Self-adaptive common-situation intelligent dialogue method and device, electronic equipment and storage medium

Abstract

The invention provides a self-adaptive common-situation intelligent dialogue method and device, electronic equipment and a storage medium. The method comprises the following steps: receiving user input; acquiring a professional guidance reply corresponding to user input; acquiring a corresponding emotion resonance reply input by the user; and determining a target reply input by the user according to the professional guidance reply and the emotion resonance reply. According to the method, professional semantic analysis and sentiment analysis are carried out on user input at the same time, so that target reply has professionality and sentiment temperature at the same time, the double requirements of the user for professional guidance and sentiment support in the dialogue process are met, effective distinguishing between the sentiment attribute and the professional attribute of reply content is achieved, and the user experience is improved. And the intelligent level of the system and the experience feeling of the user are improved.

Description

Technical Field

[0001] This invention relates to the field of intelligent dialogue technology, and in particular to an adaptive empathetic intelligent dialogue method, apparatus, electronic device, and storage medium. Background Technology

[0002] With the rapid development of artificial intelligence and natural language processing technologies, intelligent dialogue systems have been widely used in many fields. For example, in the field of mental health services, intelligent psychological counseling systems have become an important means to improve the efficiency of psychological intervention and expand the service coverage.

[0003] In existing technologies, a typical intelligent dialogue system usually includes the following processing flow: First, the system performs basic parsing of the text information input by the user using natural language processing technology to identify its semantic structure, keywords, and basic intent; then, it directly calls a pre-trained large language model to generate a response in a uniform format, without distinguishing the functional attributes carried by the response; finally, the generated response is directly fed back to the user, completing one round of interaction.

[0004] However, the aforementioned existing technical solutions have the following drawbacks: Due to the use of a single response generation mechanism, the system cannot effectively distinguish and process emotional responses aimed at emotional comfort and professional guidance responses aimed at problem-solving, resulting in numerous problems with the generated response content in practical applications. For example, responses may appear stiff and lack empathy, failing to meet the psychological support needs of users in specific emotional states; furthermore, the system cannot dynamically adjust the response style and expression according to the user's real-time emotional state, thus affecting the overall effectiveness of psychological counseling and the user experience.

[0005] Therefore, how to solve the problem that existing intelligent dialogue systems cannot effectively distinguish between the emotional and professional attributes of the response content, and cannot adaptively generate more targeted or warm responses based on the user's current emotional state, is an important issue that urgently needs to be addressed in the field of intelligent dialogue. Summary of the Invention

[0006] This invention provides an adaptive empathic intelligent dialogue method, device, electronic device, and storage medium to overcome the shortcomings of existing intelligent dialogue systems that cannot effectively distinguish between the emotional and professional attributes of response content, and cannot adaptively generate more targeted or warmer responses based on the user's current emotional state. It achieves effective distinction between the emotional and professional attributes of response content and adaptively generates more targeted and warmer responses based on the user's current emotional state, thereby improving the system's intelligence level and the user's experience.

[0007] On one hand, the present invention provides an adaptive empathic intelligent dialogue method, comprising: receiving user input; obtaining a professional guidance response corresponding to the user input; obtaining an emotional resonance response corresponding to the user input; and determining a target response for the user input based on the professional guidance response and the emotional resonance response.

[0008] Furthermore, obtaining the professional guidance response corresponding to the user input includes: obtaining the context information of the user input; and obtaining the professional guidance response based on the large language model, according to the user input and the context information.

[0009] Furthermore, obtaining the professional guidance response includes: calling the target domain knowledge base, correcting the target domain professional content in the professional guidance response, and obtaining the final professional guidance response; correspondingly, determining the target response input by the user based on the professional guidance response and the emotional resonance response includes: determining the target response input by the user based on the final professional guidance response and the emotional resonance response.

[0010] Furthermore, obtaining the emotional resonance response corresponding to the user input includes: acquiring the user's emotional state based on the user input using a pre-trained sentiment analysis model; wherein the sentiment analysis model is trained based on user input samples and corresponding sentiment tags; and adjusting the professional guidance response according to the user's emotional state to obtain the emotional resonance response.

[0011] Furthermore, it also includes: recording the user's emotional state in multi-round dialogues in real time to construct a dynamic emotional profile of the user; and based on the dynamic emotional profile, assisting in obtaining emotional resonance responses corresponding to the user's input.

