Conversational digital employee man-machine interaction implementation method and system, medium and electronic equipment

Abstract

The invention relates to the technical field of large model interaction, in particular to a dialogue type digital employee man-machine interaction implementation method and system, a medium and electronic equipment. The scheme comprises the following steps: setting a memory button, an API (Application Program Interface) and an online data access interface for the man-machine interaction system; stipulating feedback human settings for the robot; setting corresponding open scene white display for different people; presetting an enhanced retrieval mode, and performing enhanced reply according to the retrieval information; feeding back directional suggestion information; key reminding information is set and stored in the memory button, and the robot is reminded in each reply and information feedback. According to the scheme, accurate and efficient matching type question answering is realized by constructing the question answering type system of the digital employees online and combining the dynamically adjusted memory buttons.

Description

Technical Field

[0001] The present invention relates to the field of large model interaction technology, and more specifically, to a method, system, medium and electronic device for realizing conversational digital employee human-computer interaction. Background Art

[0002] In the current field of large-model interaction, the method, significance and importance of studying conversational human-computer interaction are reflected in how to achieve efficient and intelligent communication between users and large computing models through natural language processing technology. This interaction method can not only improve the user experience and enable people to communicate with machines more naturally, but also has a far-reaching impact on promoting the development of artificial intelligence technology. With the continuous advancement of technology, conversational human-computer interaction has become an important bridge connecting human wisdom and machine intelligence. It can not only promote the rapid circulation of information and the sharing of knowledge, but also play a huge role in education, medical care, customer service and other fields, greatly improving service efficiency and quality. Therefore, exploring more intelligent and personalized conversational human-computer interaction methods and their system design is crucial to meeting the needs of modern society for efficient and convenient information services.

[0003] Prior to the technology of the present invention, the existing conversational human-computer interaction methods mainly adopted a system architecture based on modules such as speech recognition, natural language processing, dialogue management, and speech synthesis. These systems convert the user's voice input into text information, then use natural language processing technology for semantic understanding and intent recognition, and finally control the dialogue process and generate responses through the dialogue management module. However, there are often the following problems: understanding depth and context problems. Although large language models can generate fluent language, they usually lack the ability to deeply understand the context behind the question and the context of a long period of time. For example, in a multi-round dialogue, it may forget the previous information or not understand it well, resulting in incoherent answers. Knowledge update lag: The training data of large language models comes from a data set at a specific point in time, so their knowledge base will have a time lag. Over time, the model cannot automatically obtain new information. Accuracy and reliability issues of generated content: Generative AI may sometimes generate inaccurate, wrong, or harmful content. For example, it may generate false information, ambiguous answers, etc. Summary of the invention

[0004] In view of the above problems, the present invention proposes a method, system, medium and electronic device for realizing conversational digital employee human-computer interaction, which realizes accurate and efficient matching question and answer by building an online question-and-answer system of digital employees and combining dynamically adjusted memory buttons.

[0005] According to a first aspect of an embodiment of the present invention, a method for implementing a conversational digital employee human-computer interaction is provided.

[0006] In one or more embodiments, preferably, the method for implementing a conversational digital employee human-computer interaction includes:

[0007] Set up memory buttons, API interfaces and online data access interfaces for the human-computer interaction system;

[0008] Define feedback personas for the robot;

[0009] Set up corresponding opening remarks for different characters;

[0010] Pre-set enhanced search mode to provide enhanced responses based on search information;

[0011] Feedback targeted advice information;

[0012] Setting key reminder information to be stored in the memory button to remind the robot in each reply and information feedback;

[0013] After each digital employee feedback response, extract the proportion of declarative opinions and the proportion of problematic opinions, wherein the proportion of declarative opinions is the number of declarative sentences divided by the total number of sentences, and the proportion of problematic opinions is the number of problematic sentences divided by the total number of sentences. According to the needs, in each questioning process, according to the needs of the proportion of declarative opinions and the proportion of problematic opinions, increase the tendency, which includes increasing the proportion of declarative opinions or the proportion of problematic opinions;

[0014] After each digital employee feedback reply, the proportion of reply sentences containing new nouns is extracted. When the proportion of new nouns is less than a preset value, it is required to increase the proportion of reply sentences containing new nouns, wherein the proportion of reply sentences containing new nouns is the number of sentences that do not include the name in the given information divided by the total number of sentences.

[0015] In one or more embodiments, preferably, the human-computer interaction system is provided with a memory button, an API interface and an online data access interface, specifically including:

[0016] Set up a memory button to store the query information online;

[0017] Set up an API interface to extract question and answer information in corresponding professional fields for matching;

[0018] Set up an online data access interface to extract online data for control.

