5G message intelligent session interaction method, storage medium and equipment

Abstract

The invention discloses a 5G message intelligent session interaction method, a storage medium and equipment. The method comprises the following steps: configuring basic information, an entry language, word slot information and a response mode of an interaction scene; a terminal user enters a chatboot interaction page, sends a 5G message to be interacted, converts the 5G message into a message text, matches an interaction scene, and selects the interaction scene; after the terminal user enters the interaction scene, the chatboot inserts a piece of cache data in the redis cache, traverses word slot positions in the word slot filling data list, generates clarified information according to word slot information of the corresponding word slot positions, sends the clarified information to the terminal user, and the terminal user gives an uplink reply, fills reply content into the corresponding word slot and writes the reply content into the redis cache; and generating a 5G response message according to the response mode configuration of the current interaction scene, and sending the 5G response message to the terminal user. According to the method, the development cost of chatboot is reduced, and the development period is shortened.

Description

Technical Field

[0001] The present invention relates to the field of 5G message session technology, and in particular to a 5G message intelligent session interaction method, storage medium and device. Background Art

[0002] 5G message intelligent conversation interaction technology is a major innovation in the field of communications. It not only integrates multiple communication methods such as 5G messages, AIM messages, multimedia text messages and text messages, but also provides users with a unified and feature-rich message service platform through the high speed, low latency and high reliability of the 5G network. As 5G technology continues to mature, 5G message intelligent conversation interaction technology has shown great application potential in many fields such as smart devices, enterprise communications, and public services. 5G message intelligent conversation interaction technology will promote the development of society in a more intelligent and digital direction and become an indispensable part of future communications. It will change the way people communicate, improve service efficiency, and bring revolutionary changes to all walks of life.

[0003] Chatbot is an application form that provides services to individual users through conversational interaction. As one of the core applications of 5G messaging, chatbot needs to provide real-time interactive experience for end users. 5G messaging chatbot can use natural language processing and machine learning technology to provide intelligent dialogue services, allowing users to communicate more naturally and intuitively with various services and applications through the 5G messaging platform. However, with the development of technology and the deepening of its application, the management and customized development of 5G messaging chatbots also face a series of challenges and difficulties.

[0004] First, as user needs diversify, 5G message chatbots need to constantly learn and adapt to new conversation scenarios and contexts, which requires developers to invest a lot of energy in training and optimizing algorithms; in addition, customized development needs to take into account the special needs of different industries and user groups, which not only increases development costs, but also increases the complexity of technical implementation. Summary of the invention

[0005] In response to the problems existing in the prior art, the present invention provides a 5G message intelligent conversation interaction method, storage medium and device, which realizes the 5G message interaction process by constructing an interaction scenario, and can extend the interaction scenario to 5G letter reading, video SMS, and text SMS sending scenarios, thereby improving the user's message interaction experience and the enterprise's 5G message delivery effect.

[0006] To achieve the above technical objectives, the present invention adopts the following technical solution: a 5G message intelligent conversation interaction method, which specifically includes the following steps:

[0007] Step 1: Configure the basic information of the interaction scene, entry language, word slot information and response method;

[0008] Step 2: After the end user enters the page for interacting with the chatbot according to the entry language, the chatbot sends a guidance message;

[0009] Step 3: The terminal user sends the 5G message to be interacted to the chatbot according to the guidance message, and the chatbot converts the 5G message to be interacted into a message text;

[0010] Step 4: The chatbot matches the interaction scenarios according to the message text and selects the interaction scenario according to the number of matched interaction scenarios;

[0011] Step 5: After the terminal user enters the interaction scene, the chatbot inserts a cache data into the redis cache according to the message to be interacted with, including: basic information of the interaction scene, word slot filling data list and interaction status;

[0012] Step 6: traverse the word slot positions in the word slot filling data list, generate a clarification message according to the word slot information of the corresponding word slot position, send it to the terminal user, the terminal user gives an upstream reply, fills the reply content into the corresponding word slot, and writes it into the redis cache;

[0013] Step 7: According to the response mode configuration of the current interaction scenario, generate a 5G response message and send it to the terminal user, mark the interaction status in the redis cache as completed, and exit the interaction scenario.