[0012] Further, determining the target response input by the user based on the professional guidance response and the emotional resonance response includes: obtaining a first weight corresponding to the professional guidance response and a second weight corresponding to the emotional resonance response; and fusing the professional guidance response and the emotional resonance response based on the first weight and the second weight to obtain the target response input by the user.

[0013] Secondly, the present invention also provides an adaptive empathy intelligent dialogue device, comprising: a user input receiving module for receiving user input; a professional guidance response acquisition module for acquiring a professional guidance response corresponding to the user input; an emotional resonance response acquisition module for acquiring an emotional resonance response corresponding to the user input; and a target response determination module for determining a target response to the user input based on the professional guidance response and the emotional resonance response.

[0014] Thirdly, the present invention also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor, when executing the computer program, implements the adaptive empathic intelligent dialogue method as described above.

[0015] Fourthly, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the adaptive empathic intelligent dialogue method as described above.

[0016] Fifthly, the present invention also provides a computer program product, including a computer program that, when executed by a processor, implements the adaptive empathic intelligent dialogue method as described above.

[0017] The adaptive empathic intelligent dialogue method provided by this invention receives user input and obtains corresponding professional guidance and emotional resonance responses. Based on these responses, the target response for the user input is determined. This method simultaneously performs professional semantic and sentiment analysis on the user input, ensuring that the target response possesses both professionalism and emotional warmth. This satisfies the user's dual needs for professional guidance and emotional support during the dialogue, effectively distinguishing between the emotional and professional attributes of the response content, thus improving the system's intelligence and user experience. Attached Figure Description

[0018] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0019] Figure 1 This is a flowchart illustrating the adaptive empathy-based intelligent dialogue method provided in an embodiment of the present invention.

[0020] Figure 2 This is a schematic diagram of the structure of the adaptive empathy intelligent dialogue device provided in an embodiment of the present invention.

[0021] Figure 3 This is a schematic diagram of the physical structure of the electronic device provided in an embodiment of the present invention. Detailed Implementation

[0022] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0023] It should be noted that most existing intelligent dialogue systems utilize natural language processing technology to perform basic parsing of text input and directly call large language models to generate uniform responses, without distinguishing between emotional and guiding attributes. This can result in responses that sound rather stiff and lack sufficient emotional warmth. Furthermore, existing intelligent dialogue systems cannot dynamically adjust their response style based on the user's real-time emotional state, impacting the user's dialogue experience.

[0024] In view of this, the present invention proposes an adaptive empathy-based intelligent dialogue method, specifically, Figure 1 A flowchart illustrating the adaptive empathy-based intelligent dialogue method provided in an embodiment of the present invention is shown.

[0025] like Figure 1 As shown, the method includes: S110, receiving user input; S120, obtaining a professional guidance response corresponding to the user input; S130, obtaining an emotional resonance response corresponding to the user input; S140, determining a target response for the user input based on the professional guidance response and the emotional resonance response.

[0026] It should be noted that the adaptive empathy intelligent dialogue method provided in this embodiment can be executed by any electronic device. The electronic device here can be not only smart home devices such as smart speakers and smart TVs, but also personal devices such as smartphones, tablets / computers, smartwatches, and smart bracelets, etc., without specific limitations.

[0027] For ease of understanding, the following will take the electronic device as the execution subject and provide a detailed description of steps S110-S140 and related steps.

[0028] S110 receives user input.

[0029] It is easy to understand that electronic devices can receive user input in real time via voice, text, or images. That is to say, the received user input can be voice input, text input, or image input, without being specifically limited here.

[0030] The professional knowledge field involved in the user input can be psychological counseling, education, or medical fields, and no specific limitation is made here.

[0031] Based on receiving user input in step S110, step S120 is further executed.

[0032] S120, Obtain the professional guidance response corresponding to the user input.

[0033] What's easy to understand is that by using a large language model combined with contextual information from multi-turn dialogues to perform semantic analysis on user input, the implicit professional content in the user input is parsed, thereby generating a professional guidance response corresponding to the user input.

[0034] Among them, the Large Language Model (LLM) is a neural network model that uses a large amount of text data for pre-training. It can understand and generate natural language and achieve zero-shot or few-shot transfer learning capabilities in a variety of natural language processing tasks.

