Virtual pet emotion intelligent processing feedback method and system and medium

Abstract

The invention provides a virtual pet emotion intelligent processing feedback method and system and a medium. The method comprises the steps that a real-time feature data set is obtained for preprocessing, a real-time decoupling feature data set is obtained, scene evaluation data is combined for processing, user micro-state label data and corresponding emotional intensity parameters are obtained, then processing is conducted through a preset virtual pet bionic memory library, memory calling data is obtained, and the user micro-state label data and the corresponding emotional intensity parameters are displayed. And performing comprehensive processing to obtain collaborative feedback output data of the virtual pet, controlling execution of the virtual pet, obtaining a feedback feature data set, performing processing in combination with scene evaluation data, obtaining user micro-state feedback label data and corresponding feedback emotion intensity parameters, evaluating effective emotion processing parameters of the virtual pet, and finally, performing evaluation on the user micro-state feedback label data and the corresponding feedback emotion intensity parameters. Determining a virtual pet emotion processing state through a threshold value; through multi-modal feature decoupling, bionic memory attenuation analysis and three-dimensional collaborative feedback, closed-loop processing of virtual pet emotion intelligent feedback is realized.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of artificial intelligence and affective computing, in particular to a virtual pet emotional intelligent processing feedback method, system and medium. BACKGROUND

[0002] At present, the traditional virtual pet emotional processing feedback technology collects full data, and the data redundancy is serious. The emotional recognition is mainly performed through feature splicing and fixed weight. For example, CN117880566A directly superimposes image and sound features, which cannot adapt to the dynamic change of scene demand. The memory storage is all static archiving processing, which does not conform to the cognitive law, and the interaction matching with the user is insufficient. At the same time, the existing feedback mainly includes image and voice output of digital terminal. For example, CN119626263A only responds through virtual pet action and voice, which lacks the immersion of physical touch and environmental linkage. Therefore, the traditional technology lacks intelligence and adaptability, and the user experience is poor. Therefore, a new virtual pet emotional intelligent processing feedback method is urgently needed.

[0003] In view of the above problems, an effective technical solution is currently needed. SUMMARY

[0004] The purpose of the present application is to provide a virtual pet emotional intelligent processing feedback method, system and medium, which can realize the closed-loop processing of virtual pet emotional intelligent feedback through multi-modal feature decoupling, bionic memory decay analysis and three-dimensional collaborative feedback.

[0005] In a first aspect, the present application provides a virtual pet emotional intelligent processing feedback method, comprising the following steps: Obtain real-time feature data set of virtual pet user, and perform feature decoupling preprocessing to obtain real-time decoupling feature data set; Obtain scene evaluation data of virtual pet user, process the scene evaluation data in combination with the real-time decoupling feature data set to obtain user micro-state label data of virtual pet user and corresponding emotional intensity parameters; Process the user micro-state label data and corresponding emotional intensity parameters through a preset virtual pet bionic memory library to obtain memory calling data; Process the user micro-state label data and memory calling data through a preset virtual pet emotional intelligent processing model to obtain collaborative feedback output data of virtual pet; Control the virtual pet to execute according to the collaborative feedback output data, and obtain feedback feature data set of the user for feature decoupling preprocessing to obtain feedback decoupling feature data set; According to the feedback decoupling feature data set in combination with the scene evaluation data, user micro-state feedback label data and corresponding feedback emotional intensity parameters of the virtual pet user are obtained; According to the user micro-state feedback label data and corresponding feedback emotional intensity parameters and the user micro-state label data and corresponding emotional intensity parameters, virtual pet emotional processing effective parameters are obtained by processing, and the virtual pet emotional processing effective parameters are processed with preset virtual pet emotional processing effective thresholds to obtain a virtual pet emotional processing state.

[0006] Optionally, in the virtual pet emotional intelligent processing feedback method, the real-time feature data set of the virtual pet user is obtained, and feature decoupling preprocessing is performed to obtain a real-time decoupling feature data set, including: The real-time feature data set of the virtual pet user includes semantic feature data, physiological feature data and behavior feature data. The semantic feature data, physiological feature data and behavior feature data are respectively subjected to feature decoupling preprocessing to obtain a real-time decoupling feature data set, including semantic decoupling feature data, physiological decoupling feature data and behavior decoupling feature data.

[0007] Optionally, in the virtual pet emotional intelligent processing feedback method, the scene evaluation data of the virtual pet user is obtained, and the real-time decoupling feature data set is processed in combination with the scene evaluation data to obtain user micro-state label data and corresponding emotional intensity parameters of the virtual pet user, including: The scene evaluation data of the virtual pet user includes user location data and environmental noise data. The user location data and environmental noise data are input into a preset user scene recognition model for processing to obtain scene category feature data. According to the scene category feature data, a preset decoupling feature weight value mapping table is queried to obtain weight values corresponding to the real-time decoupling feature data set, including semantic weight values, physiological weight values and behavior weight values. The semantic decoupling feature data, physiological decoupling feature data and behavior decoupling feature data and the semantic weight values, physiological weight values and behavior weight values are input into a preset user micro-state evaluation model for processing to obtain user micro-state label data and corresponding emotional intensity parameters of the virtual pet user.

[0008] Optionally, in the virtual pet emotional intelligent processing feedback method, the user micro-state label data and corresponding emotional intensity parameters are processed through a preset virtual pet bionic memory bank to obtain memory calling data, including: According to the user micro-state label data and the corresponding emotional intensity parameters, matching processing is performed on a preset virtual pet bionic memory bank to obtain an initial memory calling data set; Real-time memory intensity, calling frequency and scene correlation degree of the memory calling initial data in the memory calling initial data set are obtained. The real-time memory intensity, calling frequency and scene correlation degree are weighted and summed to obtain a matching value corresponding to the memory calling initial data. The matching value is arranged in descending order, and the memory calling initial data with the largest matching value is determined as the memory calling data.

[0009] Optionally, in the virtual pet emotional intelligent processing feedback method, the user micro-state label data and the memory calling data are input into a preset virtual pet emotional intelligent processing model for processing to obtain the collaborative feedback output data of the virtual pet. The user micro-state label data and the memory calling data are input into a preset virtual pet emotional intelligent processing model for processing to obtain the collaborative feedback output data of the virtual pet. The collaborative feedback output data includes digital end feedback output data, physical end feedback output data and service end feedback output data.

