Satisfaction perception method and device, electronic equipment and storage medium

Abstract

The invention provides a satisfaction perception method and device, electronic equipment and a storage medium, and the method comprises the steps: obtaining user control data and / or equipment pose data when equipment is in a running state, the user control data representing behaviors when a user controls the equipment, and the equipment pose data representing pose changes of the equipment under the control of the user; and performing user satisfaction perception based on the user control data and / or the equipment pose data to obtain satisfaction information of the user. According to the method, the device, the electronic equipment and the storage medium provided by the invention, the user control data and / or the equipment pose data of the equipment in the running state are / is actively acquired to perform satisfaction perception, so that a user does not need to execute additional operation except for controlling and using the equipment; therefore, the satisfaction degree of the user to the equipment in the process of operating and using the equipment can be dynamically obtained in real time, and conditions are provided for updating and optimizing the equipment, improving the satisfaction degree of the user to the equipment and optimizing the experience feeling of the user.

Description

Technical Field

[0001] This invention relates to the field of artificial intelligence technology, and in particular to a method, apparatus, electronic device, and storage medium for perceiving satisfaction. Background Technology

[0002] Currently, most terminal products are equipped with a satisfaction feedback mechanism. When users are dissatisfied with certain functions of the terminal product or wish to make suggestions, they can directly submit relevant feedback information through the terminal product's built-in operation interface.

[0003] However, the aforementioned satisfaction feedback mechanism is cumbersome to operate and relies excessively on user initiative. In practical applications, due to its inconvenience, user enthusiasm for feedback is usually low, and manufacturers can only collect a very limited amount of satisfaction feedback information, which directly limits the targeted improvement and optimization of end products. Summary of the Invention

[0004] This invention provides a satisfaction perception method, apparatus, electronic device, and storage medium to address the shortcomings of related technologies in satisfaction feedback mechanisms, which are cumbersome to operate and overly reliant on user initiative.

[0005] This invention provides a method for perceiving satisfaction, comprising: When the device is in operation, user control data and / or device pose data are acquired. The user control data represents the user's behavior when controlling the device, and the device pose data represents the pose changes of the device under user control. User satisfaction is perceived based on the user operation data and / or the device pose data to obtain user satisfaction information.

[0006] A satisfaction perception method provided by the present invention further includes: When the device is in operation, user emotion data and / or human-computer interaction data are acquired; The step of perceiving user satisfaction based on the user operation data and / or the device pose data to obtain user satisfaction information includes: Based on user emotion data and / or human-computer interaction data, as well as the user control data and / or the device pose data, user satisfaction is perceived to obtain user satisfaction information.

[0007] According to a satisfaction perception method provided by the present invention, the step of perceiving user satisfaction based on user emotion data and / or human-computer interaction data, and the user operation data and / or the device pose data, to obtain user satisfaction information, includes: The user emotion data and / or human-computer interaction data, as well as the user operation data and / or the device pose data, are input into the satisfaction perception model to obtain the satisfaction information output by the satisfaction perception model. The satisfaction information includes the satisfaction level and the contribution weight of various data to the satisfaction level. The satisfaction perception model is trained based on various types of sample data, the satisfaction labels of the various types of sample data, and a feature importance attribution mechanism.

[0008] A satisfaction perception method provided by the present invention further includes: If the satisfaction level in the satisfaction information is lower than the level threshold, the low satisfaction data type is determined based on the contribution weight of each type of data in the satisfaction information to the satisfaction level. Based on the aforementioned low satisfaction data, the causes of low satisfaction are identified.

[0009] A satisfaction perception method provided by the present invention further includes: Determine the optimization strategy corresponding to the causes of low satisfaction; Based on the optimization strategy, the device is optimized for application use.

[0010] According to a satisfaction perception method provided by the present invention, the step of optimizing the device based on the optimization strategy further includes: Perform feedback interaction with the user to obtain the user's optimization feedback information; If the optimization feedback is positive, the user's preference model is updated based on the optimization strategy, and the preference model is used to update the application of the device.

[0011] According to a satisfaction perception method provided by the present invention, the user operation data includes grip pressure information and / or scanning sliding pressure information; the device pose data includes at least one of scanning trajectory, repeated scanning action, and picking up confirmation action.

[0012] The present invention also provides a satisfaction sensing device, comprising: The acquisition unit is used to acquire user control data and / or device pose data when the device is in operation. The user control data represents the user's behavior when controlling the device, and the device pose data represents the pose changes of the device under user control. The sensing unit is used to perceive user satisfaction based on the user operation data and / or the device pose data, and obtain user satisfaction information.

[0013] 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 executes the program to implement the satisfaction perception method as described above.

[0014] 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 satisfaction perception method as described above.

[0015] The present invention also provides a computer program product, including a computer program that, when executed by a processor, implements the satisfaction perception method as described above.