[0019] In one or more embodiments, preferably, the step of defining a feedback personality for the robot specifically includes:

[0020] The feedback format for pre-set characters includes roles, skills, and considerations;

[0021] A table with corresponding personas is pre-set for selecting personas;

[0022] Pre-set the output method of the character's feedback.

[0023] In one or more embodiments, preferably, the setting of corresponding opening remarks for different personas specifically includes:

[0024] Choose a pre-set opening statement based on your character settings;

[0025] When a robot query of a corresponding persona is conducted online, the opening remarks corresponding to the robot of the persona are retrieved.

[0026] In one or more embodiments, preferably, the pre-setting of the enhanced search mode and the performing of an enhanced reply according to the search information specifically include:

[0027] In the pre-set enhanced search mode, set the method and order of filtering information;

[0028] By pre-setting enhanced search parameters and rules, the retrieved information can be filtered and sorted.

[0029] In one or more embodiments, preferably, the feedback-oriented suggestion information specifically includes:

[0030] According to the error information fed back by the robot in each conversation, the direction of directional suggestions is clearly defined;

[0031] Provides targeted suggestions based on the rules set by the questioner in each conversation.

[0032] In one or more embodiments, preferably, the key reminder information is stored in the memory button, and the robot is reminded in each reply and information feedback, specifically including:

[0033] Every time when receiving reply information, the clear declarative opinions are automatically extracted, numbered, and stored in the memory button as clear opinions;

[0034] Number the stated clear points;

[0035] Obtain the input upper limit of each feedback reply, and calculate the remaining word count according to the first calculation formula;

[0036] Use the second calculation formula to calculate the viewpoint number of this time;

[0037] Calculate the updated current viewpoint set using the third calculation formula;

[0038] Calculate the viewpoint number each time using the fourth calculation formula according to the updated current viewpoint set until the fifth calculation formula is satisfied;

[0039] After the fifth calculation formula is satisfied, all the clear viewpoints are updated, and the viewpoint number of this time is calculated again using the second, third, and fourth calculation formulas;

[0040] In each reply, start with "What needs to be reminded" and give feedback one by one on the clear opinions corresponding to the opinion number of this time;

[0041] The first calculation formula is:

[0042] SX-YR=SY

[0043] Among them, SX is the input upper limit, YR is the number of words entered, and SY is the number of remaining words;

[0044] The second calculation formula is:

[0045] BC=Argmin(|SY-ΣGZ|)

[0046] SY-ΣGZ>0

[0047] BC∈JH

[0048] Where Argmin is a function for extracting the number of this viewpoint that minimizes |SY-ΣGZ|, ΣGZ is the total number of words in this viewpoint, BC is the number of this viewpoint, and JH is the current viewpoint set;

[0049] The third calculation formula is:

[0050] JHq=JH-BC

[0051] Among them, JHq is the updated current viewpoint set;

[0052] The fourth calculation formula is:

[0053] BCn=Argmin(|SY-ΣGZ|)

[0054] SY-ΣGZ>0

[0055] BCn∈JHq

[0056] Among them, BCn is the viewpoint number of each time;

[0057] The fifth calculation formula is:

[0058] JHq=kong

[0059] Among them, kong is the empty set.

[0060] According to a second aspect of an embodiment of the present invention, a conversational digital employee human-computer interaction implementation system is provided.

[0061] In one or more embodiments, preferably, the conversational digital employee human-computer interaction implementation system includes:

[0062] Structural building module, used to set up memory buttons, API interfaces and online data access interfaces for the human-computer interaction system;

[0063] A character generation module is used to define a feedback character for the robot;

[0064] The opening remarks module is used to set up corresponding opening remarks for different characters;

[0065] An enhanced search module, used to pre-set an enhanced search mode and provide an enhanced reply based on the search information;

[0066] A directional suggestion module, used to feedback directional suggestion information;

[0067] The key memory information learning module is used to set key reminder information to be stored in the memory button, and remind the robot in each reply and information feedback.

[0068] According to a third aspect of an embodiment of the present invention, there is provided a computer-readable storage medium on which computer program instructions are stored. When the computer program instructions are executed by a processor, the method as described in any one of the first aspect of the embodiment of the present invention is implemented.

[0069] According to a fourth aspect of an embodiment of the present invention, there is provided an electronic device, comprising a memory and a processor, wherein the memory is used to store one or more computer program instructions, wherein the one or more computer program instructions are executed by the processor to implement any one of the methods described in the first aspect of the embodiment of the present invention.

[0070] The technical solution provided by the embodiments of the present invention may have the following beneficial effects:

[0071] In the solution of the present invention, a clear digital employee question and answer system is combined to conduct online questions and answers according to the pre-set feedback personality.

[0072] In the solution of the present invention, the question and answer information supplement based on the dynamically adjusted memory button is combined to achieve efficient and accurate question and answer feedback.

