Man-machine interaction type steel ordering method and system based on natural language

Abstract

The invention discloses an interactive steel ordering method based on natural language processing, and the method comprises the steps: converting the voice information of various services of steel input by a user into text information, recognizing the ordering operation intention of the user according to the daily term of the ordering of the user in the text information, analyzing the key service elements of the ordering demands, and carrying out the processing of the key service elements. Then comparing the key business elements with the established corresponding relation between the key business elements of the user ordering daily term and the key business elements of the steel industry special term, and if the key business elements of the ordering demand are not completely the same as the key business elements of the steel industry special term, feeding back different element names to the user, and if the key business elements of the ordering demand are not completely the same as the key business elements of the steel industry special term. And if the key business elements are completely the same, calling the steel key business elements of the enterprise product knowledge base and comparing the key business elements with the analyzed key business elements of the ordering demand, if the key business elements are inconsistent, feeding back the names of the inconsistent elements to the user, and if the key business elements are consistent, generating an ordering receipt by an enterprise internal information system.

Description

Technical Field

[0001] The invention relates to the field of steel product trade ordering systems, in particular to a natural language-based human-computer interactive steel ordering method and system. Background Art

[0002] With the rapid evolution of AI technology, in the context of the digital transformation and upgrading of the steel industry, it is necessary to further explore new models and paths to speed up and increase business efficiency to support the company's business development. Although the existing internal information systems of steel companies have undergone years of construction and iteration, the maturity of the system architecture and the completeness of functions can basically meet the needs of internal business management and external customer service, but these systems basically use menu-based functions and page-based operations. Therefore, when business personnel operate the internal information system of the enterprise, they must first find the corresponding operation page through the function menu or function favorites, and then enter key business information such as customer name, product name, delivery date, order number, etc. to retrieve the business documents and types to be operated, and then follow the established operating steps to gradually complete the entry of documents to complete steel orders, resulting in low user operation efficiency and poor operating experience. Summary of the invention

[0003] The purpose of the present invention is to provide a natural language-based human-computer interactive steel ordering method and system that can improve user operating efficiency, so as to enhance user operating experience.

[0004] The interactive steel ordering method based on natural language processing of the present invention comprises the following steps: S1. Establishing the correspondence between the key business elements of the daily language used by users for ordering and the key business elements of the special terms of the steel industry in the natural language processing module; S2. Acquire the voice information of various steel business input by the user through the user operation module; S3, the natural language processing module converts the voice information input by users for various steel business into text information, identifies the user's ordering operation intention based on the user's daily ordering language in the text information, and analyzes the key business elements of the ordering demand; S4. Compare the key business elements of the parsed order demand with the key business elements of the established user ordering daily terms and the key business elements of the steel industry special terminology. If the key business elements of the parsed order demand are not exactly the same as the key business elements of the steel industry special terminology, the inconsistent element names will be fed back to the user. If they are exactly the same, the steel key business elements of the enterprise product knowledge base will be retrieved and compared with the key business elements of the parsed order demand. If they are inconsistent, the inconsistent element names will be fed back to the user. If they are consistent, the order document will be generated by the enterprise's internal information system.

[0005] As a preferred solution of the present invention, it also includes identifying the user's ordering business scenario and parsing the description of the ordering demand; Determine whether the description of the parsed order demand is an order for new steel attribute specifications or an order based on historical steel order records; if it is an order for new steel attribute specifications, create a new business operation instruction; if it is an order based on historical steel order records, use the old business operation instruction, and the enterprise's internal information system generates an order document based on the comparison result of step S4 and the business operation instruction.

[0006] The interactive steel ordering system based on natural language processing of the present invention comprises: The user operation module is used for users to input voice information of various steel business and send it to the natural language processing module; The natural language processing module is used to establish a correspondence between the key business elements of the user's daily ordering terms and the key business elements of the special terminology of the steel industry; convert voice information into text information, identify the user's ordering operation intention based on the user's daily ordering terms in the text information, and parse out the key business elements of the ordering needs; compare the parsed key business elements of the ordering needs with the established correspondence between the key business elements of the user's daily ordering terms and the key business elements of the special terminology of the steel industry; if the parsed key business elements of the ordering needs are not exactly the same as the key business elements of the special terminology of the steel industry, the inconsistent element names are fed back to the user; if they are exactly the same, the key business elements of steel in the enterprise product knowledge base are retrieved and compared with the key business elements of the parsed ordering needs; if they are inconsistent, the inconsistent element names are fed back to the user; if they are consistent, the order document is generated by the enterprise's internal information system.