[0016] The satisfaction perception method, apparatus, electronic device, and storage medium provided by this invention actively acquire user operation data and / or device pose data during device operation, and perform satisfaction perception based on the user operation data and / or device pose data. This allows for dynamic and real-time acquisition of user satisfaction with the device during operation without requiring the user to perform any additional operations beyond using the device. This provides conditions for device updates, optimizations, improving user satisfaction with the device, and enhancing the user experience. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in this invention or related technologies, the accompanying drawings used in the description of the embodiments or related technologies will be briefly introduced below. Obviously, the accompanying drawings described below are some embodiments of this invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 This is a flowchart illustrating the satisfaction perception method provided by the present invention.

[0019] Figure 2 This is a schematic diagram of the satisfaction perception device provided by the present invention.

[0020] Figure 3 This is a schematic diagram of the structure of the electronic device provided by the present invention. Detailed Implementation

[0021] 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.

[0022] All actions involving the acquisition of signal information or data in this invention are carried out in compliance with the relevant data protection laws and policies of the country where the device is located, and with the authorization granted by the owner of the relevant device.

[0023] To understand user feedback, many consumer products, such as smartphones and smartwatches, are equipped with satisfaction feedback mechanisms. Typically, when users are dissatisfied with certain functions of a product or wish to provide opinions and suggestions, they can input and submit their feedback through the product's built-in interface. This allows manufacturers to obtain user feedback information.

[0024] Although the current satisfaction feedback mechanism can clearly reflect user satisfaction and even collect specific opinions and suggestions, it has obvious shortcomings in practical application.

[0025] Specifically, under the current satisfaction feedback mechanism, users need to manually search for and access the interface, and then manually type in the feedback information. This cumbersome process results in high costs for users to provide feedback, thus discouraging their enthusiasm. Furthermore, the current mechanism relies on user initiative, but in most everyday scenarios, unless encountering a problem that severely impacts usability, users are generally reluctant to engage in the complex process of providing feedback.

[0026] The cumbersome feedback process and low user engagement mean that end-product manufacturers can only collect a very limited amount of satisfaction feedback. As a result, they cannot fully and objectively grasp the true level of user satisfaction with their products, and therefore cannot effectively improve and optimize them based on sufficient data.

[0027] To address the above problems, embodiments of the present invention provide a method for perceiving satisfaction. Figure 1 This is a flowchart illustrating the satisfaction perception method provided by the present invention, as shown below. Figure 1 As shown, the method includes: Step 110: When the device is in operation, acquire user control data and / or device pose data, wherein the user control data represents the user's behavior when controlling the device, and the device pose data represents the pose changes of the device under user control.

[0028] The device here refers to a terminal device, which can be a portable device, such as a smartphone, tablet, dictionary pen, reading pen, handheld game console, or handheld computer.

[0029] The device is in an operational state, meaning it is in use. That is, the device remains operational while the user is using it. For example, in the case of a dictionary pen, the dictionary pen is in an operational state when the user is using it to perform operations such as scanning and looking up words.

[0030] When the device is in operation, user control data can be collected. This user control data represents the user's actions while operating the device; it can be data on the user's behavior in using the device. User control data can be detected by sensors deployed on the device; for example, pressure sensors deployed on the device can detect the pressure data applied to the device by the user.

[0031] Furthermore, device pose data can also be collected while the device is in operation. Device pose data represents the changes in the device's pose under user control, and can include data on the device's position, attitude, and changes during operation. Device pose data can be detected by sensors deployed on the device, such as gyroscopes and accelerometers.

[0032] Step 120: Based on the user control data and / or the device pose data, perform user satisfaction perception to obtain user satisfaction information.

[0033] Specifically, after obtaining user control data and / or device pose data when the device is in operation, user satisfaction can be perceived based on at least one of these two data.

[0034] User satisfaction can be perceived based on user operation data. User operation data represents the nonverbal behavior of users when operating the device. This nonverbal behavior often directly reflects the user's emotions and psychological state. Therefore, user operation data can be used to infer user satisfaction with the device. For example, if user operation data includes the pressure applied to the device by the user, consistently high pressure levels or fluctuations exceeding a certain threshold may indicate user difficulties in using the device, potentially pointing to low user satisfaction.

[0035] Additionally, user satisfaction can be perceived based on device pose data. While device pose data represents the device's position and posture, changes in these states occur during user operation. Therefore, device pose data can accurately reflect subconscious user behavior when using the device, allowing us to infer user satisfaction. For example, we can infer the smoothness of user operation or the presence of repetitive actions. If the device's movement is smooth and repetitive, the user experience is likely smooth and comfortable, indicating higher user satisfaction.