[0073] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or understood by practicing the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures particularly pointed out in the written description, claims, and drawings.

[0074] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0075] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative work.

[0076] Figure 1 The present invention is a flowchart of a method for implementing a conversational digital employee human-computer interaction according to an embodiment of the present invention.

[0077] Figure 2 The present invention is a flowchart of setting a memory button, an API interface and an online data access interface for a human-computer interaction system in a method for realizing a conversational digital employee human-computer interaction according to an embodiment of the present invention.

[0078] Figure 3 It is a flowchart of defining feedback personality for a robot in a method for implementing a conversational digital employee human-machine interaction according to an embodiment of the present invention.

[0079] Figure 4 It is a flowchart of setting corresponding opening remarks for different personalities in a method for realizing a conversational digital employee human-computer interaction according to an embodiment of the present invention.

[0080] Figure 5 The present invention is a flowchart of presetting an enhanced search mode in a method for realizing a conversational digital employee human-computer interaction and providing an enhanced reply based on search information in an embodiment of the present invention.

[0081] Figure 6 The present invention is a flowchart of feedback-oriented suggestion information in a method for implementing a conversational digital employee human-computer interaction according to an embodiment of the present invention.

[0082] Figure 7 It is a flowchart of setting key reminder information in a method for implementing a conversational digital employee human-computer interaction according to an embodiment of the present invention, storing it in the memory button, and reminding the robot in each reply and information feedback.

[0083] Figure 8 It is a structural diagram of a conversational digital employee human-computer interaction implementation system according to an embodiment of the present invention.

[0084] Fig. 9 It is a structural diagram of an electronic device in one embodiment of the present invention. DETAILED DESCRIPTION

[0085] In some of the processes described in the specification and claims of the present invention and the above-mentioned figures, multiple operations that appear in a specific order are included, but it should be clearly understood that these operations may not be executed in the order in which they appear in this article or executed in parallel. The serial numbers of the operations, such as 101, 102, etc., are only used to distinguish different operations, and the serial numbers themselves do not represent any execution order. In addition, these processes may include more or fewer operations, and these operations may be executed in sequence or in parallel. It should be noted that the descriptions of "first", "second", etc. in this article are used to distinguish different messages, devices, modules, etc., do not represent the order of precedence, and do not limit the "first" and "second" to be different types.

[0086] The following will be combined with the drawings in the embodiments of the present invention 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 those skilled in the art without creative work are within the scope of protection of the present invention.

[0087] In the current field of large-model interaction, the method, significance and importance of studying conversational human-computer interaction are reflected in how to achieve efficient and intelligent communication between users and large computing models through natural language processing technology. This interaction method can not only improve the user experience and enable people to communicate with machines more naturally, but also has a far-reaching impact on promoting the development of artificial intelligence technology. With the continuous advancement of technology, conversational human-computer interaction has become an important bridge connecting human wisdom and machine intelligence. It can not only promote the rapid circulation of information and the sharing of knowledge, but also play a huge role in education, medical care, customer service and other fields, greatly improving service efficiency and quality. Therefore, exploring more intelligent and personalized conversational human-computer interaction methods and their system design is crucial to meeting the needs of modern society for efficient and convenient information services.

[0088] Prior to the technology of the present invention, the existing conversational human-computer interaction methods mainly adopted a system architecture based on modules such as speech recognition, natural language processing, dialogue management, and speech synthesis. These systems convert the user's voice input into text information, then use natural language processing technology for semantic understanding and intent recognition, and finally control the dialogue process and generate responses through the dialogue management module. However, there are often the following problems: understanding depth and context problems. Although large language models can generate fluent language, they usually lack the ability to deeply understand the context behind the question and the context of a long period of time. For example, in a multi-round dialogue, it may forget the previous information or not understand it well, resulting in incoherent answers. Knowledge update lag: The training data of large language models comes from a data set at a specific point in time, so their knowledge base will have a time lag. Over time, the model cannot automatically obtain new information. Accuracy and reliability issues of generated content: Generative AI may sometimes generate inaccurate, wrong, or harmful content. For example, it may generate false information, ambiguous answers, etc.

[0089] In the embodiment of the present invention, a method, system, medium and electronic device for realizing a conversational digital employee human-computer interaction are provided. The solution realizes accurate and efficient matching question and answer by building a question-and-answer system of digital employees online and combining dynamically adjusted memory buttons.

[0090] According to a first aspect of an embodiment of the present invention, a method for implementing a conversational digital employee human-computer interaction is provided.

[0091] Figure 1 The present invention is a flowchart of a method for implementing a conversational digital employee human-computer interaction according to an embodiment of the present invention.

