Service plan recommendation method and electronic device

Abstract

This application discloses a service scheme recommendation method and an electronic device, relating to the field of information processing, comprising: obtaining input information of at least two modalities, the input information representing the cognitive state of an operator of the electronic device; determining the cognitive state of the operator based on the input information; and providing a service scheme for the operator based on the cognitive state of the operator, wherein different cognitive states correspond to different service schemes, and different service schemes provide different amounts of information.

Description

Technical Field

[0001] This application relates to the field of information processing, and more particularly to a service solution recommendation method and an electronic device. Background Technology

[0002] With the continuous development of Large Language Model (LLM) technology, intelligent agents are gradually evolving from the traditional passive response mode to the proactive service mode.

[0003] In multitasking and high cognitive load work scenarios, users' needs for assistive tools are no longer limited to basic information query and feedback. Instead, they expect intelligent agents to act as cognitive support tools, providing adaptive assistance to users without interfering with their normal work processes.

[0004] Traditional agents typically aim to maximize information content or semantic relevance when generating answers. Therefore, current service recommendations for users tend to provide detailed and comprehensive explanatory documents. However, reading and understanding such lengthy content requires significant time and effort, which doesn't align with users' desire for quick answers or automated execution, resulting in recommended services that don't meet their expectations. Summary of the Invention

[0005] The first aspect of this application provides a service solution recommendation method applied to electronic devices, including:

[0006] Obtain input information in at least two modalities, wherein the input information characterizes the cognitive state of the operator of the electronic device;

[0007] The cognitive state of the operator is determined based on the input information;

[0008] Based on the cognitive state of the operator, a service plan is provided to the operator. Different cognitive states correspond to different service plans, and different service plans provide different amounts of information.

[0009] In one possible implementation, determining the cognitive state of the operator based on the input information includes:

[0010] Based on the preset analysis model, the first modal information in the input information is analyzed to obtain the first cognitive state of the operator. The first cognitive state includes a first load cognitive state and a second load cognitive state. The cognitive load corresponding to the first load cognitive state is higher than the cognitive load corresponding to the second load cognitive state.

[0011] Based on the first cognitive state being a first load cognitive state, and according to the second modal information in the input information, the second cognitive state of the operator is determined. The second cognitive state includes a first sub-state and a second sub-state. The first sub-state indicates that the behavior pattern of the operator satisfies the preset behavior pattern stability condition, and the second sub-state indicates that the behavior pattern of the operator does not satisfy the preset behavior pattern stability condition.

[0012] In one possible implementation, based on the premise that the first cognitive state is a first load cognitive state, and according to the second modal information in the input information, determining the second cognitive state of the operator includes:

[0013] Analyze the second modal information to obtain the first parameter corresponding to the second modal information;

[0014] Determine the second parameter corresponding to the first load cognitive state;

[0015] Based on the first parameter and the second parameter, the target parameters of the operating body are determined;

[0016] Based on the fact that the value of the target parameter is not greater than a preset threshold, the second cognitive state of the operator is determined to be the first sub-state;

[0017] Based on the fact that the value of the target parameter is greater than a preset threshold, the second cognitive state of the operator is determined to be the second sub-state.

[0018] In one possible implementation, the analysis of the second modal information to obtain the first parameter corresponding to the second modal information includes:

[0019] Based on the second modal information containing the operational behavior information of the operator, the second modal information is analyzed to obtain the behavior pattern of the operator;

[0020] Based on the fact that the behavior pattern satisfies the preset behavior pattern stability condition, the first parameter is determined to be a first value;

[0021] Based on the fact that the behavior pattern does not meet the preset behavior pattern stability condition, the first parameter is determined to be a second value, and the second value is greater than the first value.

[0022] In one possible implementation, the service plan for the operator based on the operator's cognitive state includes:

[0023] Based on the cognitive state of the operator, the type of service provision plan is determined;

[0024] Based on the type of service plan provided, a service plan is determined, and a service plan is provided for the operating body.

[0025] In one possible implementation, determining the type of service provision based on the cognitive state of the operator includes:

[0026] Based on the determination that the cognitive state of the operator is the second load cognitive state, a first type of service plan is determined according to the second load cognitive state;

[0027] Based on the fact that the cognitive state of the operator is the first sub-state, a second type of service scheme is determined to be provided. The amount of information provided by the second type of service scheme is less than the amount of information provided by the first type of service scheme.

[0028] Based on the fact that the cognitive state of the operator is the second sub-state, it is determined to provide a third type of service. The amount of information provided by the third type of service is greater than the amount of information provided by the second type and less than the amount of information provided by the first type.

[0029] In one possible implementation, a service plan is determined based on the type of service plan, and a service plan is provided to the operating entity, including:

[0030] Based on the fact that the service plan is of the third type, at least two alternative service plans are generated;

[0031] Based on at least two decision dimensions, the expected cognitive benefit of each alternative service plan is determined, wherein the expected cognitive benefit represents the degree to which the operator recognizes the expected value of the alternative service plan;

[0032] Based on the expected cognitive benefits, a target service plan is determined from at least two alternative service plans, wherein the expected cognitive benefits of the target service plan are greater than the expected cognitive benefits of any non-target service plan.