[0014] Furthermore, the basic information of the interaction scene configured in step 1 includes: the name, type and whether to enable word slot filling of the interaction scene; if the type of the interaction scene is question-and-answer type, word slot filling is not enabled; if the type of the interaction scene is dialogue type, word slot filling is enabled.

[0015] Furthermore, the word slot information configured in the interactive scene in step 1 includes: basic information of the word slot, verification rules of the word slot text, clarification messages of the word slot and the number of clarifications; the basic information of the word slot includes: word slot name, word slot collection method, whether to enable multi-input integration, whether to parse the word slot, voice recognition interface, image recognition interface, filling completion prompt and whether to transfer the word slot.

[0016] Furthermore, the response methods for the interactive scenarios configured in step 1 include: supporting a quick response method for inputting text, a fixed 5G message response method, and an interface call to dynamically generate a 5G message response method.

[0017] Furthermore, step 4 includes the following sub-steps:

[0018] Step 4.1, the chatbot performs BERT text extraction on the message text, and performs similarity analysis between the extracted text and the interaction scenario, and selects a list of interaction scenarios whose similarity exceeds the threshold;

[0019] Step 4.2: If the number of interaction scenarios in the list is 1, directly enter the corresponding interaction scenario; if the number of interaction scenarios in the list is greater than 1, the terminal user selects one or more interaction scenarios in the list.

[0020] Furthermore, if the number of interaction scenarios in the list is 0, there is no scenario interaction, and the corresponding action or instruction is executed through customized suggested actions or replies, or the knowledge base is called through FastGPT to generate a 5G response message for the AIGC large model.

[0021] Furthermore, if the type of the interaction scene is question-answer type, the word slot filling data list cannot be filled; if the type of the interaction scene is dialogue type, step 6 includes the following sub-steps:

[0022] Step 6.1, obtaining the word slot information of the first word slot position in the word slot filling data list, generating a clarification message, and sending it to the terminal user;

[0023] Step 6.2: For the clarification message, the terminal user gives a corresponding upstream reply, and the reply is parsed through the parsing interface, and the reply is judged whether it meets the filling condition according to the verification rule of the word slot text. If it meets the condition, the reply content is filled into the corresponding word slot and written into the redis cache, and step 6.4 is executed. Otherwise, step 6.3 is executed;

[0024] Step 6.3, resend the clarification message to the end user, and repeat step 6.2 until the maximum number of clarifications is reached, and then exit the interaction scenario;

[0025] Step 6.4: Move the word slot down one position and repeat steps 6.1-6.3 until the last word slot is filled.

[0026] Furthermore, if there are multiple interaction scenarios, the multi-input unification function of the word slot is turned on, a certain word slot position in the word slot filling data list corresponds to different word slot information according to different interaction scenarios, and different clarification messages are generated according to different word slot information.

[0027] Furthermore, the present invention also provides a computer-readable storage medium storing a computer program, wherein the computer program enables a computer to execute the 5G message intelligent conversation interaction method.

[0028] Furthermore, the present invention also provides an electronic device, comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the 5G message intelligent conversation interaction method is implemented.

[0029] Compared with the prior art, the present invention has the following beneficial effects:

[0030] (1) The 5G message intelligent conversation interaction method of the present invention can select the word slot filling method according to the interaction scenario by configuring the word slot information of the interaction scenario, and establish a cross-slot matching mechanism by whether to enable the word slot function of multiple inputs in one, thereby solving the interaction problems that are difficult to handle in the traditional methods of filling multiple slots with a single input and filling a single slot with multiple inputs, and effectively enriching the usage scenarios of chatbot;

[0031] (2) In order to meet the diverse requirements of users, the present invention solves this problem by pre-configuring the interaction scenario configuration method. The interaction scenario encapsulates a set of rules, data, 5G message templates and business interfaces for the interaction between users and chatbots, and can implement specific business functions through 5G message interaction. In addition, the interaction scenarios can be freely combined and data can be transferred, thereby realizing the processing of complex business logic and the rapid replication of simple business processing scenarios.