[0036] Furthermore, user satisfaction can be perceived by combining user control data and device pose data. Specifically, user behavior reflected in user control data and device pose changes reflected in device pose data can be combined to infer user satisfaction with the device. In addition, other types of data, such as user emotion data and user interaction data with the device, can be combined with user control data and device pose data to perceive user satisfaction. This embodiment of the invention does not specifically limit this approach.

[0037] Thus, user satisfaction information can be obtained. The satisfaction information referred to here may include one of the pre-defined satisfaction levels or satisfaction categories, and may also include the reasons for an increase or decrease in satisfaction compared to before; however, this embodiment of the invention does not specifically limit this.

[0038] In the method provided in this embodiment of the invention, user operation data and / or device pose data of the device in operation are actively acquired, and satisfaction perception is performed based on the user operation data and / or device pose data. Without requiring the user to perform any additional operations outside of operating and using the device, the user's satisfaction with the device during operation and use can be dynamically and in real time obtained, providing conditions for device updates, optimization, improving user satisfaction with the device, and optimizing the user experience.

[0039] Based on the above embodiments, the method further includes: When the device is in operation, user emotion data and / or human-computer interaction data are acquired; Specifically, when the device is in operation, in addition to collecting user control data and / or device pose data, user emotion data and / or human-computer interaction data can also be collected.

[0040] User emotion data is used to reflect the user's emotional state and changes in emotional state during the operation and use of the device. User emotion data can be detected by sensors deployed on the device. For example, a camera deployed on the device can capture the user's facial image, and then emotion recognition can be performed on the facial image to obtain user emotion data; or a microphone deployed on the device can capture the user's voice, and then emotion recognition can be performed on the user's voice to obtain user emotion data.

[0041] Human-computer interaction (HCI) data reflects the interaction between a user and a device during operation. HCI data can be detected by sensors deployed on the device. For example, infrared sensors and cameras deployed on the device can capture and record the user's viewing distance and duration of interaction with the device screen, thus obtaining the length of time the user gazes at the screen, and recording this gazing time as HCI data. Alternatively, sensors deployed on the device's controls can collect user actions on various controls, such as clicking and pressing, thus recording the user's control actions as HCI data.

[0042] Accordingly, in step 120, the step of perceiving user satisfaction based on the user control data and / or the device pose data to obtain user satisfaction information includes: Based on user emotion data and / or human-computer interaction data, as well as the user control data and / or the device pose data, user satisfaction is perceived to obtain user satisfaction information.

[0043] Specifically, in addition to using user control data and / or device pose data to perceive user satisfaction, user satisfaction can also be perceived based on user emotion data and / or human-computer interaction data.

[0044] Among them, user sentiment data is the direct emotional expression of users when operating and using the device. The positive or negative emotions reflected in the user sentiment data can be directly linked to the user's satisfaction with the device. Human-computer interaction data reflects the user's interactive operations on the device. Whether these operations are repetitive or take a long time can directly reflect the smoothness of the user's operation and use of the device, thereby inferring the user's satisfaction with the device.

[0045] User emotion data and / or human-computer interaction data, as well as user control data and / or device pose data, can be input together into a pre-trained satisfaction perception model to obtain satisfaction information output by the satisfaction perception model, thereby realizing satisfaction perception based on an artificial intelligence model; alternatively, user emotion data and / or human-computer interaction data, as well as user control data and / or device pose data, can be input together into a pre-built rule engine, and the above-mentioned data can be classified into the corresponding satisfaction levels through the judgment rules pre-configured in the rule engine to obtain satisfaction information. This embodiment of the invention does not specifically limit the specific methods used.

[0046] In the method provided in this embodiment of the invention, user satisfaction perception is achieved by combining user emotion data and / or human-computer interaction data, as well as user operation data and / or device pose data. This realizes proactive perception of user satisfaction based on multi-dimensional data, which helps to improve the accuracy and reliability of user satisfaction perception.

[0047] Based on any of the above embodiments, in step 120, the step of perceiving user satisfaction based on user emotion data and / or human-computer interaction data, and user control data and / or device pose data, to obtain user satisfaction information, includes: The user emotion data and / or human-computer interaction data, as well as the user operation data and / or the device pose data, are input into the satisfaction perception model to obtain the satisfaction information output by the satisfaction perception model. The satisfaction information includes the satisfaction level and the contribution weight of various data to the satisfaction level. The satisfaction perception model is trained based on various types of sample data, the satisfaction labels of the various types of sample data, and a feature importance attribution mechanism.

[0048] Specifically, user satisfaction perception based on user emotion data and / or human-computer interaction data, as well as user operation data and / or device posture data, can be achieved through a satisfaction perception model.