[0010] Optionally, in the virtual pet emotional intelligent processing feedback method, the virtual pet is controlled according to the collaborative feedback output data, and feedback feature data sets of the user are obtained for feature decoupling preprocessing to obtain feedback decoupling feature data sets. The virtual pet is controlled according to the collaborative feedback output data, and feedback feature data sets of the user are obtained, including semantic feedback feature data, physiological feedback feature data and behavior feedback feature data. The semantic feedback feature data, physiological feedback feature data and behavior feedback feature data are respectively subjected to feature decoupling preprocessing to obtain feedback decoupling feature data sets, including semantic feedback decoupling feature data, physiological feedback decoupling feature data and behavior feedback decoupling feature data.

[0011] Optionally, in the virtual pet emotional intelligent processing feedback method, the user micro-state feedback label data and the corresponding feedback emotional intensity parameters and the user micro-state label data and the corresponding emotional intensity parameters are processed to obtain virtual pet emotional processing effective parameters, the virtual pet emotional processing effective parameters are processed with a preset virtual pet emotional processing effective threshold to obtain a virtual pet emotional processing state. input the user micro-state feedback label data and the corresponding feedback emotional intensity parameters and the user micro-state label data and the corresponding emotional intensity parameters into a preset virtual pet emotional processing evaluation model for processing to obtain virtual pet emotional processing effective parameters; process the virtual pet emotional processing effective parameters and a preset virtual pet emotional processing effective threshold to obtain a virtual pet emotional processing state; if the virtual pet emotional processing effective parameters are less than the preset virtual pet emotional processing effective threshold, it is determined that the virtual pet emotional processing state is an invalid state; if the virtual pet emotional processing effective parameters are greater than or equal to the preset virtual pet emotional processing effective threshold, it is determined that the virtual pet emotional processing state is an effective state.

[0012] Optionally, in the virtual pet emotional intelligent processing feedback method, the method further includes: obtaining a preset initial memory intensity, a preset memory decay coefficient, a memory repetition mentioning number, a memory storage emotional intensity parameter and a memory storage time length of the memory calling initial data in the memory calling initial data set; performing weighted summation processing on the memory repetition mentioning number and the memory storage emotional intensity parameter to obtain a memory decay coefficient influence factor; correcting the preset memory decay coefficient according to the memory decay coefficient influence factor to obtain a memory decay correction coefficient; processing the preset initial memory intensity and the memory storage time length according to the memory decay correction coefficient to obtain a real-time memory intensity.

[0013] In a second aspect, the present application provides a virtual pet emotional intelligent processing feedback system, which comprises a memory and a processor, wherein the memory comprises a program of a virtual pet emotional intelligent processing feedback method, and the program of the virtual pet emotional intelligent processing feedback method is executed by the processor to realize the following steps: obtaining a real-time feature data set of a virtual pet user and performing feature decoupling preprocessing to obtain a real-time decoupling feature data set; obtaining scene evaluation data of the virtual pet user and processing the scene evaluation data in combination with the real-time decoupling feature data set to obtain user micro-state label data and corresponding emotional intensity parameters of the virtual pet user; processing the user micro-state label data and the corresponding emotional intensity parameters through a preset virtual pet bionic memory bank to obtain memory calling data; processing the user micro-state label data and the memory calling data through a preset virtual pet emotional intelligent processing model to obtain cooperative feedback output data of the virtual pet; According to the cooperative feedback output data control virtual pet execution, and obtain the user's feedback feature data set for feature decoupling preprocessing, obtain the feedback decoupling feature data set; According to the feedback decoupling feature data set combined with the scene evaluation data, obtain the user micro-state feedback label data of the virtual pet user and the corresponding feedback emotional intensity parameter; According to the user micro-state feedback label data and the corresponding feedback emotional intensity parameter and the user micro-state label data and the corresponding emotional intensity parameter, obtain the virtual pet emotional processing effective parameter, and process the virtual pet emotional processing effective parameter and the preset virtual pet emotional processing effective threshold, obtain the virtual pet emotional processing state.

[0014] In a third aspect, the present application also provides a computer readable storage medium, the computer readable storage medium stores a virtual pet emotional intelligent processing feedback method program, when the virtual pet emotional intelligent processing feedback method program is executed by a processor, the steps of the virtual pet emotional intelligent processing feedback method of any one of the above are realized.

[0015] From the above, the virtual pet emotional intelligent processing feedback method, system and medium provided by the present application realize the closed-loop processing of virtual pet emotional intelligent feedback through multi-modal feature decoupling, biomimetic memory decay analysis and three-dimensional cooperative feedback.

[0016] Other features and advantages of the present application will be set forth in the following description, and in part will become apparent to those skilled in the art from the description, or can be learned by practice of the present application. The objects and other advantages of the present application can be realized and achieved by the structure particularly pointed out in the written description and the accompanying drawings. BRIEF DESCRIPTION OF DRAWINGS

[0017] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following will briefly introduce the drawings needed to be used in the embodiments of the present application. It should be understood that the following drawings only show some embodiments of the present application, and therefore should not be regarded as limiting the scope. For those skilled in the art, other related drawings can also be obtained without creative labor on the basis of these drawings.

[0018] Figure 1 A flow chart of a virtual pet emotional intelligent processing feedback method provided by the embodiments of the present application; Figure 2 A flow chart of obtaining real-time decoupling feature data set of a virtual pet emotional intelligent processing feedback method provided by the embodiments of the present application; Figure 3A flowchart of a virtual pet emotional intelligence processing feedback method provided by an embodiment of the present application for obtaining user micro-state label data of a virtual pet user and corresponding emotional intensity parameters; Figure 4 A flowchart of a virtual pet emotional intelligence processing feedback method provided by an embodiment of the present application for obtaining memory calling data. DETAILED DESCRIPTION

[0019] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the accompanying drawings of the embodiments of the present application. Obviously, the described embodiments are only some of the embodiments of the present application, but not all the embodiments of the present application. The components of the embodiments of the present application described and shown in the accompanying drawings can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present application provided in the accompanying drawings is not intended to limit the scope of the claimed present application, but only represents selected embodiments of the present application. Based on the embodiments of the present application, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present application.

[0020] It should be noted that similar reference numerals and letters represent similar items in the following drawings, so once an item is defined in one drawing, it does not need to be further defined and explained in subsequent drawings. Meanwhile, in the description of the present application, the terms “first”, “second”, etc. are only used to distinguish the description, and cannot be understood as indicating or implying relative importance.