[0007] As a preferred solution of the present invention, the natural language processing module further identifies the user's ordering business scenario and parses the description of the ordering requirements; Determine whether the description of the parsed order demand is an order for new steel attribute specifications or an order based on historical steel order records; if it is an order for new steel attribute specifications, create a new business operation instruction; if it is an order based on historical steel order records, use the old business operation instruction. The enterprise's internal information system generates an order document based on the consistent results and business operation instructions of the comparison between the key business elements of steel in the enterprise product knowledge base and the key business elements of the parsed order demand.

[0008] As a preferred solution of the present invention, the natural language processing module further includes a text conversion unit, and the text conversion unit is used to receive voice information sent by the user operation module and convert the voice information into text information.

[0009] As a preferred solution of the present invention, the natural language processing module also includes a semantic understanding unit, which is used to receive the text information sent by the text conversion unit, identify the user's ordering operation intention and ordering business scenario based on the user's daily ordering terms in the text information, and parse the key business elements of the ordering demand based on the ordering operation intention, and parse the description of the ordering demand based on the ordering business scenario.

[0010] As a preferred solution of the present invention, the natural language processing module further includes a value set mapping unit, which is used to establish a correspondence between key business elements of daily language used by users for ordering and key business elements of special terms for the steel industry.

[0011] As a preferred solution of the present invention, the natural language processing module also includes an information parsing unit, which receives the key business elements and description of the ordering requirements parsed by the semantic understanding unit, and the correspondence between the key business elements of the user's daily ordering terms and the key business elements of the special terminology in the steel industry; converts the key business elements of the ordering requirements parsed by the semantic understanding unit and the correspondence between the key business elements of the user's daily ordering terms and the key business elements of the special terminology in the steel industry into ordering operation intention parameters recognizable by the internal information system of the enterprise; and converts the description of the ordering requirements parsed by the semantic understanding unit into business operation parameters recognizable by the internal information system of the enterprise.

[0012] As a preferred solution of the present invention, the natural language processing module also includes a parameter memory unit, which receives and stores the order operation intention parameters and business operation parameters sent by the information parsing unit.

[0013] As a preferred solution of the present invention, the natural language processing module further includes a logic judgment unit for retrieving the order operation intention parameters and business operation parameters of the parameter memory unit; The key business elements of the order demand parsed from the order operation intention parameters are compared with the key business elements of the established user ordering daily terms and the key business elements of the special terminology of the steel industry. If the key business elements of the order demand parsed from the key business elements of the special terminology of the steel industry are not exactly the same, the different element names are fed back to the user. If they are exactly the same, the key business elements of steel in the enterprise product knowledge base are retrieved and compared with the key business elements of the parsed order demand. If they are inconsistent, the inconsistent element names are fed back to the user. If they are consistent, it is determined whether the description of the order demand parsed from the business operation parameters is an order for new steel attribute specifications or an order with reference to historical steel order records. If it is an order for new steel attribute specifications, a new business operation instruction is created. If it is an order with reference to historical steel order records, the old business operation instruction is used. The internal information system of the enterprise generates an order document based on the consistent results and business operation instructions of the comparison between the key business elements of steel in the enterprise product knowledge base and the key business elements of the parsed order demand.

[0014] The interactive steel ordering method and system based on natural language processing described in the present invention converts the voice information of various steel business input by the user into text information, identifies the user's ordering operation intention according to the user's daily ordering terms in the text information, parses the key business elements of the ordering demand, and then compares the corresponding relationship between the key business elements of the user's daily ordering terms and the key business elements of the special terms of the steel industry. If the key business elements of the ordering demand parsed are not completely the same as the key business elements of the special terms of the steel industry, the different element names are fed back to the user. If they are completely the same, the steel key business elements of the enterprise product knowledge base are retrieved and compared with the key business elements of the ordering demand parsed. If they are inconsistent, the inconsistent element names are fed back to the user. If they are consistent, the order document is generated by the internal information system of the enterprise, and the business personnel do not need to operate the internal information system of the enterprise. They need to first find the corresponding operation page through the function menu or function favorites, and then enter the customer name, product name, delivery date, order number and other key business information to retrieve the business document and type to be operated, thereby improving the user's operation efficiency and greatly improving the user's operation experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 The block diagram of the interactive steel ordering system based on natural language processing of the present invention; Figure 2 The present invention is a flowchart of an interactive steel ordering method based on natural language processing. DETAILED DESCRIPTION