[0049] The satisfaction perception model here can be a model obtained through supervised learning based on various types of sample data and their satisfaction labels. Specifically, it can be a lightweight neural network model, such as a combination of MobileNet (Mobile Neural Network) and LSTM (Long Short-Term Memory). The various types of sample data can include four categories: user emotion data, human-computer interaction data, user control data, and device pose data. The satisfaction labels for each type of sample data are pre-labeled labels reflecting the true satisfaction level corresponding to that type of sample data. During supervised learning based on these sample data and their satisfaction labels, the satisfaction perception model can learn the mapping relationship between the various types of sample data and the satisfaction labels, thus possessing the ability to infer the satisfaction level based on the input data. Here, the loss function used in supervised learning can be the cross-entropy loss function.

[0050] Furthermore, a feature importance attribution mechanism is incorporated into the satisfaction perception model. This mechanism is an internal computational process embedded within the model. During the process of calculating the final satisfaction level based on various input data, the mechanism performs real-time backward analysis, calculating and measuring the contribution weight of each type of input data to the final satisfaction level, thereby obtaining the contribution weight of each type of data to the satisfaction level. For example, the gradient-weighted class activation mapping (Grad-CAM) method can be used to calculate the contribution weight of four dimensions of features—user emotion, human-computer interaction, user operation, and device pose—to the satisfaction level for the feature tensors output from each layer of the satisfaction perception model.

[0051] Correspondingly, the satisfaction perception model with a feature importance attribution mechanism can associate the satisfaction labels with lower satisfaction levels with features of data types with higher contribution weights obtained through feature importance attribution mechanism during the training process, thereby forming a low satisfaction feature association library, which can be directly applied in the inference process of the satisfaction perception model.

[0052] After training the satisfaction perception model, user emotion data and / or human-computer interaction data, as well as user control data and / or device pose data, can be input into the model. The model can then infer a satisfaction level based on these data, and use a feature importance attribution mechanism to infer the contribution weights of each data type to the final satisfaction level. Therefore, the satisfaction perception model can output satisfaction information, specifically including the satisfaction level and the contribution weights of each data type to that level.

[0053] In the method provided in this embodiment of the invention, while perceiving the satisfaction level, the contribution weight of various data to the satisfaction level can also be obtained, thereby deeply analyzing the causes of changes in satisfaction and providing conditions for targeted optimization of user experience.

[0054] Based on any of the above embodiments, the satisfaction perception method further includes: If the satisfaction level in the satisfaction information is lower than the level threshold, the low satisfaction data type is determined based on the contribution weight of each type of data in the satisfaction information to the satisfaction level. Based on the aforementioned low satisfaction data, the causes of low satisfaction are identified.

[0055] Specifically, the satisfaction information obtained through satisfaction perception can include satisfaction levels and the contribution weights of various data types to those satisfaction levels. For any given data type, the contribution weight represents the importance of that data type in the calculation of the satisfaction level; in other words, it indicates the determining weight of that data type in inferring the satisfaction level. Understandably, the higher the contribution weight, the greater the influence of that data type on arriving at the satisfaction level.

[0056] For cases where the satisfaction level is below a pre-set threshold, meaning the satisfaction level reflects a low level of satisfaction, the data type for low satisfaction can be determined based on the weight of each type of data in relation to the satisfaction level. For example, data types with a weight higher than a weight threshold can be identified as low satisfaction data types.

[0057] For example, suppose there are five levels of satisfaction, from one to five, where level one represents very dissatisfied, level two represents somewhat dissatisfied, level three represents somewhat satisfied, level four represents satisfied, and level five represents very satisfied. The level threshold can be preset to three levels. For satisfaction levels below level three, i.e., levels one and two, the data type for low satisfaction can be determined based on the weighted contribution of various data to the satisfaction level.

[0058] For example, the weight threshold can be set to 30%, or it can be set to 40% or other values. With a weight threshold of 30%, when the satisfaction level is level one or two, data types with a contribution weight higher than 30% can be selected from various data categories as low satisfaction data types.

[0059] Once the data on low satisfaction levels is obtained, the underlying causes of low satisfaction can be identified. These causes are the reasons for the low satisfaction level, and the categories of causes can vary depending on the type of device. For example, for a dictionary pen, causes of low satisfaction might include "abnormal scanning and recognition," "user resistance to operation," "negative user sentiment," "insufficient definitions," "cumbersome operation," and "poor display quality."

[0060] The causes of low satisfaction can be identified based on a pre-built rule engine. This rule engine integrates judgment rules for the anomalies of various data types, and these judgment rules can be derived from a large amount of labeled data and device usage scenarios.