[0021] Please refer to Figure 1 , Figure 1 is a flowchart of a virtual pet emotional intelligence processing feedback method in some embodiments of the present application. The virtual pet emotional intelligence processing feedback method is used in a terminal device, such as a computer, a mobile phone terminal, etc. The virtual pet emotional intelligence processing feedback method includes the following steps: S11, obtaining real-time feature data set of a virtual pet user, and performing feature decoupling preprocessing to obtain real-time decoupling feature data set; S12, obtaining scene evaluation data of the virtual pet user, and processing according to the scene evaluation data in combination with the real-time decoupling feature data set to obtain user micro-state label data of the virtual pet user and corresponding emotional intensity parameters; S13, processing according to the user micro-state label data and corresponding emotional intensity parameters through a preset virtual pet bionic memory library to obtain memory calling data; S14, processing according to the user micro-state label data and memory calling data through a preset virtual pet emotional intelligence processing model to obtain collaborative feedback output data of the virtual pet; S15, control the virtual pet to execute according to the cooperative feedback output data, and obtain the feedback feature data set of the user to carry out feature decoupling preprocessing, and obtain the feedback decoupling feature data set; S16, according to the feedback decoupling feature data set, process in combination with the scene evaluation data, obtain the user micro-state feedback label data of the virtual pet user and the corresponding feedback emotional intensity parameter; S17, according to the user micro-state feedback label data and the corresponding feedback emotional intensity parameter and the user micro-state label data and the corresponding emotional intensity parameter, process, obtain the virtual pet emotional processing effective parameter, process the virtual pet emotional processing effective parameter and the preset virtual pet emotional processing effective threshold, and obtain the virtual pet emotional processing state.

[0022] It should be noted that, in order to realize the intelligent feedback of virtual pet emotional processing, first, the non-redundant real-time feature data set of the user is collected through the intelligent device of the user, and after preprocessing, the real-time decoupling feature data set is obtained, which is used to represent the semantic, physiological and behavior characteristics of the user, then, the emotion of the user is quantified in combination with the scene of the user, the user micro-state label data and the corresponding emotional intensity parameter are obtained, and then based on the principle of bionic memory decay, the matching memory calling data is called through the preset virtual pet bionic memory library, the cooperative feedback output data of the virtual pet is determined according to the user micro-state label data and the memory calling data, and the virtual pet is controlled to execute the output data, at the same time, the feedback feature data set of the user is collected, the user micro-state feedback label data before execution and the corresponding feedback emotional intensity parameter are analyzed, the virtual pet emotional processing effective parameter is determined through the pre-training model evaluation, and finally, whether the virtual pet emotional processing state is effective is determined through threshold comparison.

[0023] Please refer to Figure 2 , Figure 2 is a flow chart of obtaining real-time decoupling feature data set in a virtual pet emotional intelligent processing feedback method in some embodiments of the present application. According to the embodiment of the present application, the real-time feature data set of the virtual pet user is obtained, and feature decoupling preprocessing is carried out to obtain the real-time decoupling feature data set, which comprises: S21, obtaining the real-time feature data set of the virtual pet user, including semantic feature data, physiological feature data and behavior feature data; S22, the semantic feature data, physiological feature data and behavior feature data are respectively preprocessed by feature decoupling, and the real-time decoupling feature data set is obtained, including semantic decoupling feature data, physiological decoupling feature data and behavior decoupling feature data.

[0024] It should be noted that, in order to reduce the amount of data processing, first, the non-redundant features including semantic feature data, physiological feature data and behavior feature data are collected through the smart device of the user, such as a bracelet, a smart speaker, and then the collected real-time feature data set is processed for feature decoupling to obtain a real-time decoupling feature data set including semantic decoupling feature data, physiological decoupling feature data and behavior decoupling feature data, wherein the semantic decoupling feature data includes emotional keywords (such as sadness, joy) and semantic relationship type feature data (semantic relationship such as transition relationship, semantic relationship type feature data is represented by different identifiers), the physiological decoupling feature data includes heart rate variability frequency domain parameters (used to evaluate autonomic nervous regulation function) and skin electrical activity peak value (which refers to the peak value of skin conductance response, a golden indicator for emotion monitoring), the behavior decoupling feature data includes limb movement amplitude (such as finger tremor frequency of 3 times per second when anxious) and micro-expression dynamic change feature data (such as mouth corner lifting speed, pupil dilation rate).

[0025] Please refer to Figure 3 , Figure 3 is a flowchart of a virtual pet emotional intelligence processing feedback method in some embodiments of the present application. According to the embodiment of the present application, the scene evaluation data of the virtual pet user is obtained, and the real-time decoupling feature data set is processed according to the scene evaluation data to obtain the user micro-state label data of the virtual pet user and the corresponding emotional intensity parameters, including: S31, obtaining scene evaluation data of a virtual pet user, including user location data and environmental noise data; S32, inputting the user location data and environmental noise data into a preset user scene recognition model for processing to obtain scene category feature data; S33, querying a preset decoupling feature weight value mapping table according to the scene category feature data to obtain weight values corresponding to the real-time decoupling feature data set, including semantic weight values, physiological weight values and behavior weight values; S34, inputting the semantic decoupling feature data, physiological decoupling feature data and behavior decoupling feature data and the semantic weight values, physiological weight values and behavior weight values into a preset user micro-state evaluation model for processing to obtain user micro-state label data of the virtual pet user and corresponding emotional intensity parameters.

[0026] It should be noted that the contribution of real-time decoupling feature data is different in different scenarios. First, the user location data and environmental noise data of the environment in which the virtual pet user is located are input into a preset user scene recognition model for processing to obtain scene category feature data, such as a quiet home environment, a moving vehicle environment, and a noisy outdoor environment. The scene category feature data is represented by a unique identifier. Then, according to the determined scene category feature data, the corresponding weight value is obtained by querying the preset decoupling feature weight value mapping table, and then the pre-trained preset user micro-state evaluation model is analyzed and processed to obtain the user micro-state label data of the virtual pet user and the corresponding emotional intensity parameters. Among them, the user micro-state is, for example, happy with regret, anxious with expectation, the user micro-state label data is represented by a unique identifier, the preset user scene recognition model is obtained by training a large number of historical samples of user location data and environmental noise data and corresponding scene category feature data, the preset decoupling feature weight value mapping table is pre-constructed by the person skilled in the art according to the historical samples, and can be dynamically adjusted, and the preset user micro-state evaluation model is obtained by training a large number of historical samples of semantic decoupling feature data, physiological decoupling feature data and behavior decoupling feature data combined with semantic weight value, physiological weight value and behavior weight value, and corresponding user micro-state label data and emotional intensity parameters.