[0033] In one possible implementation, determining the target service option from at least two alternative service options based on the expected cognitive benefits includes:

[0034] The feature information of the operator is obtained, which represents the operator's preference for the amount of information provided by the service information. The feature information is obtained based on the analysis of the operator's historical input information.

[0035] Based on the expected cognitive benefits and the feature information, a target service solution is determined from at least two alternative service solutions. The degree of matching between the target service solution and the feature information, as well as the expected cognitive benefits, are better than the degree of matching between the non-target service solutions and the feature information, as well as the expected cognitive benefits.

[0036] In one possible implementation, obtaining input information from at least two modalities includes:

[0037] Obtain first modal information, which includes at least two consecutively arranged images, each containing an image of the operator;

[0038] The second modal information is obtained, which includes input operation information of each input device in the electronic device within a preset time period, at least two display interfaces displayed by the display device in the electronic device within the preset time period, and the display duration of each display interface.

[0039] A second aspect of this application provides an electronic device, comprising:

[0040] At least two interfaces are provided for receiving input information in at least two modalities, wherein the input information characterizes the cognitive state of the operator of the electronic device.

[0041] The processor is configured to determine the cognitive state of the operator based on the input information; and to provide a service plan for the operator based on the cognitive state, wherein different cognitive states correspond to different service plans and different service plans provide different amounts of information.

[0042] A third aspect of this application provides a computer program product including computer-readable instructions that, when executed on an electronic device, cause the electronic device to implement the service scheme recommendation method of the first aspect or any implementation thereof.

[0043] A fourth aspect of this application provides an electronic device, including at least one processor and a memory connected to the processor, wherein:

[0044] The memory is used to store computer programs;

[0045] The processor is used to execute the computer program so that the electronic device can implement the service scheme recommendation method of the first aspect or any implementation thereof.

[0046] The fifth aspect of this application provides a computer storage medium carrying one or more computer programs, which, when executed by an electronic device, enable the electronic device to recommend a service scheme according to the first aspect or any implementation thereof. Attached Figure Description

[0047] The above and other features, advantages, and aspects of the embodiments of this disclosure will become more apparent from the accompanying drawings and the following detailed description. Throughout the drawings, the same or similar reference numerals denote the same or similar elements. It should be understood that the drawings are schematic, and the originals and elements are not necessarily drawn to scale.

[0048] Figure 1This is a flowchart illustrating a service solution recommendation method provided in an embodiment of this application;

[0049] Figure 2 This is a flowchart illustrating how the cognitive state of an operator is determined based on input information, as provided in an embodiment of this application.

[0050] Figure 3 This is a flowchart illustrating the process of determining the second cognitive state of an operator based on the second modal information in the input information, according to an embodiment of this application, which is based on the first cognitive state being the first load cognitive state.

[0051] Figure 4 This is a flowchart illustrating the process of analyzing the second modal information to obtain the first parameter corresponding to the second modal information, provided in an embodiment of this application.

[0052] Figure 5 This is a flowchart illustrating a service solution for an operator based on its cognitive state, as provided in an embodiment of this application.

[0053] Figure 6 This is a flowchart illustrating the process of determining a service plan based on the type of service plan provided in this application embodiment, and providing a service plan for the operating entity.

[0054] Figure 7 This is a flowchart illustrating a service solution recommendation method provided in this application embodiment in an application scenario;

[0055] Figure 8 This is a schematic diagram of the structure of the intelligent agent provided in the embodiments of this application;

[0056] Figure 9 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Detailed Implementation

[0057] The embodiments of this application are described below with reference to the accompanying drawings. The terminology used in the implementation section of this application is for explaining specific embodiments only and is not intended to limit the scope of this application.

[0058] The embodiments of this application will now be described with reference to the accompanying drawings. Those skilled in the art will recognize that, with technological advancements and the emergence of new scenarios, the technical solutions provided in the embodiments of this application are equally applicable to similar technical problems.

[0059] The terms "first," "second," etc., used in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such terms are interchangeable where appropriate; this is merely a way of distinguishing objects with the same attributes in the embodiments of this application. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion, so that a process, method, system, product, or apparatus that comprises a series of elements is not necessarily limited to those elements, but may include other elements not explicitly listed or inherent to those processes, methods, products, or apparatuses.

[0060] Reference Figure 1 , Figure 1 This is a flowchart illustrating a service recommendation method provided in an embodiment of this application, such as... Figure 1 As shown in the embodiment of this application, a service solution recommendation method may include steps 101 to 103, which are described in detail below.