[0016] The following will be combined with the accompanying drawings to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0017] like Figure 1 As shown, an interactive steel ordering system based on natural language processing includes a user operation module and a natural language processing module. The user operation module is used for users to input voice information of various steel businesses and send it to the natural language processing module. It is also used for users to input text information of various steel businesses. It is also used to receive the key business elements of the ordering requirements parsed by the natural language processing module and the key business elements of the steel industry special terms. The element names are also used for the key business elements of the steel in the enterprise product knowledge base and the key business elements of the ordering requirements that are inconsistent. It is also used to receive the order documents generated by the internal information system of the enterprise. It can provide users with an interface for specific business operations, which can be an independent system application terminal or a local application function of the internal information system of the enterprise.

[0018] The natural language processing module includes a text conversion unit, a semantic understanding unit, a value set mapping unit, an information parsing unit, a parameter memory unit, and a logic judgment unit.

[0019] The text conversion unit is used to receive the voice information sent by the user operation module and convert the voice information into text information.

[0020] The semantic understanding unit is used to receive the text information sent by the text conversion unit, identify the user's ordering operation intention and ordering business scenario based on the user's daily ordering language in the text information, and parse the key business elements of the ordering demand based on the ordering operation intention, and parse the description of the ordering demand based on the ordering business scenario. The user's operation intention specifically refers to the type of business operation that the user wants to perform. In this embodiment, it is the ordering operation intention. Other operation intentions include contract allocation, printing delivery orders, financial reconciliation, etc. The ordering business scenario determines whether the user directly describes the relevant steel product attributes to place an order or refers to the historical order records to place an order through the description of the ordering demand in the text information.

[0021] The value set mapping unit is used to establish the corresponding relationship between the key business elements of the daily terms used by users for ordering and the key business elements of the special terms of the steel industry.

[0022] The information parsing unit receives and parses the key business elements of ordering requirements and the corresponding relationship between the key business elements of daily user ordering terms and the key business elements of special terms in the steel industry, and converts them into ordering operation intention parameters that can be identified by the internal information system of the enterprise. The key business elements include steel varieties, grades, delivery dates, specifications, order quantities, etc. Receive the description of the parsed ordering requirements and convert them into business operation parameters that can be identified by the internal information system of the enterprise.

[0023] The parameter memory unit receives the order operation intention parameters and business operation parameters sent by the information analysis unit and stores them. In multiple rounds of dialogue for an order interaction, the parameter results of multiple analysis and conversion need to be memorized and stored.

[0024] The logic judgment unit retrieves the order operation intention parameters and business operation parameters of the parameter memory unit; compares the key business elements of the order demand parsed from the order operation intention parameters with the key business elements of the established user ordering daily terms and the key business elements of the steel industry special terms; if the key business elements of the order demand parsed from the key business elements of the steel industry special terms are not completely the same, the different element names are fed back to the user; if they are completely the same, the steel key business elements of the enterprise product knowledge base are retrieved and compared with the key business elements of the order demand parsed; if they are inconsistent, the inconsistent element names are fed back to the user; if they are consistent, it is determined whether the description of the order demand parsed from the business operation parameters is an order for new steel attribute specifications or an order with reference to historical steel order records; if it is an order for new steel attribute specifications, a new business operation instruction is created; if it is an order with reference to historical steel order records, the old business operation instruction is used, and the enterprise internal information system generates an order document based on the consistent results and business operation instructions of the comparison between the steel key business elements of the enterprise product knowledge base and the key business elements of the order demand parsed. The key steel business elements of the enterprise product knowledge base record the brand, specifications and other element information that the enterprise can produce. The enterprise's internal information system uses the API interface and the logic judgment unit parameters to copy, modify and add order cards. The order documents generated by the enterprise's internal information system are fed back to the user operation module through the natural language processing module, so that users can know the current process progress and business document number.