[0027] Please refer to Figure 4 , Figure 4 is a flowchart of a virtual pet emotional intelligence processing feedback method in some embodiments of the present application. According to the embodiment of the present application, the memory call data is obtained by processing the user micro-state label data and the corresponding emotional intensity parameters through the preset virtual pet bionic memory bank, including: S41, according to the user micro-state label data and the corresponding emotional intensity parameters, the memory call initial data set is obtained by matching processing through the preset virtual pet bionic memory bank; S42, the real-time memory strength, the number of calls and the scene correlation degree of the memory call initial data in the memory call initial data set are obtained; S43, the real-time memory strength, the number of calls and the scene correlation degree are weighted and summed to obtain the matching value corresponding to the memory call initial data; S44, the matching value is arranged in descending order, and the memory call initial data with the largest matching value is determined as the memory call data.

[0028] It should be noted that, first, based on the determined user micro-state label data and the corresponding emotional intensity parameters, data matching processing is performed from the preset virtual pet bionic memory library to obtain a memory calling initial data set. In order to determine the most suitable memory calling data, the real-time memory intensity, the calling frequency and the scene correlation degree of the memory calling initial data are obtained. The real-time memory intensity is obtained by evaluating the preset initial memory intensity, the preset memory attenuation coefficient, the memory repetition reference frequency, the memory storage emotional intensity parameter and the memory storage time length by the person skilled in the art. The calling frequency refers to the number of times that the memory calling initial data is called within a preset time period. The scene correlation degree is 0-100%, which is evaluated by the person skilled in the art according to the user micro-state label data when the memory is generated and called. For example, if the user micro-state label data when the memory is generated and called is happy, the scene correlation degree is 100%. The real-time memory intensity, the calling frequency and the scene correlation degree are normalized and weighted summed to obtain a matching value corresponding to the memory calling initial data. The memory calling initial data with the maximum matching value is determined as the memory calling data.

[0029] According to the embodiment of the present application, the user micro-state label data and the memory calling data are input into the preset virtual pet emotional intelligent processing model for processing to obtain the collaborative feedback output data of the virtual pet. The user micro-state label data and the memory calling data are input into the preset virtual pet emotional intelligent processing model for processing to obtain the collaborative feedback output data of the virtual pet. The collaborative feedback output data includes digital end feedback output data, physical end feedback output data and service end feedback output data.

[0030] It should be noted that, wherein the preset virtual pet emotional intelligent processing model is obtained by training a large number of historical sample user micro-state label data and memory calling data and corresponding collaborative feedback output data. The digital end feedback output data includes visual and auditory output. The visual output is, for example, that the APP interface automatically switches from the default daily accompanying scene to the relaxation scene, and the virtual pet performs slow blinking and tail sweeping actions. The auditory output is, for example, that the synthetic voice matching the user's sound line is adopted, and white noise is played synchronously. The physical end feedback output data includes real tactile output of touch and body sense, such as when the digital end plays white noise, the micro-vibration motor built in the pendant vibrates continuously at a frequency of 30Hz and an amplitude of 0.1mm, simulating the touch of a light pat on the back of the hand. The service end feedback output data is, for example, that the APP bottom folding bar displays relaxation recommendations, the left side is a 10-minute overtime relaxation meditation course, and the right side is a nearby 24-hour coffee shop coupon of Meituan, realizing precise and scene-based output. The three types of collaborative feedback output data realize time synchronization, emotional linkage and user-led collaboration through the emotional main line.

[0031] According to the embodiment of the present application, the virtual pet is controlled to execute according to the cooperative feedback output data, and a feedback feature data set of the user is acquired for feature decoupling preprocessing to obtain a feedback decoupling feature data set, including: The virtual pet is controlled to execute according to the cooperative feedback output data, and a feedback feature data set of the user is acquired, including semantic feedback feature data, physiological feedback feature data and behavior feedback feature data; The semantic feedback feature data, the physiological feedback feature data and the behavior feedback feature data are respectively subjected to feature decoupling preprocessing to obtain a feedback decoupling feature data set, including semantic feedback decoupling feature data, physiological feedback decoupling feature data and behavior feedback decoupling feature data.

[0032] It should be noted that after the virtual pet executes according to the cooperative feedback output data, the feedback feature data set of the user is acquired in real time, and feature decoupling preprocessing is performed to obtain a feedback decoupling feature data set.

[0033] According to the embodiment of the present application, the user micro-state feedback label data and the corresponding feedback emotional intensity parameters and the user micro-state label data and the corresponding emotional intensity parameters are processed to obtain virtual pet emotional processing effective parameters, the virtual pet emotional processing effective parameters are processed with preset virtual pet emotional processing effective thresholds to obtain a virtual pet emotional processing state, including: The user micro-state feedback label data and the corresponding feedback emotional intensity parameters and the user micro-state label data and the corresponding emotional intensity parameters are input into a preset virtual pet emotional processing evaluation model for processing to obtain virtual pet emotional processing effective parameters; The virtual pet emotional processing effective parameters are processed with preset virtual pet emotional processing effective thresholds to obtain a virtual pet emotional processing state; If the virtual pet emotional processing effective parameters are less than the preset virtual pet emotional processing effective thresholds, it is determined that the virtual pet emotional processing state is an invalid state; If the virtual pet emotional processing effective parameters are greater than or equal to the preset virtual pet emotional processing effective thresholds, it is determined that the virtual pet emotional processing state is an effective state.

[0034] It should be noted that, in order to evaluate whether the virtual pet emotion processing is effective, the user micro-state feedback label data and the corresponding feedback emotion intensity parameters and the user micro-state label data and the corresponding emotion intensity parameters after the virtual pet executes the action are processed through a preset virtual pet emotion processing evaluation model to obtain virtual pet emotion processing effective parameters, and then the threshold is compared to evaluate whether the virtual pet emotion processing state is effective. The preset virtual pet emotion processing evaluation model is trained by obtaining a large number of historical sample user micro-state feedback label data and corresponding feedback emotion intensity parameters and user micro-state label data and corresponding emotion intensity parameters and corresponding virtual pet emotion processing effective parameters.