[0035] Professional guidance responses refer to replies to professional content entered by the user, such as professional psychological counseling, medicine, law, mathematics, etc.

[0036] Based on obtaining the professional guidance response corresponding to the user input in step S120, step S130 is further executed.

[0037] S130, Obtain the emotional resonance response corresponding to the user input.

[0038] The easy-to-understand approach is to first use a pre-trained sentiment analysis model to perform sentiment analysis on user input, accurately identifying the user's emotional state contained in the input; then, based on the identified user emotional state, adjust the style of the professional guidance response to obtain an emotionally resonant response corresponding to the user input.

[0039] The sentiment analysis model can be trained and optimized based on user input samples and their corresponding sentiment labels. The sentiment analysis model can be built based on convolutional neural networks or other network structures; no specific limitations are specified here.

[0040] Emotion labels can be set according to actual needs, and no specific limitations are made here. For example, in a specific embodiment, emotion labels include negative emotions, positive emotions, and neutral emotions, or emotion labels include happiness, sadness, anger, fear, disgust, surprise, and anxiety.

[0041] Emotional resonance response refers to a response that addresses the emotional content entered by the user, evoking a sense of resonance in the user. These emotional contents include feelings such as sadness, melancholy, and calmness.

[0042] Based on obtaining the emotional resonance response corresponding to the user input in step S130, step S140 is further executed.

[0043] S140, Based on the professional guidance response and the emotional resonance response, determine the target response input by the user.

[0044] It's easy to understand that by adaptively fusing the professional guidance and emotionally resonant responses obtained, the target response corresponding to the user's input can be obtained.

[0045] Adaptive fusion can be achieved by setting corresponding weight values ​​for professional guidance responses and emotional resonance responses.

[0046] The target responses cover both professional knowledge and emotional feedback.

[0047] In this embodiment, by receiving user input and obtaining corresponding professional guidance and emotional resonance responses, the system determines the target response based on both the professional guidance and emotional resonance responses. This method simultaneously performs professional semantic and sentiment analysis on the user input, ensuring that the target response possesses both professionalism and emotional warmth. This satisfies the user's dual needs for professional guidance and emotional support during the conversation, effectively distinguishing between the emotional and professional attributes of the response content, and enhancing the system's intelligence and user experience.

[0048] Based on the above embodiments, the process of obtaining professional guidance responses will be described in detail below.

[0049] In one specific embodiment, obtaining a professional guidance response corresponding to user input includes: obtaining contextual information of user input; and obtaining a professional guidance response based on a large language model, according to user input and contextual information.

[0050] The straightforward approach is to obtain the contextual information corresponding to the user input after receiving it—that is, the contextual information in a multi-turn dialogue. Then, a contextual representation is constructed based on the user input and the contextual information. This contextual representation is then used as input to a large language model to obtain the output, providing professional guidance.

[0051] In particular, the introduction of contextual information in multi-turn dialogues enables large language models to extract users' true intentions and needs from the semantic analysis of user input within a richer historical context.

[0052] Furthermore, after obtaining the professional guidance response, the process includes: accessing the target domain knowledge base, correcting the target domain professional content in the professional guidance response, and obtaining the final professional guidance response.

[0053] Specifically, the professional content in the target domain of the professional guidance response is matched with the target domain knowledge base to correct the professional content in the professional guidance response, so as to achieve accurate analysis of the explicit / implicit professional content of the user's needs and obtain the final professional guidance response.

[0054] The "target domain" in the target domain knowledge base can be any domain that meets the needs, including but not limited to education, healthcare, law, finance, enterprise services, creative assistance, and other fields.

[0055] After receiving the final professional guidance response, further fine-tuning can be made based on contextual information to ensure that the final professional guidance response is both accurate and meets the requirements of the dialogue task.

[0056] Finally, based on the final professional guidance response and the emotionally resonant response, the target response for the user's input is determined.

[0057] In another specific embodiment, obtaining the professional guidance response corresponding to the user input includes: obtaining the professional guidance response based on the user input using a pre-trained professional response inference model.

[0058] The professional response reasoning model can be trained and optimized using user input samples labeled with professional guidance responses. Through training, the professional response reasoning model can effectively learn how to transform semantic representations into responses with professional structure.