[0092] In one or more embodiments, preferably, the method for implementing a conversational digital employee human-computer interaction includes:

[0093] S101, setting a memory button, an API interface and an online data access interface for a human-computer interaction system;

[0094] S102, defining a feedback personality for the robot;

[0095] S103, setting corresponding opening remarks for different characters;

[0096] S104, pre-setting an enhanced search mode, and providing an enhanced reply based on the search information;

[0097] S105, feedback of targeted suggestion information;

[0098] S106, setting key reminder information to be stored in the memory button, and reminding the robot in each reply and information feedback.

[0099] In the embodiment of the present invention, the core is to create a digital robot, log in to the button platform, click the button to generate the robot, enter the robot design page, create a persona for the robot, describe the identity and tasks of the robot in the persona and reply logic panel on the left, and write a plan. The robot's persona and reply logic define its basic persona, which will continue to affect its reply effect in all conversations. In order to achieve the desired effect, the role of the model is specified in the persona and reply logic, the language style of the reply is designed, the range of the model's answer is limited, etc., so that the conversation is more in line with user expectations.

[0100] In addition, in order to make the robot's feedback better meet the needs of asking questions, after each digital employee's feedback, the proportion of declarative opinions and the proportion of problematic opinions are extracted, wherein the proportion of declarative opinions is the number of declarative sentences divided by the total number of sentences, and the proportion of problematic opinions is the number of problematic sentences divided by the total number of sentences. According to the needs, in each questioning process, according to the needs of the proportion of declarative opinions and the proportion of problematic opinions, the tendency is increased, and the tendency includes increasing the proportion of declarative opinions or the proportion of problematic opinions; after each digital employee's feedback, the proportion of reply sentences containing new nouns is extracted. When the proportion of new nouns is less than the preset value, it is required to increase the proportion of reply sentences containing new nouns, wherein the proportion of reply sentences containing new nouns is the number of sentences that do not include the name in the given information divided by the total number of sentences.

[0101] Figure 2 The present invention is a flowchart of setting a memory button, an API interface and an online data access interface for a human-computer interaction system in a method for realizing a conversational digital employee human-computer interaction according to an embodiment of the present invention.

[0102] like Figure 2 As shown, in one or more embodiments, preferably, the human-computer interaction system is provided with a memory button, an API interface and an online data access interface, specifically including:

[0103] S201, setting a memory button for storing query information online;

[0104] S202, setting an API interface for extracting question and answer information in corresponding professional fields for matching;

[0105] S203: Setting an online data access interface for online extraction of online data for control.

[0106] In an embodiment of the present invention, an efficient human-computer interaction system is constructed, which optimizes the user experience and improves the responsiveness of the system by integrating memory buttons, API interfaces and online data access interfaces. First, the system sets a functional module called "memory buttons", which functions similarly to the memory function of humans and can store the user's previous inquiry information online. This storage mechanism allows the system to reference previous information in subsequent conversations, thereby achieving a more personalized and coherent conversation experience. For example, if the user has asked questions about the weather before, the system can actively provide weather forecasts in subsequent conversations without the user having to ask questions again. Secondly, the system is configured with API interfaces, which are specifically used to extract question and answer information in corresponding professional fields for matching. By connecting to an external knowledge base or database, the API interface can quickly retrieve and provide accurate answers based on the user's query request. For example, when a user asks medical-related questions, the API interface can connect to a professional medical database to obtain and return relevant medical information. Finally, the system also sets an online data access interface, the function of which is to extract online data for control online. This means that the system can monitor dynamic data such as network status and user behavior in real time, and adjust its own operation strategy based on these data. For example, if an increase in network latency is detected, the system may automatically reduce the frequency of data transmission to ensure the smoothness of the conversation. In summary, through the collaborative work of these three key components, the human-computer interaction system of the present invention can not only provide more intelligent and personalized services, but also adapt to changes in the external environment in real time, thereby significantly improving the user's interactive experience.

[0107] Figure 3 It is a flowchart of defining feedback personality for a robot in a method for implementing a conversational digital employee human-machine interaction according to an embodiment of the present invention.

[0108] like Figure 3 As shown, in one or more embodiments, preferably, the step of defining a feedback personality for the robot specifically includes:

[0109] S301, pre-setting the feedback format of the character setting including roles, skills and precautions;

[0110] S302, presetting a table corresponding to a persona for selecting a persona;

[0111] S303, pre-setting the output method of the personality feedback.