[0035] According to the embodiments of the present application, the method further comprises: obtaining a preset initial memory strength, a preset memory decay coefficient, a memory repetition mentioning number, a memory storage emotion intensity parameter and a memory storage time length of the memory calling initial data in the memory calling initial data set; performing weighted sum processing on the memory repetition mentioning number and the memory storage emotion intensity parameter to obtain a memory decay coefficient influence factor; correcting the preset memory decay coefficient according to the memory decay coefficient influence factor to obtain a memory decay correction coefficient; processing the preset initial memory strength and the memory storage time length according to the memory decay correction coefficient to obtain a real-time memory strength.

[0036] It should be noted that, in order to accurately evaluate the real-time memory strength of the memory calling initial data, first, the memory repetition mentioning number in a preset time period and the memory storage emotion intensity parameter at the memory storage time are normalized and then weighted and summed to obtain a memory decay coefficient influence factor. Then, the preset memory decay coefficient is corrected according to the memory decay coefficient influence factor to obtain a memory decay correction coefficient. For example, if the memory decay coefficient influence factor is 0.2 and the preset memory decay coefficient is 0.3, then 0.3*(1-0.2)=0.24 is the memory decay correction coefficient. Finally, the preset initial memory strength and the memory storage time length (in days) are processed according to the memory decay correction coefficient to obtain a real-time memory strength. For example, if the preset initial memory strength is a and the memory storage time length is t, then a*e (-0.24*t) is the real-time memory strength. The preset initial memory strength and the preset memory decay coefficient are pre-constructed by a person skilled in the art according to specific applications and can be dynamically adjusted.

[0037] The application further discloses a virtual pet emotional intelligence processing feedback system, comprising a memory and a processor, wherein the memory comprises a virtual pet emotional intelligence processing feedback method program, and the virtual pet emotional intelligence processing feedback method program is executed by the processor to realize the following steps: Real-time feature data set of the virtual pet user is acquired, and feature decoupling preprocessing is performed to obtain real-time decoupling feature data set; Scene evaluation data of the virtual pet user is acquired, and the real-time decoupling feature data set is processed according to the scene evaluation data to obtain user micro-state label data and corresponding emotional intensity parameters of the virtual pet user; Memory calling data is obtained by processing the user micro-state label data and the corresponding emotional intensity parameters through a preset virtual pet bionic memory bank; Cooperative feedback output data of the virtual pet is obtained by processing the user micro-state label data and the memory calling data through a preset virtual pet emotional intelligence processing model; The virtual pet is controlled according to the cooperative feedback output data, and feedback feature data set of the user is acquired for feature decoupling preprocessing to obtain feedback decoupling feature data set; User micro-state feedback label data and corresponding feedback emotional intensity parameters of the virtual pet user are obtained by processing the feedback decoupling feature data set in combination with the scene evaluation data; Virtual pet emotional processing effective parameters are obtained by processing the user micro-state feedback label data and the corresponding feedback emotional intensity parameters and the user micro-state label data and the corresponding emotional intensity parameters, the virtual pet emotional processing effective parameters are processed in combination with preset virtual pet emotional processing effective threshold values, and a virtual pet emotional processing state is obtained.

[0038] It should be noted that, in order to realize intelligent feedback of virtual pet emotional processing, first, non-redundant real-time feature data set is collected through a user's intelligent device, and after preprocessing, real-time decoupling feature data set is obtained, which is used to represent semantic, physiological and behavioral characteristics of the user, then, the user's emotion is quantified in combination with a scene in which the user is located to obtain user micro-state label data and corresponding emotional intensity parameters, and then, based on the bionic memory attenuation principle, matching memory calling data is called through a preset virtual pet bionic memory bank constructed in advance, cooperative feedback output data of the virtual pet is determined according to the user micro-state label data and the memory calling data, and the virtual pet is controlled to output the data, meanwhile, feedback feature data set of the user is collected, the user micro-state feedback label data and the corresponding feedback emotional intensity parameters before execution are analyzed in combination, virtual pet emotional processing effective parameters are determined through a pre-training model, and finally, whether the virtual pet emotional processing state is effective is determined through threshold comparison.

[0039] According to the embodiment of the present application, the real-time feature data set of the virtual pet user is obtained, and feature decoupling preprocessing is performed to obtain a real-time decoupling feature data set, comprising: The real-time feature data set of the virtual pet user is obtained, including semantic feature data, physiological feature data and behavior feature data; The semantic feature data, physiological feature data and behavior feature data are respectively subjected to feature decoupling preprocessing to obtain a real-time decoupling feature data set, including semantic decoupling feature data, physiological decoupling feature data and behavior decoupling feature data.

[0040] It should be noted that, in order to reduce the amount of data processing, first, the non-redundant features including semantic feature data, physiological feature data and behavior feature data are collected by the user's smart device, such as a bracelet, a smart speaker, and then the collected real-time feature data set is subjected to feature decoupling processing to obtain a real-time decoupling feature data set including semantic decoupling feature data, physiological decoupling feature data and behavior decoupling feature data, wherein the semantic decoupling feature data includes emotional keywords (such as sadness, joy) and semantic relationship type feature data (semantic relationship such as transition relationship, semantic relationship type feature data is represented by different identifiers), the physiological decoupling feature data includes heart rate variability frequency domain parameters (used to evaluate autonomic nervous regulation function) and skin electrical activity peak value (which refers to the peak value of skin conductance response, a gold indicator for emotion monitoring), the behavior decoupling feature data includes limb movement amplitude (such as finger tremor frequency of 3 times per second when anxious) and micro-expression dynamic change feature data (such as mouth corner lifting speed, pupil dilation rate).

[0041] According to the embodiment of the present application, the scene evaluation data of the virtual pet user is obtained, and the real-time decoupling feature data set is processed according to the scene evaluation data to obtain user micro-state label data and corresponding emotional intensity parameters of the virtual pet user, comprising: The scene evaluation data of the virtual pet user is obtained, including user location data and environmental noise data; The user location data and environmental noise data are input into a preset user scene recognition model for processing to obtain scene category feature data; According to the scene category feature data, a preset decoupling feature weight value mapping table is queried to obtain weight values corresponding to the real-time decoupling feature data set, including semantic weight values, physiological weight values and behavior weight values; The semantic decoupling feature data, physiological decoupling feature data and behavior decoupling feature data, and the semantic weight values, physiological weight values and behavior weight values are input into a preset user micro-state evaluation model for processing to obtain user micro-state label data and corresponding emotional intensity parameters of the virtual pet user.