[0025] An interactive steel ordering method based on natural language processing, such as Figure 2 As shown, the following steps are included: S1. Establish a correspondence between the key business elements of the daily language used by users for ordering and the key business elements of the special terms in the steel industry in the natural language processing module.

[0026] S2. Acquire the voice information of various steel business input by the user through the user operation module.

[0027] S3. The natural language processing module converts the voice information input by users for various steel businesses into text information, identifies the user's ordering intention based on the user's daily ordering terms in the text information, and analyzes the key business elements of the ordering needs.

[0028] S4. Compare the key business elements of the parsed order demand with the key business elements of the established user ordering daily terms and the key business elements of the steel industry special terminology. If the key business elements of the parsed order demand are not exactly the same as the key business elements of the steel industry special terminology, the inconsistent element names will be fed back to the user. If they are exactly the same, the steel key business elements of the enterprise product knowledge base will be retrieved and compared with the key business elements of the parsed order demand. If they are inconsistent, the inconsistent element names will be fed back to the user. If they are consistent, the order document will be generated by the enterprise's internal information system.

[0029] It also includes identifying the user's ordering business scenario and parsing the description of the ordering requirements, determining whether the parsed description of the ordering requirements is an order for new steel attribute specifications or an order based on historical steel ordering records; if it is an order for new steel attribute specifications, creating a new business operation instruction; if it is an order based on historical steel ordering records, using the old business operation instruction, and the enterprise's internal information system generates an order document based on the comparison result of step S4 and the business operation instruction.

[0030] This method requires business personnel to operate the company's internal information system. They must first find the corresponding operation page through the function menu or function favorites, and then enter key business information such as customer name, product name, delivery date, order number, etc. to retrieve the business documents and types to be operated. Therefore, it can improve user operation efficiency and greatly enhance the user operation experience.

[0031] The above embodiments are only used to illustrate the detailed scheme of the present invention, and the present invention is not limited to the above detailed scheme, that is, it does not mean that the present invention must rely on the above detailed scheme to be implemented. Those skilled in the art should understand that any improvement of the present invention, equivalent replacement of the raw materials of the product of the present invention, addition of auxiliary components, selection of specific methods, etc., are all within the protection scope and disclosure scope of the present invention.

Claims

1. A human-computer interactive steel ordering method based on natural language, characterized in that: include: S1. Establishing the correspondence between the key business elements of the daily language used by users for ordering and the key business elements of the special terms of the steel industry in the natural language processing module; S2. Acquire the voice information of various steel business input by the user through the user operation module; S3, the natural language processing module converts the voice information input by users for various steel business into text information, identifies the user's ordering operation intention based on the user's daily ordering language in the text information, and analyzes the key business elements of the ordering demand; S4. Compare the key business elements of the parsed order demand with the key business elements of the established user ordering daily terms and the key business elements of the steel industry special terminology. If the key business elements of the parsed order demand are not exactly the same as the key business elements of the steel industry special terminology, the inconsistent element names will be fed back to the user. If they are exactly the same, the steel key business elements of the enterprise product knowledge base will be retrieved and compared with the key business elements of the parsed order demand. If they are inconsistent, the inconsistent element names will be fed back to the user. If they are consistent, the order document will be generated by the enterprise's internal information system.

2. The human-computer interactive steel ordering method based on natural language according to claim 1, characterized in that: It also includes identifying the user's ordering business scenario and parsing the description of the ordering requirements; Determine whether the description of the parsed order demand is an order for new steel attribute specifications or an order based on historical steel order records; if it is an order for new steel attribute specifications, create a new business operation instruction; if it is an order based on historical steel order records, use the old business operation instruction, and the enterprise's internal information system generates an order document based on the comparison result of step S4 and the business operation instruction.