[0112] In an embodiment of the present invention, a highly customizable and interactive robot feedback system is designed. The system guides the robot's behavior and communication methods through pre-set personas (role settings), thereby providing a more personalized and contextualized user experience. First, the system specifies the feedback persona for the robot, which includes three core elements: role, skills, and precautions. Roles define the robot's identity in a specific situation, such as "friendly tour guide" or "professional doctor's assistant", which helps shape the robot's personality and behavior pattern. Skills clarify the scope of tasks that the robot can perform. For example, a robot set as "IT support expert" will have the ability to solve technical problems. Precautions are to ensure that the robot follows specific behavioral codes during the interaction process, such as being polite and avoiding sensitive topics. Next, the system pre-sets a table corresponding to the persona, which, as part of the user interface, allows users to select different personas as needed. Users can easily switch the robot's role through this table to adapt to different conversation scenarios. For example, if users need health consultation, they can choose the "medical consultant" persona; if they need travel information, they can choose the "travel guide" persona. Finally, the system also pre-sets the output method of the persona feedback. This means that when the robot interacts with the user, the content and style of its feedback will be consistent with its personality. For example, when the robot is set as a "humorous friend", its answers may contain more humorous elements and a relaxed tone; while as a "serious teacher", its feedback will be more educational and instructive. In this way, the embodiments of the present invention not only enhance the robot's interactive capabilities, enabling it to better adapt to the diverse needs of users, but also improve the quality of user experience, enabling the robot to provide more natural and appropriate services in different situations.

[0113] Figure 4 It is a flowchart of setting corresponding opening remarks for different personalities in a method for realizing a conversational digital employee human-computer interaction according to an embodiment of the present invention.

[0114] like Figure 4 As shown, in one or more embodiments, preferably, the corresponding opening remarks are displayed for different personas, specifically including:

[0115] S401, selecting a pre-set opening statement according to the character setting;

[0116] S402: When a robot corresponding to the character setting is queried online, the opening remarks corresponding to the robot corresponding to the character setting are retrieved.

[0117] In an embodiment of the present invention, an intelligent dialogue system is constructed, which can display corresponding opening remarks according to different personas (role settings) to enhance user experience and improve the naturalness of interaction. First, the system pre-sets a series of opening remarks corresponding to different personas. These opening remarks are carefully designed sentences or paragraphs, which are intended to reflect the personality, style and function of a specific persona. For example, for a robot set as a "friendly travel guide", its opening remarks may be: "Hello! I am your travel assistant. Are you ready to explore this beautiful city together?" For a robot set as a "professional legal advisor", its opening remarks may be: "Welcome to use the legal consulting service. I will provide you with the most professional advice." Then, when the user asks the robot corresponding to the persona online, the system will automatically retrieve the opening remarks corresponding to the robot of the persona. This process is implemented by an algorithm, which will identify the most suitable persona according to the user's query content and context information, and trigger the corresponding opening remarks. For example, if the user enters a question about travel through the chat interface, the system will recognize that the user needs travel-related information, so it will automatically select the "friendly travel guide" persona and display its corresponding opening remarks. This implementation not only improves the robot's response efficiency, but also enhances the user's immersion and satisfaction through personalized opening remarks. Users can immediately feel the robot's role and personality, making it easier to build trust and willingness to interact. In addition, this approach also enables the robot to be flexibly applied in a variety of scenarios. Whether it is providing customer service, educational guidance, or entertainment interaction, it can provide appropriate opening remarks based on different personalities to meet the diverse needs of users.

[0118] Figure 5 The present invention is a flowchart of presetting an enhanced search mode in a method for realizing a conversational digital employee human-computer interaction and providing an enhanced reply based on search information in an embodiment of the present invention.

[0119] like Figure 5 As shown, in one or more embodiments, preferably, the pre-setting of the enhanced search mode and the enhanced reply according to the search information specifically include:

[0120] S501, in a preset enhanced search mode, setting a method and order for filtering information;

[0121] S502: Filter and sort the retrieved information by presetting enhanced search parameters and rules.

[0122] In an embodiment of the present invention, an enhanced retrieval mode is designed to improve the accuracy and efficiency of information retrieval. This mode filters and sorts the retrieved information through the pre-set way and order of filtering information, as well as the parameters and rules of enhanced retrieval, so as to provide users with more accurate and relevant replies. First, the system pre-sets the enhanced retrieval mode, which includes defining the way and order of how to filter information. For example, when a user queries for "best restaurants", the system may first filter out restaurants located near the user's current location, and then further filter according to the user's historical preferences (such as cuisine, price range, etc.), and finally sort them according to the rating or recommendation. This multi-level screening mechanism ensures that only the information that best meets the user's needs will be displayed. Then, the system optimizes the retrieval results through the pre-set enhanced retrieval parameters and rules. These parameters may include keyword weights, time range restrictions, source credibility assessment, etc. For example, if the user is looking for the latest technology news, the system can set a time range parameter to only retrieve articles published in the past week; at the same time, through the source credibility assessment, reports from well-known technology media are given priority. Take a specific example to illustrate this process: suppose the user wants to know information about "the latest progress in artificial intelligence". The system first identifies the keywords "artificial intelligence" and "latest progress", and then applies the pre-set enhanced retrieval mode. In this mode, the system may first screen out relevant articles published in the past month, and then conduct credibility assessments based on the sources of the articles (such as top academic journals, authoritative news websites, etc.), and give articles from highly credible sources higher rankings. Finally, the system presents the filtered and sorted results to the user to ensure that the user can quickly obtain the most relevant and reliable information. In this way, the enhanced retrieval mode of the present invention not only improves the efficiency of information retrieval, but also significantly improves the user's retrieval experience and satisfaction through a multi-dimensional screening and sorting mechanism.