[0042] It should be noted that the contribution of real-time decoupling feature data is different in different scenarios. First, the user location data and environmental noise data of the environment in which the virtual pet user is located are input into a preset user scene recognition model for processing to obtain scene category feature data, such as a quiet home environment, a moving vehicle environment, and a noisy outdoor environment. The scene category feature data is represented by a unique identifier. Then, according to the determined scene category feature data, the corresponding weight value is obtained by querying the preset decoupling feature weight value mapping table, and then the pre-trained preset user micro-state evaluation model is analyzed and processed to obtain the user micro-state label data of the virtual pet user and the corresponding emotional intensity parameters. The user micro-state is, for example, happy with regret, anxious with expectation. The user micro-state label data is represented by a unique identifier. The preset user scene recognition model is trained by obtaining a large number of historical sample user location data and environmental noise data and corresponding scene category feature data. The preset decoupling feature weight value mapping table is pre-constructed by the person skilled in the art according to the historical samples, and can be dynamically adjusted. The preset user micro-state evaluation model is trained by obtaining a large number of historical sample semantic decoupling feature data, physiological decoupling feature data and behavior decoupling feature data, combining semantic weight value, physiological weight value and behavior weight value, and corresponding user micro-state label data and emotional intensity parameters.

[0043] According to the embodiment of the application, the user micro-state label data and the corresponding emotional intensity parameters are processed by a preset virtual pet bionic memory bank to obtain memory call data, including: According to the user micro-state label data and the corresponding emotional intensity parameters, the memory call initial data set is obtained by matching processing through the preset virtual pet bionic memory bank. Obtain the real-time memory intensity, call frequency and scene correlation degree of the memory call initial data in the memory call initial data set; The real-time memory intensity, call frequency and scene correlation degree are weighted and summed to obtain the matching value corresponding to the memory call initial data; The matching value is arranged in descending order, and the memory call initial data with the largest matching value is determined as the memory call data.

[0044] It should be noted that, first, based on the determined user micro-state label data and the corresponding emotional intensity parameters, data matching processing is performed from the preset virtual pet bionic memory library to obtain a memory calling initial data set. In order to determine the most suitable memory calling data, the real-time memory intensity, the calling frequency and the scene correlation degree of the memory calling initial data are obtained. The real-time memory intensity is obtained by evaluating the preset initial memory intensity, the preset memory attenuation coefficient, the memory repetition reference frequency, the memory storage emotional intensity parameter and the memory storage time length by the person skilled in the art. The calling frequency refers to the number of times that the memory calling initial data is called within a preset time period. The scene correlation degree is 0-100%, which is evaluated by the person skilled in the art according to the user micro-state label data when the memory is generated and called. For example, if the user micro-state label data when the memory is generated and called is happy, the scene correlation degree is 100%. The real-time memory intensity, the calling frequency and the scene correlation degree are normalized and weighted summed to obtain a matching value corresponding to the memory calling initial data. The memory calling initial data with the maximum matching value is determined as the memory calling data.

[0045] According to the embodiment of the present application, the user micro-state label data and the memory calling data are input into the preset virtual pet emotional intelligent processing model for processing to obtain the collaborative feedback output data of the virtual pet. The user micro-state label data and the memory calling data are input into the preset virtual pet emotional intelligent processing model for processing to obtain the collaborative feedback output data of the virtual pet. The collaborative feedback output data includes digital end feedback output data, physical end feedback output data and service end feedback output data.

[0046] It should be noted that, wherein the preset virtual pet emotional intelligent processing model is obtained by training a large number of historical sample user micro-state label data and memory calling data and corresponding collaborative feedback output data. The digital end feedback output data includes visual and auditory output. The visual output is, for example, that the APP interface automatically switches from the default daily accompanying scene to the relaxation scene, and the virtual pet performs slow blinking and tail sweeping actions. The auditory output is, for example, that the synthetic voice matching the user's sound line is adopted, and white noise is played synchronously. The physical end feedback output data includes real tactile output of touch and body sense, such as when the digital end plays white noise, the micro-vibration motor built in the pendant vibrates continuously at a frequency of 30Hz and an amplitude of 0.1mm, simulating the touch of a light pat on the back of the hand. The service end feedback output data is, for example, that the APP bottom folding bar displays relaxation recommendations, the left side is a 10-minute overtime relaxation meditation course, and the right side is a nearby 24-hour coffee shop coupon of Meituan, realizing precise and scene-based output. The three types of collaborative feedback output data realize time synchronization, emotional linkage and user-led collaboration through the emotional main line.

[0047] According to the embodiment of the present application, the virtual pet is controlled to execute according to the cooperative feedback output data, and a feedback feature data set of the user is acquired for feature decoupling preprocessing to obtain a feedback decoupling feature data set, including: The virtual pet is controlled to execute according to the cooperative feedback output data, and a feedback feature data set of the user is acquired, including semantic feedback feature data, physiological feedback feature data and behavior feedback feature data; The semantic feedback feature data, the physiological feedback feature data and the behavior feedback feature data are respectively subjected to feature decoupling preprocessing to obtain a feedback decoupling feature data set, including semantic feedback decoupling feature data, physiological feedback decoupling feature data and behavior feedback decoupling feature data.

[0048] It should be noted that after the virtual pet executes according to the cooperative feedback output data, the feedback feature data set of the user is acquired in real time, and feature decoupling preprocessing is performed to obtain a feedback decoupling feature data set.

[0049] According to the embodiment of the present application, the user micro-state feedback label data and the corresponding feedback emotional intensity parameters and the user micro-state label data and the corresponding emotional intensity parameters are processed to obtain virtual pet emotional processing effective parameters, the virtual pet emotional processing effective parameters are processed with preset virtual pet emotional processing effective thresholds to obtain a virtual pet emotional processing state, including: The user micro-state feedback label data and the corresponding feedback emotional intensity parameters and the user micro-state label data and the corresponding emotional intensity parameters are input into a preset virtual pet emotional processing evaluation model for processing to obtain virtual pet emotional processing effective parameters; The virtual pet emotional processing effective parameters are processed with preset virtual pet emotional processing effective thresholds to obtain a virtual pet emotional processing state; If the virtual pet emotional processing effective parameters are less than the preset virtual pet emotional processing effective thresholds, it is determined that the virtual pet emotional processing state is an invalid state; If the virtual pet emotional processing effective parameters are greater than or equal to the preset virtual pet emotional processing effective thresholds, it is determined that the virtual pet emotional processing state is an effective state.