[0059] Similarly, in the process of obtaining professional guidance responses using a professional response reasoning model, the contextual information of the user input can be obtained. This contextual information, along with the user input, is then fed into a pre-trained professional response reasoning model to obtain the output professional guidance response.

[0060] Subsequently, the target domain knowledge base can be accessed to revise the target domain professional content in the professional guidance response, resulting in the final professional guidance response.

[0061] Finally, based on the final professional guidance response and the emotionally resonant response, the target response for the user's input is determined.

[0062] In this embodiment, by acquiring the contextual information of user input, a professional guidance response is obtained based on the user input and its contextual information. The target domain knowledge base is then invoked to correct the professional content of the target domain in the professional guidance response, resulting in the final professional guidance response. Finally, based on the final professional guidance response and the emotional resonance response, the target response input by the user is determined. This method simultaneously performs professional semantic analysis and sentiment analysis on user input, ensuring that the target response possesses both professionalism and emotional warmth. This satisfies the user's dual needs for professional guidance and emotional support during the dialogue process, effectively distinguishing between the emotional and professional attributes of the response content, and improving the system's intelligence level and user experience.

[0063] Based on the above embodiments, the process of obtaining emotional resonance responses will be described in detail below.

[0064] Obtaining emotionally resonant responses to user input includes: acquiring the user's emotional state based on a pre-trained sentiment analysis model; wherein the sentiment analysis model is trained based on user input samples and corresponding sentiment tags; and adjusting professional guidance responses according to the user's emotional state to obtain emotionally resonant responses.

[0065] It is easy to understand that this embodiment pre-trains a sentiment analysis model. During training, the sentiment analysis model takes user input samples as input, predicts the emotional state as output, and uses the difference between the predicted emotional state and the corresponding emotional label as training loss. The sentiment analysis model is iteratively optimized and trained until convergence, resulting in a pre-trained sentiment analysis model.

[0066] During inference, user input is used as input to the sentiment analysis model to obtain the output of the user's emotional state corresponding to the user input.

[0067] The user's emotional state includes the type of emotion and the corresponding intensity of emotion.

[0068] Furthermore, during multi-turn dialogues, the system records the user's emotional state in real time, thereby constructing a dynamic emotional profile of the user and understanding long-term emotional trends in the dialogue history. When identifying the user's emotional state, the dynamic emotional profile can be used along with the user's input as input to the sentiment analysis model to obtain the current user's emotional state, thus assisting in obtaining emotionally resonant responses to the user's input.

[0069] Furthermore, based on the user's emotional state, a specially designed emotional resonance response generation module can be used to adjust the professional guidance response through predefined emotional expression templates. It can automatically select statements that match the user's emotional state, such as expressions of care, comfort, or empathy, to ensure that the generated emotional resonance response can evoke emotional resonance in the user and reflect sincerity, warmth, and humanistic care.

[0070] In this embodiment, a user's emotional state is obtained based on user input using a pre-trained sentiment analysis model. This model is trained using user input samples and corresponding sentiment tags. The professional guidance response is adjusted according to the user's emotional state to obtain an emotionally resonant response. Finally, the target response for the user's input is determined based on both the final professional guidance response and the emotionally resonant response. This method simultaneously performs professional semantic analysis and sentiment analysis on user input, ensuring that the target response possesses both professionalism and emotional warmth. This satisfies the user's dual needs for professional guidance and emotional support during the conversation, effectively distinguishing between the emotional and professional attributes of the response content, thus improving the system's intelligence and user experience.

[0071] Based on the above embodiments, the following will further describe in detail the process of integrating professional guidance responses and emotional resonance responses.

[0072] Based on the professional guidance response and the emotional resonance response, determine the target response input by the user, including: obtaining the first weight corresponding to the professional guidance response and the second weight corresponding to the emotional resonance response; and merging the professional guidance response and the emotional resonance response based on the first weight and the second weight to obtain the target response input by the user.

[0073] The process is straightforward: after obtaining the professional guidance response and the emotional resonance response corresponding to the user's input, the first step is to obtain the weight values ​​for each response—the first weight for the professional guidance response and the second weight for the emotional resonance response. Then, the professional guidance response and the emotional resonance response are weighted and merged using these two weights to obtain the final output, which is the target response corresponding to the user's input.