[0123] Figure 6 The present invention is a flowchart of feedback-oriented suggestion information in a method for implementing a conversational digital employee human-computer interaction according to an embodiment of the present invention.

[0124] like Figure 6 As shown, in one or more embodiments, preferably, the feedback directional suggestion information specifically includes:

[0125] S601. According to the error information fed back by the robot in each conversation, clarify the direction of directional suggestions;

[0126] S602. Propose targeted suggestions based on the rules set by the questioner in each conversation.

[0127] In the embodiment of the present invention, an intelligent dialogue system is designed, which has the function of feeding back targeted suggestion information. This function aims to provide more accurate and targeted suggestions by analyzing the wrong information fed back by the robot in each dialogue and according to the rules set by the questioner.

[0128] Figure 7 It is a flowchart of setting key reminder information in a method for implementing a conversational digital employee human-computer interaction according to an embodiment of the present invention, storing it in the memory button, and reminding the robot in each reply and information feedback.

[0129] like Figure 7 As shown, in one or more embodiments, preferably, the key reminder information is stored in the memory button, and the robot is reminded in each reply and information feedback, specifically including:

[0130] S701, each time when the reply information is obtained, automatically extract the clear declarative opinions therein, number them, and store them in the memory button as clear opinions;

[0131] S702, numbering the clear viewpoints;

[0132] S703, obtaining the upper limit of input for each feedback reply, and calculating the number of remaining words according to the first calculation formula;

[0133] S704, calculating the current viewpoint number using the second calculation formula;

[0134] S705, calculating the updated current viewpoint set using a third calculation formula;

[0135] S706, calculating the viewpoint number each time using the fourth calculation formula according to the updated current viewpoint set until the fifth calculation formula is satisfied;

[0136] S707, after the fifth calculation formula is satisfied, all the clear viewpoints are updated, and the second, third, and fourth calculation formulas are used again to calculate the viewpoint number of this time;

[0137] S708. In each reply, start with "What needs to be reminded" and give feedback one by one on the clear opinions corresponding to the opinion number of this time;

[0138] The first calculation formula is:

[0139] SX-YR=SY

[0140] Among them, SX is the input upper limit, YR is the number of words entered, and SY is the number of remaining words;

[0141] The second calculation formula is:

[0142] BC=Argmin(|SY-ΣGZ|)

[0143] SY-ΣGZ>0

[0144] BC∈JH

[0145] Where Argmin is a function for extracting the number of this viewpoint that minimizes |SY-ΣGZ|, ΣGZ is the total number of words in this viewpoint, BC is the number of this viewpoint, and JH is the current viewpoint set;

[0146] The third calculation formula is:

[0147] JHq=JH-BC

[0148] Among them, JHq is the updated current viewpoint set;

[0149] The fourth calculation formula is:

[0150] BCn=Argmin(|SY-ΣGZ|)

[0151] SY-ΣGZ>0

[0152] BCn∈JHq

[0153] Among them, BCn is the viewpoint number of each time;

[0154] The fifth calculation formula is:

[0155] JHq=kong

[0156] Among them, kong is the empty set.

[0157] In an embodiment of the present invention, an intelligent dialogue system is designed, which has the function of feeding back targeted suggestion information. This function is intended to provide more accurate and targeted suggestions by analyzing the error information fed back by the robot in each dialogue and according to the rules set by the questioner. Specifically, the system first records and analyzes the error information fed back by the robot in each dialogue. These error messages may include questions that the robot cannot understand, inaccurate or irrelevant answers, etc. Through in-depth analysis of these error messages, the system can identify common types of questions and user demand patterns. For example, if the robot fails to correctly understand questions about a specific area many times, the system will mark this area as a focus for improvement. Next, the system will make targeted suggestions based on the rules set by the questioner in each dialogue. These rules can be customized by users to reflect their specific needs or preferences. For example, users can set rules that require the robot to give clothing suggestions and outdoor activity suggestions when providing weather information. When the robot detects that the user is asking about the weather, it will not only report the current weather conditions, but also provide additional suggestions based on the user's rules, such as "The temperature is low today, please add appropriate clothes" or "Today is sunny and suitable for outdoor sports." To illustrate this process with a specific example: suppose the user asks for advice on healthy eating, and the robot provides some basic dietary guidelines. However, the user then indicates that they have a specific health condition and need more specific advice. In this case, the system adjusts its answer strategy based on the user's feedback and the set rules. In future conversations, when similar health conditions are mentioned, the robot will be able to provide more personalized and targeted advice, such as recommending low-sugar recipes or high-fiber food options. In this way, the intelligent dialogue system of the present invention can not only learn from mistakes and improve its performance, but also provide more considerate services based on the personalized needs of users. This mechanism of feedback of targeted suggestion information significantly improves the user experience and enables the robot to better meet the diverse needs of users.