[0050] It should be noted that, in order to evaluate whether the virtual pet emotion processing is effective, the user micro-state feedback label data and the corresponding feedback emotion intensity parameters and the user micro-state label data and the corresponding emotion intensity parameters after the virtual pet executes the action are processed through a preset virtual pet emotion processing evaluation model to obtain virtual pet emotion processing effective parameters, and then the threshold is compared to evaluate whether the virtual pet emotion processing state is effective. The preset virtual pet emotion processing evaluation model is trained by obtaining a large number of historical sample user micro-state feedback label data and corresponding feedback emotion intensity parameters and user micro-state label data and corresponding emotion intensity parameters and corresponding virtual pet emotion processing effective parameters.

[0051] According to the embodiments of the present application, the method further comprises: obtaining a preset initial memory strength, a preset memory decay coefficient, a memory repetition mentioning number, a memory storage emotion intensity parameter and a memory storage time length of the memory calling initial data in the memory calling initial data set; performing weighted sum processing on the memory repetition mentioning number and the memory storage emotion intensity parameter to obtain a memory decay coefficient influence factor; correcting the preset memory decay coefficient according to the memory decay coefficient influence factor to obtain a memory decay correction coefficient; processing the preset initial memory strength and the memory storage time length according to the memory decay correction coefficient to obtain a real-time memory strength.

[0052] It should be noted that, in order to accurately evaluate the real-time memory strength of the memory calling initial data, first, the memory repetition mentioning number and the memory storage emotion intensity parameter at the memory storage time in a preset time period are normalized and then weighted and summed to obtain a memory decay coefficient influence factor. Then, the preset memory decay coefficient is corrected according to the memory decay coefficient influence factor to obtain a memory decay correction coefficient. For example, if the memory decay coefficient influence factor is 0.2 and the preset memory decay coefficient is 0.3, then 0.3*(1-0.2)=0.24 is the memory decay correction coefficient. Finally, the preset initial memory strength and the memory storage time length (in days) are processed according to the memory decay correction coefficient to obtain a real-time memory strength. For example, if the preset initial memory strength is a and the memory storage time length is t, then a*e (-0.24*t) is the real-time memory strength. The preset initial memory strength and the preset memory decay coefficient are pre-constructed by a person skilled in the art according to specific applications and can be dynamically adjusted.

[0053] The third aspect of the present application provides a readable storage medium, wherein a virtual pet emotional intelligence processing feedback method program is stored in the readable storage medium, and the virtual pet emotional intelligence processing feedback method program is executed by a processor to implement the steps of the virtual pet emotional intelligence processing feedback method according to any one of the preceding aspects.

[0054] The virtual pet emotional intelligence processing feedback method, system and medium disclosed by the present application realize the closed-loop processing of virtual pet emotional intelligence feedback through multi-modal feature decoupling, bionic memory decay analysis and three-dimensional collaborative feedback.

[0055] In several embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other manners. The above described device embodiments are merely schematic, for example, the division of the units is merely a logical function division, and there can be another division manner in actual implementation, for example, a plurality of units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the coupling or direct coupling or communication connection between the various components shown or discussed can be indirect coupling or communication connection through some interfaces, devices or units, and can be electrical, mechanical or other forms.

[0056] The units described as separate components can or can not be physically separate, and the components shown as units can or can not be physical units; they can be located in one place, or distributed on a plurality of network units; and some or all of the units can be selected according to actual needs to achieve the purpose of the present embodiment.

[0057] In addition, each functional unit in each embodiment of the present application can be integrated into one processing unit, or each unit can be a separate unit, or two or more units can be integrated into one unit; the integrated unit can be realized in the form of hardware, or in the form of hardware plus software functional unit.

[0058] Those skilled in the art can understand that all or part of the steps of the above method embodiments can be completed by program instruction related hardware, and the foregoing program can be stored in a readable storage medium, and the program is executed to perform the steps of the above method embodiments; and the foregoing storage medium includes mobile storage devices, read-only memories (ROMs), random access memories (RAMs), magnetic discs or optical discs, and various media that can store program codes.

[0059] Alternatively, the above-mentioned integrated unit of the present application, if realized in the form of a software function module and sold or used as an independent product, can also be stored in a readable storage medium. Based on such understanding, the technical solutions of the embodiments of the present application can be embodied in the form of a software product, which is stored in a storage medium, includes several instructions to make a computer device (which can be a personal computer, a server, or a network device, etc.) execute all or part of the methods described in the various embodiments of the present application. The aforementioned storage medium includes: mobile storage devices, ROM, RAM, magnetic or optical disks, and various media that can store program codes.

Claims

1. A method for processing and providing feedback on the emotional intelligence of virtual pets, characterized in that, Includes the following steps: Obtain the real-time feature dataset of virtual pet users and perform feature decoupling preprocessing to obtain the real-time decoupled feature dataset; The scene evaluation data of virtual pet users is obtained, and the scene evaluation data is processed in combination with the real-time decoupled feature dataset to obtain the user micro-state label data of virtual pet users and the corresponding emotional intensity parameters. Based on the user's micro-state tag data and the corresponding emotional intensity parameters, the data is processed through a preset virtual pet bionic memory bank to obtain memory retrieval data; The user's micro-state tag data and memory recall data are processed through a preset virtual pet emotional intelligence processing model to obtain the virtual pet's collaborative feedback output data. The virtual pet is controlled to execute based on the collaborative feedback output data, and the user's feedback feature dataset is obtained for feature decoupling preprocessing to obtain the feedback decoupling feature dataset; The feedback decoupling feature dataset is processed in conjunction with the scene evaluation data to obtain the user micro-state feedback tag data of virtual pet users and the corresponding feedback emotional intensity parameters. The user's micro-state feedback tag data and corresponding feedback emotion intensity parameters are processed to obtain effective parameters for virtual pet emotion processing. The effective parameters for virtual pet emotion processing are then processed with a preset effective threshold for virtual pet emotion processing to obtain the virtual pet emotion processing status.

2. The virtual pet emotional intelligence processing and feedback method according to claim 1, characterized in that, The process of acquiring the real-time feature dataset of virtual pet users and performing feature decoupling preprocessing to obtain the real-time decoupled feature dataset includes: Obtain real-time feature datasets of virtual pet users, including semantic feature data, physiological feature data, and behavioral feature data; The semantic feature data, physiological feature data, and behavioral feature data are respectively subjected to feature decoupling preprocessing to obtain a real-time decoupled feature dataset, including semantic decoupled feature data, physiological decoupled feature data, and behavioral decoupled feature data.