3. A human-computer interactive steel ordering system based on natural language, characterized in that: include: The user operation module is used for users to input voice information of various steel business and send it to the natural language processing module; The natural language processing module is used to establish a correspondence between the key business elements of the user's daily ordering terms and the key business elements of the special terminology of the steel industry; convert voice information into text information, identify the user's ordering operation intention based on the user's daily ordering terms in the text information, and parse out the key business elements of the ordering needs; compare the parsed key business elements of the ordering needs with the established correspondence between the key business elements of the user's daily ordering terms and the key business elements of the special terminology of the steel industry; if the parsed key business elements of the ordering needs are not exactly the same as the key business elements of the special terminology of the steel industry, the inconsistent element names are fed back to the user; if they are exactly the same, the key business elements of steel in the enterprise product knowledge base are retrieved and compared with the key business elements of the parsed ordering needs; if they are inconsistent, the inconsistent element names are fed back to the user; if they are consistent, the order document is generated by the enterprise's internal information system.

4. The natural language-based human-computer interactive steel ordering system according to claim 3 is characterized in that: The natural language processing module also identifies the user's ordering business scenario and parses the description of the ordering requirements; Determine whether the description of the parsed order demand is an order for new steel attribute specifications or an order based on historical steel order records; if it is an order for new steel attribute specifications, create a new business operation instruction; if it is an order based on historical steel order records, use the old business operation instruction. The enterprise's internal information system generates an order document based on the consistent results and business operation instructions of the comparison between the key business elements of steel in the enterprise product knowledge base and the key business elements of the parsed order demand.

5. The natural language-based human-computer interactive steel ordering system according to claim 4 is characterized in that: The natural language processing module also includes a text conversion unit, which is used to receive voice information sent by the user operation module and convert the voice information into text information.

6. The natural language-based human-computer interactive steel ordering system according to claim 5, characterized in that: The natural language processing module also includes a semantic understanding unit, which is used to receive text information sent by the text conversion unit, identify the user's ordering operation intention and ordering business scenario based on the user's daily ordering terms in the text information, parse out the key business elements of the ordering demand based on the ordering operation intention, and parse out the description of the ordering demand based on the ordering business scenario.

7. The natural language-based human-computer interactive steel ordering system according to claim 6, characterized in that: The natural language processing module further comprises a value set mapping unit, which is used to establish a correspondence between key business elements of daily language used by users for ordering and key business elements of special terms for the steel industry.

8. The natural language-based human-computer interactive steel ordering system according to claim 7, characterized in that: The natural language processing module also includes an information parsing unit, which receives the key business elements and description of the ordering requirements parsed by the semantic understanding unit, and the correspondence between the key business elements of the user's daily ordering terms and the key business elements of the special terminology in the steel industry; converts the key business elements of the ordering requirements parsed by the semantic understanding unit and the correspondence between the key business elements of the user's daily ordering terms and the key business elements of the special terminology in the steel industry into ordering operation intention parameters that can be recognized by the internal information system of the enterprise; and converts the description of the ordering requirements parsed by the semantic understanding unit into business operation parameters that can be recognized by the internal information system of the enterprise.

9. The natural language-based human-computer interactive steel ordering system according to claim 8, characterized in that: The natural language processing module also includes a parameter memory unit, which receives and stores the order operation intention parameters and business operation parameters sent by the information parsing unit.

10. The natural language-based human-computer interactive steel ordering system according to claim 9, characterized in that: The natural language processing module also includes a logic judgment unit, which retrieves the order operation intention parameters and business operation parameters from the parameter memory unit; The key business elements of the order demand parsed from the order operation intention parameters are compared with the key business elements of the established user ordering daily terms and the key business elements of the special terminology of the steel industry. If the key business elements of the order demand parsed from the key business elements of the special terminology of the steel industry are not exactly the same, the different element names are fed back to the user. If they are exactly the same, the key business elements of steel in the enterprise product knowledge base are retrieved and compared with the key business elements of the parsed order demand. If they are inconsistent, the inconsistent element names are fed back to the user. If they are consistent, it is determined whether the description of the order demand parsed from the business operation parameters is an order for new steel attribute specifications or an order with reference to historical steel order records. If it is an order for new steel attribute specifications, a new business operation instruction is created. If it is an order with reference to historical steel order records, the old business operation instruction is used. The internal information system of the enterprise generates an order document based on the consistent results and business operation instructions of the comparison between the key business elements of steel in the enterprise product knowledge base and the key business elements of the parsed order demand.

Images