[0158] According to a second aspect of an embodiment of the present invention, a conversational digital employee human-computer interaction implementation system is provided.

[0159] Figure 8 It is a structural diagram of a conversational digital employee human-computer interaction implementation system according to an embodiment of the present invention.

[0160] In one or more embodiments, preferably, the conversational digital employee human-computer interaction implementation system includes:

[0161] The structure construction module 801 is used to set memory buttons, API interfaces and online data access interfaces for the human-computer interaction system;

[0162] A persona generation module 802, used to define a feedback persona for the robot;

[0163] The opening remarks module 803 is used to set corresponding opening remarks for different characters;

[0164] An enhanced search module 804 is used to pre-set an enhanced search mode and provide an enhanced reply based on the search information;

[0165] A directional suggestion module 805, used to feed back directional suggestion information;

[0166] The key memory information learning module 806 is used to set the key reminder information to be stored in the memory button, and remind the robot in each reply and information feedback.

[0167] In the embodiment of the present invention, a system suitable for different structures is realized through a series of modular designs. The system can achieve closed-loop, reliable and efficient execution through collection, analysis and control.

[0168] According to a third aspect of an embodiment of the present invention, there is provided a computer-readable storage medium on which computer program instructions are stored. When the computer program instructions are executed by a processor, the method as described in any one of the first aspect of the embodiment of the present invention is implemented.

[0169] According to a fourth aspect of an embodiment of the present invention, an electronic device is provided. Fig. 9 It is a structural diagram of an electronic device in one embodiment of the present invention. Fig. 9 The electronic device shown is a conversational digital employee human-computer interaction implementation device, which includes a general computer hardware structure, which at least includes a processor 901 and a memory 902. The processor 901 and the memory 902 are connected via a bus 903. The memory 902 is suitable for storing instructions or programs executable by the processor 901. The processor 901 can be an independent microprocessor or a collection of one or more microprocessors. Thus, the processor 901 executes the instructions stored in the memory 902, thereby executing the method flow of the embodiment of the present invention as described above to realize the processing of data and the control of other devices. The bus 903 connects the above-mentioned multiple components together, and at the same time connects the above-mentioned components to the display controller 904 and the display device and the IO device 905. The IO device 905 can be a mouse, a keyboard, a modem, a network interface, a touch input device, a somatosensory input device, a printer, and other devices known in the art. Typically, the IO device 905 is connected to the system through the IO controller 906.

[0170] The technical solution provided by the embodiments of the present invention may have the following beneficial effects:

[0171] In the solution of the present invention, a clear digital employee question and answer system is combined to conduct online questions and answers according to the pre-set feedback personality.

[0172] In the solution of the present invention, the question and answer information supplement based on the dynamically adjusted memory button is combined to achieve efficient and accurate question and answer feedback.

[0173] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage and optical storage, etc.) containing computer-usable program code.

[0174] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0175] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0176] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0177] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A method for implementing a conversational digital employee human-computer interaction, characterized in that: The method includes: Set up memory buttons, API interfaces and online data access interfaces for the human-computer interaction system; Define feedback personas for the robot; Set up corresponding opening remarks for different characters; Pre-set enhanced search mode to provide enhanced responses based on search information; Feedback targeted advice information; Setting key reminder information to be stored in the memory button to remind the robot in each reply and information feedback; Specifically, the key reminder information is stored in the memory button, and the robot is reminded in each reply and information feedback process, including: in each reply and information feedback process, the information in the memory button is automatically read, and all the declarative viewpoints in the memory button are extracted, and the declarative viewpoints are numbered, and according to the input upper limit of each reply, the number of the current viewpoint that meets the second calculation formula is supplemented in all the declarative viewpoints, and after all the declarative viewpoints are continuously replied and fully fed back for multiple times, all the clear viewpoints are updated; The second calculation formula is: BC=Argmin(|SY-ΣGZ|) SY-ΣGZ>0 BC∈JH Wherein, Argmin is a function for extracting the number of this viewpoint that minimizes |SY-ΣGZ|, ΣGZ is the total number of words in this viewpoint, BC is the number of this viewpoint, JH is the current viewpoint set, and SY is the number of remaining words; After each digital employee feedback response, extract the proportion of declarative opinions and the proportion of problematic opinions, wherein the proportion of declarative opinions is the number of declarative sentences divided by the total number of sentences, and the proportion of problematic opinions is the number of problematic sentences divided by the total number of sentences. According to the needs, in each questioning process, according to the needs of the proportion of declarative opinions and the proportion of problematic opinions, increase the tendency, which includes increasing the proportion of declarative opinions or the proportion of problematic opinions; After each digital employee feedback reply, the proportion of reply sentences containing new nouns is extracted. When the proportion of new nouns is less than a preset value, it is required to increase the proportion of reply sentences containing new nouns, wherein the proportion of reply sentences containing new nouns is the number of sentences that do not include the name in the given information divided by the total number of sentences.