3. The virtual pet emotional intelligence processing and feedback method according to claim 2, characterized in that, The process of acquiring scene evaluation data of virtual pet users, and processing the scene evaluation data in conjunction with the real-time decoupled feature dataset, yields user micro-state tag data and corresponding emotional intensity parameters for virtual pet users, including: Acquire scene evaluation data of virtual pet users, including user location data and environmental noise data; The user location data and environmental noise data are input into a preset user scene recognition model for processing to obtain scene category feature data; Based on the scenario category feature data, query the preset decoupling feature weight value mapping table to obtain the weight values ​​corresponding to the real-time decoupling feature dataset, including semantic weight values, physiological weight values ​​and behavioral weight values; The semantic decoupling feature data, physiological decoupling feature data, and behavioral decoupling feature data, along with the semantic weight value, physiological weight value, and behavioral weight value, are input into a preset user micro-state evaluation model for processing to obtain the user micro-state label data of the virtual pet user and the corresponding emotional intensity parameters.

4. The virtual pet emotional intelligence processing and feedback method according to claim 3, characterized in that, The process of obtaining memory retrieval data by processing the user's micro-state tag data and corresponding emotional intensity parameters through a preset virtual pet bionic memory bank includes: Based on the user's micro-state tag data and the corresponding emotional intensity parameters, the data is matched and processed through a preset virtual pet bionic memory library to obtain the initial dataset for memory retrieval; Obtain the real-time memory strength, number of calls, and scene relevance of the initial memory call data in the initial memory call dataset; The real-time memory strength, number of calls, and scene relevance are weighted and summed to obtain the matching value corresponding to the initial data of memory call; The matching values ​​are sorted in descending order, and the initial data for memory retrieval with the largest matching value is determined as the memory retrieval data.

5. The virtual pet emotional intelligence processing and feedback method according to claim 4, characterized in that, The process of processing the user's micro-state tag data and memory retrieval data using a preset virtual pet emotional intelligence processing model to obtain the virtual pet's collaborative feedback output data includes: The user's micro-state tag data and memory recall data are input into a preset virtual pet emotional intelligence processing model for processing to obtain the virtual pet's collaborative feedback output data; The collaborative feedback output data includes digital terminal feedback output data, physical terminal feedback output data, and server terminal feedback output data.

6. The virtual pet emotional intelligence processing and feedback method according to claim 5, characterized in that, The step of controlling the virtual pet to execute based on the collaborative feedback output data, and obtaining the user's feedback feature dataset for feature decoupling preprocessing to obtain the feedback decoupling feature dataset includes: The virtual pet is controlled to perform actions based on the collaborative feedback output data, and the user's feedback feature dataset is obtained, including semantic feedback feature data, physiological feedback feature data, and behavioral feedback feature data. The semantic feedback feature data, physiological feedback feature data, and behavioral feedback feature data are respectively subjected to feature decoupling preprocessing to obtain a feedback decoupling feature dataset, including semantic feedback decoupling feature data, physiological feedback decoupling feature data, and behavioral feedback decoupling feature data.

7. The virtual pet emotional intelligence processing and feedback method according to claim 6, characterized in that, The process of processing the user's micro-state feedback tag data and corresponding feedback emotion intensity parameters to obtain effective parameters for virtual pet emotion processing, and then processing these effective parameters with a preset effective threshold for virtual pet emotion processing to obtain the virtual pet's emotion processing state, includes: The user micro-state feedback tag data and the corresponding feedback emotion intensity parameters are input into a preset virtual pet emotion processing evaluation model for processing to obtain effective parameters for virtual pet emotion processing. The effective parameters for virtual pet emotion processing are processed with the preset effective threshold for virtual pet emotion processing to obtain the virtual pet emotion processing status; If the effective parameter of the virtual pet emotion processing is less than the preset effective threshold of the virtual pet emotion processing, the virtual pet emotion processing status is determined to be invalid. If the effective parameter of virtual pet emotion processing is greater than or equal to the preset effective threshold of virtual pet emotion processing, then the virtual pet emotion processing status is determined to be effective.

8. The virtual pet emotional intelligence processing and feedback method according to claim 7, characterized in that, Also includes: Obtain the preset initial memory strength, preset memory decay coefficient, number of repeated mentions of memory, memory storage emotional intensity parameter, and memory storage duration of the memory retrieval initial data in the memory retrieval initial dataset; The memory decay coefficient influencing factor is obtained by weighted summation of the number of times the memory is repeatedly mentioned and the emotional intensity parameter of the memory storage. The preset memory decay coefficient is corrected based on the memory decay coefficient influence factor to obtain the memory decay correction coefficient; The real-time memory strength is obtained by processing the memory decay correction coefficient in combination with the preset initial memory strength and memory storage duration.

9. A virtual pet emotional intelligence processing and feedback system, characterized in that, It includes a memory and a processor. The memory contains a program for a virtual pet emotional intelligence processing feedback method. When the program for the virtual pet emotional intelligence processing feedback method is executed by the processor, it performs the following steps: Obtain the real-time feature dataset of virtual pet users and perform feature decoupling preprocessing to obtain the real-time decoupled feature dataset; The scene evaluation data of virtual pet users is obtained, and the scene evaluation data is processed in combination with the real-time decoupled feature dataset to obtain the user micro-state label data of virtual pet users and the corresponding emotional intensity parameters. Based on the user's micro-state tag data and the corresponding emotional intensity parameters, the data is processed through a preset virtual pet bionic memory bank to obtain memory retrieval data; The user's micro-state tag data and memory recall data are processed through a preset virtual pet emotional intelligence processing model to obtain the virtual pet's collaborative feedback output data. The virtual pet is controlled to execute based on the collaborative feedback output data, and the user's feedback feature dataset is obtained for feature decoupling preprocessing to obtain the feedback decoupling feature dataset; The feedback decoupling feature dataset is processed in conjunction with the scene evaluation data to obtain the user micro-state feedback tag data of virtual pet users and the corresponding feedback emotional intensity parameters. The user's micro-state feedback tag data and corresponding feedback emotion intensity parameters are processed to obtain effective parameters for virtual pet emotion processing. The effective parameters for virtual pet emotion processing are then processed with a preset effective threshold for virtual pet emotion processing to obtain the virtual pet emotion processing status.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a virtual pet emotional intelligence processing feedback method program, which, when executed by a processor, implements the steps of a virtual pet emotional intelligence processing feedback method as described in any one of claims 1 to 8.

Images