[0032] In addition to 5G message interaction, the present invention also extends the interaction scenarios to 5G letter reading, video SMS, and text SMS sending scenarios, improving the user's message interaction experience and the enterprise's 5G message delivery effect. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] Figure 1 This is a flow chart of the 5G message intelligent conversation interaction method of the present invention;

[0034] Figure 2 It is a schematic diagram of the configuration of the interactive scene in the present invention;

[0035] Figure 3 A schematic diagram of a strategy for filling multiple word slots in a single interaction in the present invention;

[0036] Figure 4 A schematic diagram of the strategy of filling word slots interactively multiple times in the present invention; DETAILED DESCRIPTION

[0037] The technical solution of the present invention is further explained below in conjunction with the accompanying drawings.

[0038] like Figure 1 The flowchart of the 5G message intelligent conversation interaction method of the present invention specifically includes the following steps:

[0039] Step 1. Configure the basic information, entry and exit words, word slot information, confirmation message and response method of the interaction scene. The present invention can effectively improve the efficiency of building 5G message conversation interaction applications and achieve minute-level online by pre-configuring the interaction scene. On the other hand, it can refine the complex interaction process into a set of scene interaction templates with high reusability. After the configuration is completed, confirm the configuration information of the interaction scene, support preview of the interaction scene in H5 mode on the web page, and the preview effect is consistent with the native 5G message interaction effect of the mobile phone. The operator can perform functional verification of the interaction scene. After the configuration is confirmed and the function is verified, click Submit to save the interaction scene. After saving, it will be in a disabled state. After clicking Enable, the interaction scene will take effect in real time.

[0040] like Figure 2 ,Configuring the basic information of the interaction scene includes: the name, type and whether to enable word slot filling of the interaction scene; if the type of the interaction scene is question-and-answer type, word slot filling is not enabled; if the type of the interaction scene is dialogue type, word slot filling is enabled.

[0041] Configure the entry and exit words of the interaction scene, where the entry word refers to the triggering words for entering the interaction scene, and the exit word refers to the words for exiting the interaction scene. For the entry and exit words, you can introduce the synonym replacement and text rewriting technology in NLP to generate several synonymous texts of the entry and exit words and add them to the original entry and exit word lists.

[0042] Configuring the word slot information of the interactive scenario includes: basic information of the word slot, basic information of the word slot includes: word slot name, word slot collection method, whether to enable multi-input integration, whether to parse the word slot, voice recognition interface, image recognition interface, filling completion prompt and whether to transfer the word slot; configuring the clarification message of the word slot and the number of clarifications. The clarification message of the word slot refers to how to ask the terminal user about the content of the word slot filling; configuring the verification rules of the word slot text, and the content that does not meet the verification rules will not be filled into the word slot.

[0043] Configure the confirmation message for the interactive scenario. You can choose not to confirm, which will directly trigger the scenario response. If you choose to display the confirmation message, the content of all the word slots filled in the dialogue will be displayed for confirmation. The end user can choose to submit after confirmation or return to modify a specific word slot. After modification, the modified content will be confirmed again.