[0061] For example, if the low satisfaction data type includes device pose data, and "repeated scanning ≥ 3 times and trajectory overlap ≥ 80%" is detected in the device pose data, the cause of the low satisfaction can be determined to be "abnormal scanning recognition". For low satisfaction data types including user operation data, and if "grip pressure change ≥ 50%" is detected in the user operation data, it can be determined that the cause of low satisfaction is "user operation resistance". For low satisfaction data types including user emotion data, and for which "facial expression recognition shows impatience and lasts for ≥2 seconds", it can be determined that the cause of low satisfaction is "negative user emotion". The low satisfaction data types include user operation data and device pose data. In the user operation data, "increased grip pressure" was detected, and in the device pose data, "repeated scanning" was detected. It can be determined that the cause of low satisfaction is "inaccurate interpretation". For low satisfaction data types, including human-computer interaction data, and the detection of "staring at the screen for more than 5 seconds" and "clicking for more explanation" in the human-computer interaction data, it can be determined that the cause of low satisfaction is "insufficient explanation". For low satisfaction data types including device pose data and user operation data, and for the detection of "rapid sliding of device trajectory" and "repeated pauses (overlap ≥70%)" in device pose data, and for the detection of "grip pressure fluctuation" in user operation data, it can be determined that the cause of low satisfaction is "recognition response delay". The low satisfaction data includes human-computer interaction data and user emotion data. The detection of "multiple accidental touches of the back button" in the human-computer interaction data and "user frustration (frowning)" in the user emotion data indicates that the cause of low satisfaction is "cumbersome operation and interaction". The low satisfaction data includes human-computer interaction data, and the detection of "frequent adjustment of screen brightness" and "staring at the screen for more than 8 seconds" in the human-computer interaction data indicates that the cause of low satisfaction is "poor display effect".

[0062] In the method provided in this embodiment of the invention, after obtaining the contribution weight of various types of data to the satisfaction level based on the feature importance attribution mechanism, the low satisfaction data type is determined, and the low satisfaction data type is combined with the preset feature anomaly judgment rules to determine the cause of low satisfaction. In this way, the key parts of various types of data that reduce the user satisfaction level can be accurately located, providing a clear optimization direction for the subsequent execution of the device optimization program.

[0063] Based on any of the above embodiments, the satisfaction perception method further includes: Determine the optimization strategy corresponding to the causes of low satisfaction; Based on the optimization strategy, the device is optimized for application use.

[0064] Specifically, once the causes of low satisfaction are identified, the corresponding optimization strategies can be determined. It can be understood that these optimization strategies are used to optimize device applications, thereby resolving or improving the problems manifested by the causes of low satisfaction.

[0065] Here, the correspondence between low satisfaction triggers and optimization strategies can be pre-defined. For example, optimization strategies corresponding to the low satisfaction trigger "inaccurate interpretation" include adjusting the semantic similarity weights of the translation model; optimization strategies corresponding to the low satisfaction trigger "insufficient interpretation" include supplementing professional domain interpretations, such as adding subject explanations for academic terms; optimization strategies corresponding to the low satisfaction trigger "recognition response delay" include local caching of high-frequency query terms and adjusting the threshold of the image recognition algorithm, thereby shortening the delay from scanning to displaying results; optimization strategies corresponding to the low satisfaction trigger "cumbersome operation and interaction" include optimizing by adding a "one-click return to query page" shortcut button to the scanning interface and simplifying multi-level menu operations; and optimization strategies corresponding to the low satisfaction trigger "poor display effect" include automatically adjusting screen brightness / contrast based on ambient light sensor data and supporting users to long-press the screen to quickly enlarge the interpretation font.

[0066] Once the optimization strategy is determined, it can be executed to optimize the device application. As a result, during the user's real-time operation of the device, proactive satisfaction perception can be achieved without additional user intervention. Furthermore, when satisfaction is low, the system can automatically analyze the causes of low satisfaction and perform targeted self-optimization based on these causes, thereby improving device user satisfaction and optimizing the user experience.

[0067] Furthermore, during the application optimization process based on the optimization strategy, the time consumed by the device's local optimization engine can be evaluated, i.e., the local optimization time can be assessed. If the local optimization time is less than the time threshold, it can be determined that the application optimization is performed based on the device's local optimization engine; if the local optimization time exceeds the time threshold, it can be determined that the device's local optimization engine cannot meet the optimization requirements, and the application optimization is performed by calling the cloud-based optimization engine over the network. For example, for optimization tasks such as parsing complex sentence structures or matching data from multiple domains, the device's local computing power is insufficient to support the rapid completion of the above tasks. In such cases, the cloud-based translation engine can be called for interpretation optimization and supplementation, and the device's local translation knowledge base can be updated synchronously.

[0068] Furthermore, during application optimization, if there are insufficient local resources, such as when optimizing translation tasks for niche subject terms or the latest technical vocabulary, and the local translation knowledge base lacks corresponding professional definitions, or when the accuracy of the local recognition model is lower than the accuracy threshold for special texts such as handwritten or blurred fonts, it can be determined that the device's local optimization engine cannot meet the optimization requirements. In such cases, the application can be optimized by calling the cloud-based optimization engine over the network.