2. A method for implementing a conversational digital employee human-computer interaction as claimed in claim 1, characterized in that: The human-computer interaction system is provided with a memory button, an API interface and an online data access interface, specifically including: Set up a memory button to store the query information online; Set up an API interface to extract question and answer information in corresponding professional fields for matching; Set up an online data access interface to extract online data for control.

3. A method for implementing a conversational digital employee human-computer interaction as claimed in claim 1, characterized in that: The feedback personality of the robot is specifically provided as follows: The feedback format for pre-set characters includes roles, skills, and considerations; A table with corresponding personas is pre-set for selecting personas; Pre-set the output method of the character's feedback.

4. A method for implementing a conversational digital employee human-computer interaction as claimed in claim 1, characterized in that: The corresponding opening remarks for different character settings specifically include: Choose a pre-set opening statement based on your character settings; When a robot query of a corresponding persona is conducted online, the opening remarks corresponding to the robot of the persona are retrieved.

5. A method for implementing a conversational digital employee human-computer interaction as claimed in claim 1, characterized in that: The pre-setting enhanced search mode and providing an enhanced reply according to the search information specifically include: In the pre-set enhanced search mode, set the method and order of filtering information; By pre-setting enhanced search parameters and rules, the retrieved information can be filtered and sorted.

6. A method for implementing a conversational digital employee human-computer interaction as claimed in claim 1, characterized in that: The feedback-oriented suggestion information specifically includes: According to the error information fed back by the robot in each conversation, the direction of directional suggestions is clearly defined; Provides targeted suggestions based on the rules set by the questioner in each conversation.

7. A method for implementing a conversational digital employee human-computer interaction as claimed in claim 1, characterized in that: The key reminder information is stored in the memory button, and the robot is reminded in each reply and information feedback, specifically including: Every time when receiving reply information, the clear declarative opinions are automatically extracted, numbered, and stored in the memory button as clear opinions; Number the stated clear points; Obtain the input upper limit of each feedback reply, and calculate the remaining word count according to the first calculation formula; Use the second calculation formula to calculate the viewpoint number of this time; Calculate the updated current viewpoint set using the third calculation formula; Calculate the viewpoint number each time using the fourth calculation formula according to the updated current viewpoint set until the fifth calculation formula is satisfied; After the fifth calculation formula is satisfied, all the clear viewpoints are updated, and the viewpoint number of this time is calculated again using the second, third, and fourth calculation formulas; In each reply, start with "What needs to be reminded" and give feedback one by one on the clear opinions corresponding to the opinion number of this time; The first calculation formula is: SX-YR=SY Among them, SX is the input upper limit, YR is the number of words entered, and SY is the number of remaining words; The third calculation formula is: JHq=JH-BC Among them, JHq is the updated current viewpoint set; The fourth calculation formula is: BCn=Argmin(|SY-ΣGZ|) SY-ΣGZ>0 BCn∈JHq Among them, BCn is the viewpoint number of each time; The fifth calculation formula is: JHq=kong Among them, kong is the empty set.

8. A conversational digital employee human-computer interaction implementation system, characterized in that: The system is used to implement the method according to any one of claims 1 to 7, and the system comprises: Structural building module, used to set up memory buttons, API interfaces and online data access interfaces for the human-computer interaction system; A character generation module is used to define a feedback character for the robot; The opening remarks module is used to set up corresponding opening remarks for different characters; An enhanced search module, used to pre-set an enhanced search mode and provide an enhanced reply based on the search information; A directional suggestion module, used to feedback directional suggestion information; The key memory information learning module is used to set key reminder information to be stored in the memory button, and remind the robot in each reply and information feedback.

9. A computer-readable storage medium storing computer program instructions, characterized in that: The computer program instructions implement the method according to any one of claims 1 to 7 when executed by a processor.

10. An electronic device comprising a memory and a processor, characterized in that: The memory is used to store one or more computer program instructions, wherein the one or more computer program instructions are executed by the processor to implement the method according to any one of claims 1 to 7.

Images