[0044] The response methods for configuring interactive scenarios include: supporting quick response methods such as text input, fixed 5G message response methods, and dynamically generating 5G message response methods through interface calls. The interface call dynamically generates 5G message response methods. It is necessary to upload the definition file of the business system interface through a standardized interface definition file in advance, such as swag ui, and pass the system automatic call test. The dynamically generated 5G message must contain placeholder information, and the placeholder is dynamically replaced according to the system interface call result to generate the final 5G response message for delivery. The specific process is as follows:

[0045] A. Generate basic information of http interface by uploading pre-edited swag ui file, including: interface address, interface input parameter, interface output parameter, parameter example;

[0046] B. When creating a 5G message card, pre-fill the placeholder information in the card, supporting four types: ordinary text placeholder, object placeholder, loop placeholder and material placeholder;

[0047] C. Select the data source of the interface input parameter. You can select three types: word slot content, cache content or fixed value. Except for the word slot, you need to fill in the relevant key name as the input parameter of the interface call;

[0048] D. Perform placeholder matching and associate the output parameter content of the interface with the placeholder of the 5G message card. The object type placeholder is associated with the object type output parameter, and the loop type placeholder corresponds to the array type output parameter.

[0049] Step 2: After the terminal user enters the page for interacting with the chatbot according to the entry language, the chatbot sends a guidance message to guide the terminal user to interact.

[0050] Step 3. The terminal user sends the 5G message to be interacted to the chatbot according to the guidance message, and the chatbot converts the 5G message to be interacted into message text; specifically, the 5G message to be interacted includes: text messages, picture messages and voice messages. For picture messages, OCR recognition is performed to obtain text messages; for voice messages, ASR processing is performed to obtain text messages.

[0051] Step 4: The chatbot matches the interaction scenarios according to the message text and selects the interaction scenario according to the number of matched interaction scenarios. The specific steps include the following:

[0052] Step 4.1, the chatbot performs BERT text extraction on the message text, and performs similarity analysis between the extracted text and the interaction scenarios, and selects a list of interaction scenarios whose similarity exceeds the threshold; the advantage of BERT text extraction in similarity analysis lies in its powerful language understanding ability, which can deeply learn language patterns and contextual relationships, and gain insight into semantic differences. The vectors generated by BERT text extraction are rich in semantic information, which can accurately represent the meaning of the text and improve the accuracy and efficiency of similarity judgment. At the same time, the context-sensitive word vectors extracted by BERT text can capture the diversity of word meanings and context changes, and enhance the accuracy of text understanding.

[0053] Step 4.2: If the number of interaction scenarios in the list is 1, directly enter the corresponding interaction scenario; if the number of interaction scenarios in the list is greater than 1, the terminal user selects one or more interaction scenarios in the list.

[0054] In one technical solution of the present invention, if the number of interaction scenarios in the list is 0, there is no scenario interaction, and the corresponding action or instruction is executed through customized suggested actions or replies, or the knowledge base is called through FastGPT to generate a 5G response message for the AIGC large model.

[0055] Step 5: After the terminal user enters the interaction scene, the chatbot inserts a cache data into the redis cache according to the message to be interacted with, including: basic information of the interaction scene, a list of word slot filling data, and the interaction status.

[0056] Step 6: If the type of interaction scenario is question-answer type, the word slot filling data list cannot be filled; if the type of interaction scenario is dialogue type, traverse the word slot positions in the word slot filling data list, generate clarification messages according to the word slot information of the corresponding word slot position, and send them to the terminal user. The terminal user gives an upward reply, fills the reply content into the corresponding word slot, and writes it into the redis cache. The redis cache can record the user's effective input information in real time, and can quickly read the data used for message generation and cross-interaction scenario transmission from the redis memory; Figure 3 , including the following sub-steps:

[0057] Step 6.1, obtain the word slot information of the first word slot position in the word slot filling data list, generate a clarification message, and send it to the terminal user; if the clarification message is configured to be dynamically generated, call the interface according to the input parameter configuration, and replace the output parameter with the placeholder of the clarification message 5G message card to generate a complete clarification message, and send the clarification message to the terminal user by calling the MaaP platform interface;

[0058] Step 6.2: For the clarification message, the terminal user gives an uplink reply of text, voice, picture, video or geographic location type. The reply is parsed through the parsing interface, and the reply is judged whether it meets the filling condition according to the verification rule of the word slot text. If it meets the condition, the reply content is filled into the corresponding word slot and written into the redis cache, and step 6.4 is executed. Otherwise, step 6.3 is executed;

[0059] Step 6.3, resend the clarification message to the end user, and repeat step 6.2 until the maximum number of clarifications is reached, and then exit the interaction scenario;

[0060] Step 6.4: Move the word slot down one position and repeat steps 6.1-6.3 until the last word slot is filled.