[0069] Furthermore, during the application optimization process, if local optimization is ineffective, for example, if the satisfaction level obtained from the re-collected data is still low after local optimization, or if users directly report "dissatisfaction" with the optimization results, it can be determined that the device's local optimization engine cannot meet the optimization requirements, and the application optimization can be performed by calling the cloud-based optimization engine through the network.

[0070] Based on any of the above embodiments, the step of optimizing the device based on the optimization strategy further includes: Perform feedback interaction with the user to obtain the user's optimization feedback information; If the optimization feedback is positive, the user's preference model is updated based on the optimization strategy, and the preference model is used to update the application of the device.

[0071] Specifically, after optimizing the application based on the optimization strategy, feedback interaction with users can be performed. This can involve displaying the optimized content on the device screen along with a prompt related to satisfaction information. For example, if the cause of low satisfaction is "insufficient explanation," a prompt could be displayed: "We have optimized the explanation for you. Does it meet your needs?" By displaying prompts related to satisfaction information, users can be encouraged to pay attention to application optimizations based on satisfaction data and provide feedback and suggestions on the optimizations. For instance, while displaying prompts related to satisfaction information, users can also provide feedback on the application optimizations by tapping a "Yes / No" control.

[0072] This allows us to obtain user feedback on optimizations. Here, optimization feedback refers to information provided by users regarding application optimizations. Specifically, it can be information like "yes / no" directly indicating whether the optimization can improve user satisfaction, or it can be manually entered feedback text or voice feedback recorded by the user.

[0073] After receiving user feedback, it can be determined whether the feedback is positive or negative. Positive feedback indicates that the trend expressed in the feedback is positive, such as "yes" to "does it meet my needs," or that the semantics of the feedback text are positive. In essence, positive feedback means that the user acknowledges the results of the application optimization. The optimization strategy meets the user's needs and preferences when operating the device, so the optimization strategy can be applied to update the user's preference model. This optimization strategy is then recorded in the user's preference model, allowing the device to directly access the user's preference model to update the device application when the user subsequently operates the device. This ensures that the device application naturally aligns with user preferences, thereby optimizing the user experience.

[0074] Based on any of the above embodiments, the user control data includes grip pressure information and / or scanning sliding pressure information; the device pose data includes at least one of scanning trajectory, repeated scanning action, and pick-up confirmation action.

[0075] Specifically, for the case of a dictionary pen, user control data can include grip pressure information and / or scanning sliding pressure information. Grip pressure information can be obtained from the user's grip force detected by a pressure sensor located on the dictionary pen body, and can include the user's grip pressure value and its changing trend during dictionary pen operation. Scanning sliding pressure information can be obtained from the user's sliding force on the pen tip on the medium detected by a pressure sensor located on the pen tip, and can include the pen tip pressure value and its changing trend during the dictionary pen's word scanning process.

[0076] Additionally, for devices that are dictionary pens, device pose data can include at least one of the following: scanning trajectory, repeated scanning actions, and pick-up confirmation actions. The scanning trajectory refers to the sliding trajectory of the dictionary pen during the word-capturing process. Repeated scanning actions are determined by whether the sliding trajectories of multiple scans during the word-capturing process are identical, indicating whether there are repeated scans of the same content, and the movement trajectory of these actions. For example, a case where the overlap of the preceding and following trajectories is ≥80% can be identified as repeated scanning. The pick-up confirmation action is determined by the scanning sliding trajectory and the rotational displacement of the pen body during the word-capturing process, indicating whether there are multiple interruptions in the scanning process to check the scan results, and the movement trajectory of these actions.

[0077] By combining at least one of the above-mentioned grip pressure information, scanning sliding pressure information, scanning trajectory, repeated scanning action, and picking up confirmation action, proactive perception of user satisfaction can be achieved, thereby effectively improving user satisfaction for dictionary pens.

[0078] Based on any of the above embodiments, for the case where the device is a dictionary pen, the satisfaction perception method includes the following steps: During the process of users using the dictionary pen to scan and look up words, the dictionary pen can dynamically collect at least one of the following: user emotion data, user operation data, device posture data, and human-computer interaction data.

[0079] Among them, user emotion data can be detected by sensors deployed on the device. For example, the user's facial image can be captured by the camera deployed on the device, and then the user's emotion data can be obtained by performing emotion recognition on the facial image. User control data may include grip pressure information and / or scanning sliding pressure information. Grip pressure information can be detected by a pressure sensor located on the body of the dictionary pen, and scanning sliding pressure information can be detected by a pressure sensor located on the tip of the dictionary pen. Device pose data can include at least one of the following: scanning trajectory, repeated scanning action, and pick-up confirmation action. Device pose information can be detected by motion sensors installed on the dictionary pen, such as gyroscopes and accelerometers. Human-computer interaction data reflects the human-computer interaction between users and devices during the operation and use of the devices.