[0074] In a specific embodiment, in order to achieve an organic combination of professional guidance responses and emotional resonance responses, an adaptive fusion module can be designed, which is specifically used to dynamically adjust the weights of professional guidance responses and emotional resonance responses according to the real-time user emotional state and task scenario.

[0075] For example, when a user's emotional state is relatively negative, the weight of emotional resonance responses is increased to ensure that more care and emotional support content appears in the target response; when a user has a strong need for professional information or is in a neutral or positive emotional state, professional guidance responses take the lead, while being supplemented with appropriate emotional embellishment.

[0076] In another specific embodiment, to achieve an organic combination of professional guidance responses and emotionally resonant responses, a series of preset rules were designed to automatically select a fusion strategy in different contexts. For example, the preset rules can predefine multiple fusion templates based on parameters such as task scenario, user emotional state, and dialogue context; during application, the corresponding fusion template is matched according to the actual situation to achieve personalized adjustments to the response style.

[0077] Within each integration template, you can set the relative order, length, and bias of professional guidance responses and emotionally resonant responses, thereby ensuring that the target response remains both professional and warm.

[0078] In this embodiment, a target response that simultaneously encompasses both professionalism and emotional resonance is obtained by employing a weight-based adaptive fusion method or a rule-based adaptive strategy to fuse professional guidance responses and emotional resonance responses. This method performs both professional semantic analysis and sentiment analysis on user input, ensuring that the target response possesses both professionalism and emotional warmth. This satisfies the user's dual needs for professional guidance and emotional support during the conversation, effectively distinguishing between the emotional and professional attributes of the response content, and enhancing the system's intelligence and user experience.

[0079] In some embodiments, a data acquisition module based on video or physiological sensors is added to monitor parameters such as the user's facial expressions, changes in voice emotion, and heart rate, to assist the emotion analysis model in judging the user's emotional state and improve the accuracy of the user's emotional state recognition.

[0080] In other embodiments, generative adversarial networks are used to fuse professional guidance responses and emotionally resonant responses to improve the balance between professionalism and emotional expression in the target response.

[0081] In some embodiments, a personalized model can be created for users who use the system for a long time. The personalized model can adjust the production strategy of the target response based on the user's historical dialogue data and feedback to achieve a highly personalized dialogue service.

[0082] Furthermore, it is worth mentioning that the adaptive empathy-based intelligent dialogue method provided in this invention can be applied not only to psychological counseling scenarios but also to intelligent customer service, educational consultation, and medical Q&A scenarios. Utilizing domain-customized data and experience rules, the system can flexibly switch between professional guidance response mode, emotional resonance response mode, or a fusion of both in different scenarios, thereby significantly improving the overall user experience.

[0083] Corresponding to the adaptive empathy intelligent dialogue method described in the above embodiments, the present invention also provides an adaptive empathy intelligent dialogue device.

[0084] Specifically, Figure 2 A schematic diagram of the adaptive empathy intelligent dialogue device provided in an embodiment of the present invention is shown.

[0085] like Figure 2 As shown, the device includes: a user input receiving module 210 for receiving user input; a professional guidance response acquisition module 220 for acquiring a professional guidance response corresponding to the user input; an emotional resonance response acquisition module 230 for acquiring an emotional resonance response corresponding to the user input; and a target response determination module 240 for determining a target response to the user input based on the professional guidance response and the emotional resonance response.

[0086] In this embodiment, user input is received by user input receiving module 210, professional guidance response acquisition module 220 acquires the professional guidance response corresponding to the user input, and emotional resonance response acquisition module 230 acquires the emotional resonance response corresponding to the user input. Then, target response determination module 240 determines the target response for the user input based on the professional guidance response and the emotional resonance response. This device simultaneously performs professional semantic analysis and sentiment analysis on user input, ensuring that the target response possesses both professionalism and emotional warmth. This satisfies the user's dual needs for professional guidance and emotional support during the dialogue process, effectively distinguishing between the emotional and professional attributes of the response content, and improving the system's intelligence level and user experience.

[0087] It should be noted that the adaptive empathy intelligent dialogue device provided in the embodiments of the present invention can be referred to in correspondence with the adaptive empathy intelligent dialogue methods described in the above embodiments, and will not be repeated here.