[0061] In a technical solution of the present invention, a confirmation message can be generated based on the content of the word slots filled by the terminal user, the content currently filled in each word slot can be displayed, and a confirmation button and a separate modification button for each word slot can be provided through a floating menu.

[0062] In a technical solution of the present invention, if there are multiple interaction scenarios, such as Figure 4 , turn on the multi-input integration function of the word slot, a certain word slot position in the word slot filling data list corresponds to different word slot information according to different interaction scenarios, and different clarification messages are generated according to different word slot information. This strategy can support the list type of word slot filling method, make the data content of a single word slot richer, and expand the business coverage of the interaction scenario. By configuring the word slot information of the interaction scenario, the present invention can select the word slot filling method according to the interaction scenario, and establish a cross-slot matching mechanism by whether to turn on the word slot function of multi-input integration, which solves the interaction problem that is difficult to handle with the two traditional methods of single input filling multiple slots and multiple input filling single slots, effectively enriches the use scenarios of chatbot, improves the business processing capabilities of the interaction scenario, and can effectively handle complex actual business needs.

[0063] Step 7. Generate a 5G response message and send it to the terminal user according to the response mode configuration of the current interaction scenario. If the 5G response message is configured to be dynamically generated, call the interface according to the input parameter configuration, and replace the output parameter with the placeholder of the clarification message 5G message card to generate a complete 5G response message. Send the 5G response message to the terminal user by calling the MaaP platform interface, mark the interaction status in the redis cache as completed, and exit the interaction scenario.

[0064] The 5G message intelligent conversation interaction method of the present invention is also applicable to 5G message reading, video SMS, and SMS interaction when the mobile phone terminal is not compatible with 5G messages. It can be achieved by configuring the 5G message reading, video SMS, and SMS fallback message of the clarification message when configuring the word slot to clarify the message.

[0065] Example

[0066] The 5G message intelligent conversation interaction method of the present invention is applied to the self-service vehicle moving interaction scenario:

[0067] The basic information of a basic conversational interaction scenario is such as "self-service car moving, car moving, conversational, automatic outbound call notification of car moving through license plate, default grouping".

[0068] Configure the entry and exit words of the interactive scene, specifically configure the self-service moving interactive scene to enter self-service moving and parking space occupied, and configure the exit word to exit moving.

[0069] By calling the synonym replacement and text rewriting technology in NLP, the generated entry synonym texts are self-service towing, self-service car moving, and parking space used by others; the generated exit synonym texts are exit towing, exit car moving, terminate car moving, terminate car moving, terminate towing, and no car moving.

[0070] For the interactive scenario of self-service car moving, the user needs to provide at least three pieces of information, namely, location, reason and license plate number, to complete the notification of the car moving requirement. Therefore, in this embodiment, three corresponding word slots are configured.

[0071] The name of word slot one is moving location, and its alias is location. The collection method only supports geographic location and no parsing is performed. The clarification message is configured as a text 5G message with a floating menu. The message text is "Please click on my location to report the need to move the car". A floating menu is configured with the text "My location". The recommended operation after clicking is configured to obtain the user's geographic location. The number of clarifications is set to 3. No dictionary information is configured, and the word slot order is configured to 1.

[0072] The name of word slot 2 is Reason for Moving, and its alias is Reason. The collection method supports text and voice, and supports parsing and configuration of voice recognition interface. The clarification message is configured as a text 5G message with two floating menus. The message text is "Please select, enter or say the reason why you need to move your car." The text of floating menu 1 is "Occupying other people's parking spaces", and the time after clicking is configured as a recommended operation. The text of floating menu 2 is "Obstructing public transportation", and the time after clicking is configured as a recommended operation. The number of clarifications is set to 3. No dictionary information is configured, and the word slot order is configured to 2.

