information can be integrated. When the audit model is audited later, the audit model facilitates the analysis of the order information.

[0035] After extracting the order information from the email information, the order is not generated directly, and the order information needs to be reviewed. The review module 13 executes step S5, connects to each production system, and uses the review model to verify the order information. Among them, the production system includes at least one of the following: inventory management system, production scheduling system, address verification system. Obtain production information related to the order information from each production system, for example, obtain commodity inventory information consistent with the commodity name from the inventory management system. Input the production information and order information into the review model, and output the verification result information.

[0036] The audit model can verify the order information. The audit model verifies the order information as follows: Verify the content of the order information to determine whether the order information is complete. The content of the order information includes the delivery address, contact number, product information, quantity, price, etc.; obtain verification information from the verification system. The verification information is the number of digits in the phone number, postal code information, etc. to perform format verification on the order information to determine whether the format in the order information is correct. If the order format is correct, the output verification result information is passed; obtain high-risk keyword information from the risk identification system, screen the order information for high-risk keywords or sensitive information, and determine whether the order information does not contain high-risk keywords or words of sensitive information. If it does not contain high-risk keywords or words of sensitive information, the output verification result information is passed; obtain the shipping address of the restricted area from the address verification system. Determine whether the shipping information in the order information is a shipping address in a restricted area. If the shipping information in the order information is not a shipping address in a restricted area, the verification result information is output as passed. Obtain the inventory information of the product that is consistent with the product name from the inventory management system to determine whether the product inventory is sufficient. If the product inventory is sufficient, the verification result information is output as passed. If the product inventory is insufficient, further obtain the production rate of the product in the production system to determine whether the production can be completed within the order delivery cycle. If the production can be completed within the order delivery cycle, the verification result information is output as passed. Perform risk assessment on the order information, obtain the historical order information of the customer of the order information from the customer management system, and determine whether the order information is low risk, that is, use the historical order information to assess the return risk, credit risk, etc. of the order. If the order information is low risk, the verification result information is output as passed. Filter potential invalid orders at an early stage through the audit model to ensure that the created orders are safe and valid, and ensure the effectiveness of production when subsequent orders are produced.

[0037] Before the audit model is used, it needs to be trained. The staff will establish an audit database, which includes order information and production information, and mark the order information and production information to mark whether the order information can pass the audit. The audit database will be input into the audit model to train the audit model.

[0038] After the audit model verifies the order information, it outputs the verification result information. After obtaining the verification result information, step S6 is executed to determine whether the order information meets the production requirements of each production system, that is, whether the verification result information of the audit model is all passed. If the order information meets the production requirements of each production system, the order generation module 14 executes step S7 to create an order, send the order result to the customer, which can be sent to the customer by email, and also send the order result to each production system, and each production system can perform production operations.

[0039] After obtaining the verification result information, the audit model can be trained using the order information, production information, and verification result information. The audit model is trained with new data to make the prediction of the audit model more accurate.

[0040] If the order information does not meet the production requirements of each production system, step S8 is executed to mark the new email as a manual review email, and the email is subsequently manually reviewed.

[0041] In this embodiment, all steps from obtaining a new email from a customer to creating an order are automated, reducing manual intervention and improving the efficiency of the order processing chain. This embodiment connects to various production systems to ensure efficient and secure data flow. The order information is audited through the audit model to generate a verified order. The staff does not need to check the production system and create an order after meeting the requirements of each production system, which reduces the manual audit steps and the workload of the auditors.

[0042] Computer readable storage medium embodiment: The order creation method in the computer device described in the above embodiment can be stored in a computer readable storage medium in the form of a computer program. When the computer program is executed by the processing module, the steps of the order creation method embodiment in the above order creation device can be completed. The computer readable medium can be a computer readable signal medium or a computer readable storage medium. The computer readable storage medium can be, for example, but not limited to: an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer readable storage media (a non-exhaustive list) include: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above.

[0043] The above are only preferred embodiments of the present invention, but the design concept of the invention is not limited thereto. Without departing from the concept of the present invention, more other equivalent embodiments may be included. It is obvious to those skilled in the art that various changes, readjustments and substitutions can be made without departing from the protection scope of the present invention.

Claims

1. Order creation method, including: After receiving a new email, obtaining email information of the new email; Extracting order word information from the email information using the first language model, and determining whether the email information contains order word information; If yes, extract order information from the email message using the second language model, the order information including at least one of the following: customer name, product name, address, product quantity, and delivery date; Features: Connect with each production system through the communication interface, use the audit model to verify the order information, and determine whether the order information meets the production requirements of each production system. If so, create an order and send the order processing results to the customer.

2. The order creation method according to claim 1, characterized in that: The production system includes at least one of the following: an inventory management system, a production scheduling system, and an address verification system.

3. The order creation method according to claim 2, characterized in that: The steps to verify order information using the audit model include: Acquire production information related to order information from each production system, input the production information and the order information into the audit model, and output verification result information.

4. The order creation method according to any one of claims 1 to 3, characterized in that: After extracting the order information from the email using the second language model, the following is also performed: The order information is integrated into a preset template.

5. The order creation method according to any one of claims 1 to 3, characterized in that: After receiving new mail, also execute: Determine whether the sender information of the new email is in the whitelist, and if so, use the first language model to extract the order word information from the email information; If the sender information of the new email is not in the whitelist, the new email is marked as a manually reviewed email.

6. The order creation method according to claim 4, characterized in that: The first language model is used to extract order word information from the email information, and the following steps are further performed: The email information is cleaned.

7. The order creation method according to claim 3, characterized in that: After outputting the verification result information, the following is also executed: The audit model is trained using the order information, the production information and the verification result information.

8. Order creation device, characterized in that: It comprises a processing module and a memory, wherein the memory stores a computer program, and the processing module implements the order creation method according to any one of claims 1 to 7 when executing the computer program.

9. The order creation device according to claim 8, characterized in that: The processing module includes an email receiving module, an information extraction module, an audit module and an order generation module. The information extraction module is connected to the email receiving module, the audit module is connected to the information extraction module, and the order generation module is connected to the audit module.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by the test system, the order creation method according to any one of claims 1 to 7 is implemented.

Images