[0088] Figure 3 An example is a schematic diagram of the physical structure of an electronic device, such as... Figure 3As shown, the electronic device may include a processor 310, a communications interface 320, a memory 330, and a communication bus 340, wherein the processor 310, communications interface 320, and memory 330 communicate with each other via the communication bus 340. The processor 310 can invoke logical instructions in the memory 330 to execute an adaptive empathetic intelligent dialogue method, which includes: receiving user input; obtaining a professional guidance response corresponding to the user input; obtaining an emotional resonance response corresponding to the user input; and determining a target response for the user input based on the professional guidance response and the emotional resonance response.

[0089] Furthermore, the logical instructions in the aforementioned memory 330 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, essentially, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0090] On the other hand, the present invention also provides a computer program product, which includes a computer program that can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer is able to execute the adaptive empathic intelligent dialogue method provided by the above methods. The method includes: receiving user input; obtaining a professional guidance response corresponding to the user input; obtaining an emotional resonance response corresponding to the user input; and determining a target response for the user input based on the professional guidance response and the emotional resonance response.

[0091] In another aspect, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon. When executed by a processor, the computer program implements an adaptive empathic intelligent dialogue method provided by the methods described above. The method includes: receiving user input; obtaining a professional guidance response corresponding to the user input; obtaining an emotional resonance response corresponding to the user input; and determining a target response for the user input based on the professional guidance response and the emotional resonance response.

[0092] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0093] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0094] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. An adaptive empathic intelligent dialogue method, characterized in that, include: Receive user input; Obtain the professional guidance response corresponding to the user input; Obtain emotionally resonant responses corresponding to the user input; Based on the professional guidance response and the emotional resonance response, determine the target response input by the user.

2. The adaptive empathy-based intelligent dialogue method according to claim 1, characterized in that, The process of obtaining the professional guidance response corresponding to the user input includes: Obtain the context information of the user input; Based on the large language model, the professional guidance response is obtained according to the user input and the context information.

3. The adaptive empathy-based intelligent dialogue method according to claim 2, characterized in that, The process of obtaining the professional guidance response then includes: The target domain knowledge base is invoked to correct the target domain professional content in the professional guidance response, resulting in the final professional guidance response; Accordingly, determining the target response input by the user based on the professional guidance response and the emotional resonance response includes: Based on the final professional guidance response and the emotionally resonant response, the target response input by the user is determined.

4. The adaptive empathy-based intelligent dialogue method according to claim 1, characterized in that, The step of obtaining the emotional resonance response corresponding to the user input includes: Based on a pre-trained sentiment analysis model, the user's emotional state is obtained according to the user input; wherein, the sentiment analysis model is trained based on user input samples and corresponding sentiment labels; Based on the user's emotional state, the professional guidance response is adjusted to obtain the emotionally resonant response.

5. The adaptive empathy-based intelligent dialogue method according to claim 4, characterized in that, Also includes: Real-time recording of user emotional states during multi-turn conversations to build dynamic emotional profiles of users; Based on the dynamic emotional profile, the system assists in obtaining emotional resonance responses corresponding to the user input.

6. The adaptive empathic intelligent dialogue method according to any one of claims 1-5, characterized in that, The process of determining the target response input by the user based on the professional guidance response and the emotional resonance response includes: Obtain the first weight corresponding to the professional guidance response and the second weight corresponding to the emotional resonance response; Based on the first weight and the second weight, the professional guidance response and the emotional resonance response are combined to obtain the target response input by the user.

7. An adaptive empathic intelligent dialogue device, characterized in that, include: The user input receiving module is used to receive user input. The professional guidance response acquisition module is used to acquire the professional guidance response corresponding to the user input; The emotional resonance response acquisition module is used to acquire the emotional resonance response corresponding to the user input; The target response determination module is used to determine the target response input by the user based on the professional guidance response and the emotional resonance response.

8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that, When the processor executes the computer program, it implements the adaptive empathic intelligent dialogue method as described in any one of claims 1 to 6.

9. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the adaptive empathic intelligent dialogue method as described in any one of claims 1 to 6.

10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the adaptive empathic intelligent dialogue method as described in any one of claims 1 to 6.

Images