[0073] The name of slot three is license plate number, with an alias of license plate. The collection methods support text, voice, and pictures, and support parsing and configuring voice recognition and image recognition interfaces; the clarification message is configured as a text 5G message, and the message text is "Please provide the license plate number of the vehicle you need to move by inputting, speaking, or taking a photo"; the dictionary is configured as a custom dictionary and the verification rule is set as the regular expression / ^([津沪渝冀豫云辽黑湘皖鲁苏浙赣鄂桂晋陕吉闽贵粤川琼A-Z]{1}[a-zA-Z](([DF]((?![IO])[a-zA-Z0-9](?![IO]))[0-9]{4})|([0-9]{5}[DF]))|[津沪渝冀豫云辽黑湘皖鲁苏浙赣鄂桂晋陕吉闽贵粤川琼A-Z]{1}[A-Z]{1}[A-Z0-9]{4}[A-Z0-9挂学]{1})$ / which is used to verify the correct license plate number, and the slot order is configured as 3.

[0074] Configure to enable the display of the confirmation message, and the interactive scenario response is configured as an interface call to dynamically generate a 5G message response method. The response message configuration content is a text 5G message of "Your vehicle moving request has been submitted, and the processing result is ${result}", where ${result} is a placeholder with the identification name of result, and this placeholder is a string placeholder in this embodiment. The request interface includes three input parameters loc (location), reason (reason), and lp (license plate number), corresponding to the three slot values of this scenario, namely the vehicle moving location, the vehicle moving reason, and the license plate number. The two output parameters are code (return code) and msg (return information), where msg matches the result placeholder of the response message. The interface function is to query the mobile phone number reserved by the vehicle owner corresponding to the license plate number, and inform the vehicle owner of the vehicle moving location and reason through AI intelligent outbound calls, and return whether it is connected.

[0075] After the configuration confirmation and function verification are correct, click Submit and the self-service vehicle moving scenario is successfully saved. Click Enable and it will take effect immediately.