[0080] After obtaining at least one of the aforementioned user emotion data, user operation data, device pose data, and human-computer interaction data, this data can be input into the satisfaction perception model. This satisfaction perception model can be a lightweight neural network model with ≤5 million parameters and an inference latency ≤100ms. The satisfaction perception model can output a satisfaction level. For example, inputting user emotion data indicating a relatively agitated user, user operation data indicating a change in control intensity exceeding 50%, and device pose data indicating more than three repetitions of the trajectory into the satisfaction perception model will result in a satisfaction level of level two, indicating relatively low satisfaction.

[0081] After obtaining the satisfaction level, it can be compared with a threshold. If the satisfaction level is lower than the threshold, user satisfaction is considered low. In this case, the factors causing low user satisfaction can be identified from the dynamically collected data, and corresponding optimization strategies can be determined. The dictionary pen can then be optimized based on these strategies.

[0082] After application optimization is completed, feedback interaction can be performed with the user to obtain their optimization feedback information. If the optimization feedback is positive, the optimization strategy is applied to update the user's preference model. This optimization strategy is then recorded in the user's preference model, so that when the user subsequently operates the device, the device can directly call the user's preference model to update the device application. For example, when a user is reading an English scientific journal, scanning the sentence "CRISPR-Cas9 mediated geneediting," the dictionary pen displays the translation "CRISPR-Cas9 mediated gene editing." During this process, the dictionary pen collects device pose data ("scanned 3 times"), user control data ("grip pressure 250g"), and user emotion data ("corner of the mouth drooping"). Based on this, the satisfaction perception model outputs a probability of 0.68 for a level 1 satisfaction rating, thus determining low user satisfaction.

[0083] In this scenario, the application can be optimized based on an optimization strategy to supplement the professional interpretation: "CRISPR-Cas9-mediated gene editing: a gene editing technology that uses guide RNA to guide Cas9 nucleases to cut specific DNA sites, widely used in biomedical research and gene therapy." Following this, feedback can be provided, displaying a prompt related to satisfaction information alongside the translation results: "The professional interpretation has been optimized for you. Does it meet your needs?" If the user's feedback is "yes," it is determined that the user is satisfied with the application optimization. The optimization strategy "scientific journal scenario + professional interpretation preference" is recorded in the user preference model, and subsequent scans of similar texts will prioritize outputting professional interpretations.

[0084] The satisfaction perception device provided by the present invention is described below. The satisfaction perception device described below and the satisfaction perception method described above can be referred to in correspondence.

[0085] Figure 2 This is a schematic diagram of the satisfaction perception device provided by the present invention, as shown below. Figure 2 As shown, the device includes: The acquisition unit 210 is used to acquire user control data and / or device pose data when the device is in operation. The user control data represents the user's behavior when controlling the device, and the device pose data represents the pose change of the device under user control. The sensing unit 220 is used to perceive user satisfaction based on the user operation data and / or the device pose data, and obtain user satisfaction information.

[0086] In the device provided in the embodiments of the present invention, user operation data and / or device pose data of the device in operation are actively acquired, and satisfaction perception is performed based on the user operation data and / or device pose data. Without requiring the user to perform any additional operations outside of operating and using the device, the user's satisfaction with the device during operation and use can be dynamically and in real time obtained, providing conditions for device updates, optimization, improving user satisfaction with the device, and optimizing the user experience.

[0087] Based on any of the above embodiments, the acquisition unit is further configured to: When the device is in operation, user emotion data and / or human-computer interaction data are acquired; The sensing unit is specifically used for: Based on user emotion data and / or human-computer interaction data, as well as the user control data and / or the device pose data, user satisfaction is perceived to obtain user satisfaction information.

[0088] Based on any of the above embodiments, the sensing unit is specifically used for: The user emotion data and / or human-computer interaction data, as well as the user operation data and / or the device pose data, are input into the satisfaction perception model to obtain the satisfaction information output by the satisfaction perception model. The satisfaction information includes the satisfaction level and the contribution weight of various data to the satisfaction level. The satisfaction perception model is trained based on various types of sample data, the satisfaction labels of the various types of sample data, and a feature importance attribution mechanism.

[0089] Based on any of the above embodiments, the sensing unit is further configured to: If the satisfaction level in the satisfaction information is lower than the level threshold, the low satisfaction data type is determined based on the contribution weight of each type of data in the satisfaction information to the satisfaction level. Based on the aforementioned low satisfaction data, the causes of low satisfaction are identified.

[0090] Based on any of the above embodiments, the device further includes an optimization unit, used for: Determine the optimization strategy corresponding to the causes of low satisfaction; Based on the optimization strategy, the device is optimized for application use.

[0091] Based on any of the above embodiments, the optimization unit is further configured to: Perform feedback interaction with the user to obtain the user's optimization feedback information; If the optimization feedback is positive, the user's preference model is updated based on the optimization strategy, and the preference model is used to update the application of the device.