[0076] After the user with the mobile phone number 153xxxx9628 enters the chatbot interaction page, he sends a voice message such as "I want to move my car", which is parsed into the text content "I want to move my car" by the ASR engine. Then, after BERT analysis with the platform's entry language, the highest similarity is 84.13% with the entry language "move the car", which is higher than the preset matching threshold of 80%. Therefore, the number of interaction scenarios matched is 1, and the matching scenario is self-service moving. The user enters the self-service moving scene, and at the same time, the user data cache of this interaction is initialized in the redis cache as a Hash type data, with the key 153xxxx9628_f6448dd8d1ea, indicating that the data is the cache data of the user 153xxxx9628 in the interaction scene numbered f6448dd8d1ea. The corresponding value of the Hash data is {pos:1,loc:null,reason:null,lp:null}, indicating that the current clarification word slot position is 1, and the three word slots loc, reason and lp are not assigned values. Since the maximum number of clarifications for slot 1 is less than 3 times, a clarification message about the location of the moving car in slot 1 is sent to the mobile phone number 153xxxx9628. After receiving the message, the user clicks the floating menu "My Location" and reports the current location information as {latitude:32.026391, longitude:118.787639, label:"Tongfu Building, No. 501, Zhongshan South Road, Qinhuai District"}. Since slot 1 is not configured with dictionary information, this information will be filled into slot 1, and the updated Hash data value is {pos:2, loc:"Tongfu Building, No. 501, Zhongshan South Road, Qinhuai District", reason:null, lp:null}. Since the maximum number of clarifications for slot 2 is less than 3 times, a clarification message about the reason for moving the car in slot 2 is sent to 153xxxx9628. After receiving the message, the user sends a reply through voice. After ASR recognition, the reason for moving the car is "his car blocked my car and my car can't get out". Since the word slot 2 is not configured with dictionary information, the information will be filled into the word slot 2, and the updated Hash data value is {pos:3,loc:"Tongfu Building, No. 501, Zhongshan South Road, Qinhuai District",reason:"His car blocked my car and my car couldn't get out",lp:null}. Since the current maximum number of clarifications in the word slot 3 is not exceeded, a clarification message of the license plate number in the word slot 3 is sent to 153xxxx9628. After receiving the message, the user sends a reply to the license plate number information Su 5Q8Z1 through text. Because the word slot 3 is configured with a custom dictionary and uses regular expression verification, Su 5Q8Z1 does not meet the regular matching rules after verification, and the information cannot be filled into the word slot 3.The current clarification times are 1, which does not reach the maximum clarification times of word slot 3. The word slot 3 clarification message is sent again to 153xxxx9628. After receiving the message this time, the user chooses to take a photo and upload it. The license plate number obtained by the user through the license plate number recognition interface is Su xxxxxx. After verification, it meets the regular expression rules, so Su xxxxxx is filled into word slot 3. The updated Hash data value is {pos:3,loc:"Tongfu Building, No. 501, Zhongshan South Road, Qinhuai District",reason:"His car blocked my car and my car can't get out",lp:"Su xxxxxx"}. Because the confirmation message is enabled in this interaction scenario, and the three word slots are currently filled, a confirmation message is sent to the user number. The message content is "Please confirm whether the input information is correct / nMoving location: Tongfu Building, No. 501, Zhongshan South Road, Qinhuai District / nMoving reason: His car blocked my car and my car can't get out / nLicense plate number: Su xxxxxx". Because the moving location is not accurate enough, the user chooses to modify the moving location slot content to parking space 16 in the ground parking lot of Tongfu Building, No. 501, Zhongshan South Road, Qinhuai District. After sending it via text, the confirmation message is sent again. The user clicks the confirmation menu this time. The moving interface is called, and the result returned is that your moving request has been submitted, the processing result is successful, and it is sent to the user number 153xxxx9628, and this session interaction ends.

[0077] In one technical solution of the present invention, a computer-readable storage medium is also provided, storing a computer program, wherein the computer program enables a computer to execute the 5G message intelligent conversation interaction method.

[0078] In one technical solution of the present invention, an electronic device is also provided, including: a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the 5G message intelligent conversation interaction method is implemented.

[0079] In the embodiments disclosed in the present application, the computer storage medium can be a tangible medium that can contain or store a program for use by or in conjunction with an instruction execution system, device or equipment. The computer storage medium can include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices or equipment, or any suitable combination of the above. More specific examples of computer storage media can include electrical connections based on one or more lines, portable computer disks, hard disks, random access memories (RAM), read-only memories (ROM), erasable programmable read-only memories (EPROM or flash memory), optical fibers, portable compact disk read-only memories (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the above.

[0080] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed in this application can be implemented in electronic hardware or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0081] The above are only preferred embodiments of the present invention, and the protection scope of the present invention is not limited to the above embodiments. All technical solutions under the concept of the present invention belong to the protection scope of the present invention. It should be pointed out that for ordinary technicians in this technical field, some improvements and modifications without departing from the principle of the present invention should be regarded as the protection scope of the present invention.

Claims

1. A 5G message intelligent conversation interaction method, characterized in that: The specific steps include: Step 1: Configure the basic information of the interaction scene, entry language, word slot information and response method; Step 2: After the end user enters the page for interacting with the chatbot according to the entry language, the chatbot sends a guidance message; Step 3: The terminal user sends the 5G message to be interacted to the chatbot according to the guidance message, and the chatbot converts the 5G message to be interacted into a message text; Step 4: The chatbot matches the interaction scenarios according to the message text and selects the interaction scenario according to the number of matched interaction scenarios; Step 5: After the terminal user enters the interaction scene, the chatbot inserts a cache data into the redis cache according to the message to be interacted with, including: basic information of the interaction scene, word slot filling data list and interaction status; Step 6: traverse the word slot positions in the word slot filling data list, generate a clarification message according to the word slot information of the corresponding word slot position, send it to the terminal user, the terminal user gives an upstream reply, fills the reply content into the corresponding word slot, and writes it into the redis cache; Step 7: According to the response mode configuration of the current interaction scenario, generate a 5G response message and send it to the terminal user, mark the interaction status in the redis cache as completed, and exit the interaction scenario.