[0092] Based on any of the above embodiments, the user control data includes grip pressure information and / or scanning sliding pressure information; the device pose data includes at least one of scanning trajectory, repeated scanning action, and pick-up confirmation action.

[0093] Figure 3 An example is a schematic diagram of the physical structure of an electronic device, such as... Figure 3 As 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, the communications interface 320, and the memory 330 communicate with each other via the communication bus 340. The processor 310 can call logical instructions in the memory 330 to execute a satisfaction perception method, which includes: When the device is in operation, user control data and / or device pose data are acquired. The user control data represents the user's behavior when controlling the device, and the device pose data represents the pose changes of the device under user control. User satisfaction is perceived based on the user operation data and / or the device pose data to obtain user satisfaction information.

[0094] 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, or the part that contributes to related technologies, or a portion 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.

[0095] On the other hand, the present invention also provides a computer program product, the computer program product comprising a computer program that can be stored on a non-transitory computer-readable storage medium, wherein when the computer program is executed by a processor, the computer is able to execute the satisfaction perception method provided by the above methods, the method comprising: When the device is in operation, user control data and / or device pose data are acquired. The user control data represents the user's behavior when controlling the device, and the device pose data represents the pose changes of the device under user control. User satisfaction is perceived based on the user operation data and / or the device pose data to obtain user satisfaction information.

[0096] In another aspect, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, is implemented to perform the satisfaction perception method provided by the methods described above, the method comprising: When the device is in operation, user control data and / or device pose data are acquired. The user control data represents the user's behavior when controlling the device, and the device pose data represents the pose changes of the device under user control. User satisfaction is perceived based on the user operation data and / or the device pose data to obtain user satisfaction information.

[0097] 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.

[0098] 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 parts that contribute to the related technology, can be embodied in the form of software products. 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.

[0099] 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. A method for perceiving satisfaction, characterized in that, include: When the device is in operation, user control data and / or device pose data are acquired. The user control data represents the user's behavior when controlling the device, and the device pose data represents the pose changes of the device under user control. User satisfaction is perceived based on the user operation data and / or the device pose data to obtain user satisfaction information.

2. The satisfaction perception method according to claim 1, characterized in that, Also includes: When the device is in operation, user emotion data and / or human-computer interaction data are acquired; The step of perceiving user satisfaction based on the user control data and / or the device pose data to obtain user satisfaction information includes: Based on user emotion data and / or human-computer interaction data, as well as the user control data and / or the device pose data, user satisfaction is perceived to obtain user satisfaction information.

3. The satisfaction perception method according to claim 2, characterized in that, The process of perceiving user satisfaction based on user emotion data and / or human-computer interaction data, as well as user control data and / or device pose data, to obtain user satisfaction information includes: The user emotion data and / or human-computer interaction data, as well as the user operation data and / or the device pose data, are input into the satisfaction perception model to obtain the satisfaction information output by the satisfaction perception model. The satisfaction information includes the satisfaction level and the contribution weight of various data to the satisfaction level. The satisfaction perception model is trained based on various types of sample data, the satisfaction labels of the various types of sample data, and a feature importance attribution mechanism.

4. The satisfaction perception method according to claim 1, characterized in that, Also includes: If the satisfaction level in the satisfaction information is lower than the level threshold, the low satisfaction data type is determined based on the contribution weight of each type of data in the satisfaction information to the satisfaction level. Based on the aforementioned low satisfaction data, the causes of low satisfaction are identified.

5. The satisfaction perception method according to claim 4, characterized in that, Also includes: Determine the optimization strategy corresponding to the causes of low satisfaction; Based on the optimization strategy, the device is optimized for application use.

6. The satisfaction perception method according to claim 5, characterized in that, The process of optimizing the device based on the optimization strategy further includes: Perform feedback interaction with the user to obtain the user's optimization feedback information; If the optimization feedback is positive, the user's preference model is updated based on the optimization strategy, and the preference model is used to update the application of the device.

7. The satisfaction perception method according to any one of claims 1 to 6, characterized in that, The user control data includes grip pressure information and / or scanning sliding pressure information; the device pose data includes at least one of scanning trajectory, repeated scanning action, and picking up confirmation action.

8. A satisfaction perception device, characterized in that, include: The acquisition unit is used to acquire user control data and / or device pose data when the device is in operation. The user control data represents the user's behavior when controlling the device, and the device pose data represents the pose changes of the device under user control. The sensing unit is used to perceive user satisfaction based on the user operation data and / or the device pose data, and obtain user satisfaction information.

9. 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 satisfaction perception method as described in any one of claims 1 to 7.

10. 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 satisfaction perception method as described in any one of claims 1 to 7.

11. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the satisfaction perception method as described in any one of claims 1 to 7.

Images