2. A 5G message intelligent conversation interaction method according to claim 1, characterized in that: The basic information of the interaction scene configured in step 1 includes: the name and type of the interaction scene and whether to enable word slot filling; if the type of the interaction scene is question-and-answer type, word slot filling is not enabled; if the type of the interaction scene is dialogue type, word slot filling is enabled.

3. A 5G message intelligent conversation interaction method according to claim 1, characterized in that: The word slot information configured in the interactive scene in step 1 includes: basic information of the word slot, verification rules of the word slot text, clarification message of the word slot and number of clarifications; the basic information of the word slot includes: word slot name, word slot collection method, whether to enable multi-input integration, whether to parse the word slot, voice recognition interface, image recognition interface, filling completion prompt and whether to transfer the word slot.

4. A 5G message intelligent conversation interaction method according to claim 1, characterized in that: The response methods for the interactive scenarios configured in step 1 include: supporting quick response methods such as text input, fixed 5G message response methods, and dynamically generating 5G message response methods through interface calls.

5. A 5G message intelligent conversation interaction method according to claim 1, characterized in that: Step 4 includes the following sub-steps: Step 4.1, the chatbot performs BERT text extraction on the message text, and performs similarity analysis between the extracted text and the interaction scenario, and selects a list of interaction scenarios whose similarity exceeds the threshold; Step 4.2: If the number of interaction scenarios in the list is 1, directly enter the corresponding interaction scenario; if the number of interaction scenarios in the list is greater than 1, the terminal user selects one or more interaction scenarios in the list.

6. A 5G message intelligent conversation interaction method according to claim 5, characterized in that: If the number of interaction scenarios in the list is 0, there is no scenario interaction. The corresponding actions or instructions are executed through custom suggested actions or replies, or the knowledge base is called through FastGPT to generate a 5G response message for the AIGC large model.

7. A 5G message intelligent conversation interaction method according to claim 3, characterized in that: If the type of the interaction scenario is question-answer type, the word slot filling data list cannot be filled; if the type of the interaction scenario is dialogue type, step 6 includes the following sub-steps: Step 6.1, obtaining the word slot information of the first word slot position in the word slot filling data list, generating a clarification message, and sending it to the terminal user; Step 6.2: For the clarification message, the terminal user gives a corresponding upstream reply, and the reply is parsed through the parsing interface, and the reply is judged whether it meets the filling condition according to the verification rule of the word slot text. If it meets the condition, the reply content is filled into the corresponding word slot and written into the redis cache, and step 6.4 is executed. Otherwise, step 6.3 is executed; Step 6.3, resend the clarification message to the end user, and repeat step 6.2 until the maximum number of clarifications is reached, and then exit the interaction scenario; Step 6.4: Move the word slot down one position and repeat steps 6.1-6.3 until the last word slot is filled.

8. A 5G message intelligent conversation interaction method according to claim 3, characterized in that: If there are multiple interaction scenarios, the multi-input unification function of the word slot is turned on. A certain word slot position in the word slot filling data list corresponds to different word slot information according to different interaction scenarios, and different clarification messages are generated according to different word slot information.

9. A computer-readable storage medium storing a computer program, characterized in that: The computer program enables the computer to execute the 5G message intelligent conversation interaction method as described in any one of claims 1-8.

10. An electronic device, characterized in that: include: A memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the 5G message intelligent conversation interaction method as described in any one of claims 1 to 8 is implemented.

Images