[0061] 101. Obtain input information in at least two modalities, whereby the input information characterizes the cognitive state of the operator of the electronic device;

[0062] The service solution recommended in this paper is applied to electronic devices equipped with intelligent agents. The operator interacts with the intelligent agent, which provides the operator with the service solution. The operator is a user performing operations on the electronic device. The service solution can be a passive service generated by the user actively inputting a session through an input device, or an active service provided based on the user's cognitive state.

[0063] The recommended approach in this application is to provide proactive services to the operator.

[0064] The operator obtains multimodal input information, which can characterize the operator's cognitive state.

[0065] Accordingly, the input information for at least two modalities is first obtained.

[0066] The modalities included in this input information can be set according to actual conditions. Through multimodal input information, the cognitive state of the operator can be represented from multiple information dimensions. For example, the input information may include information entered through a keyboard and mouse on an electronic device, audio information collected through a microphone, and image information collected through a camera.

[0067] 102. Determine the cognitive state of the operator based on the input information;

[0068] This intelligent agent can process input information and determine the current cognitive state of the operator.

[0069] The cognitive state is divided based on the cognitive load of the operator. The cognitive state of the operator can be multiple, such as a high cognitive load state and a low cognitive load state. Furthermore, the high cognitive load state can be further subdivided into a first sub-state and a second sub-state, etc. This will be explained in detail in subsequent embodiments, and will not be elaborated here.

[0070] 103. Based on the cognitive state of the operator, provide a service plan for the operator. Different cognitive states correspond to different service plans, and different service plans provide different amounts of information.

[0071] After determining the cognitive state of the operator, a service plan corresponding to the cognitive state is provided to the operator.

[0072] Intelligent agents can provide different service solutions for different cognitive states, and the amount of information provided to the agent varies in different service solutions.

[0073] Because the amount of information the operator needs to obtain from the outside varies depending on the operator's current cognitive state, corresponding service solutions are provided based on that state.

[0074] As an example, if an operator is in a certain cognitive state and needs to learn more information from the outside world, then the service plan provided to it can provide a large amount of information, such as a detailed reasoning / explanation process. Conversely, the service plan provided to it can provide a small amount of information, such as only providing the direct result without providing a detailed reasoning process.

[0075] In this embodiment, at least two modal input information is obtained, representing the cognitive state of the operator of the electronic device. The cognitive state of the operator is determined based on the input information. A service plan is provided to the operator based on the cognitive state, with different service plans corresponding to different cognitive states, and each service plan providing a different amount of information. By using multimodal input information, the cognitive state of the operator can be determined more accurately. Furthermore, by providing different information service plans for different cognitive states, the system can match the user's need for external information and improve the effectiveness of user assistance and support.

[0076] In one possible implementation, obtaining input information in at least two modalities includes: obtaining first modal information, which includes at least two consecutively arranged images, each image containing an image of an operator; and obtaining second modal information, which includes input operation information of each input device in the electronic device within a preset time period, at least two display interfaces displayed by the display device in the electronic device within the preset time period, and the display duration of each display interface.

[0077] The input information, which includes at least two modalities, contains images, input operation information, and other information related to the input operation information.

[0078] The image information can be a real-time video composed of multiple consecutively arranged images, and each frame of the image contains an image of the operator. Based on the image information, physiological signals such as facial expressions of the operator can be detected, and the current cognitive state of the operator can be determined through these physiological signals.

[0079] For example, if the operator frowns or pouts, it can be considered that it is in a high-load cognitive state; conversely, if the operator relaxes its eyebrows or smiles, it can be considered that it is in a low-load cognitive state.

[0080] The input operation information is generated by the operation performed by the operator through the input device, which can be a keyboard, mouse, touch screen, touchpad, etc. The operator performs the input operation through the input device.

[0081] The second modal information is the input operation information collected by the input device in the electronic device. The collected input operation information can be acquired within a preset time period and analyzed based on the preset time period to determine the index parameters such as the operation frequency of the operator.

[0082] The duration of this preset time period can be set according to actual conditions. For example, it can be set to 1 hour, 10 minutes, 1 minute, etc. This application does not impose any restrictions on the duration of this preset time period.

[0083] The second modal information may also include other information related to the input operation, such as multiple display interfaces displayed by the display device and the display duration of each interface. The switching of the display interface is implemented according to the switching operation in the input operation information. This other information is also acquired through a preset time period, and analysis is performed based on this preset time period and the input operation information to determine the intent corresponding to the input operation. This intent can be purposeful or aimless input information.

[0084] In one possible implementation, the intention of the operator is clear, and its input operations and the duration of viewing on each display interface conform to the viewing and input patterns. However, when the intention is unclear, the input operations and the duration of viewing on each display interface do not conform to the viewing and input patterns. Therefore, the intention of the operator is determined by combining the input operation information and the viewing duration on each display interface.

[0085] The duration of the displayed interface represents the length of time the user's attention remains on that interface. The duration of attention reflects the amount of information the user needs from that interface; a longer dwell time indicates a greater need for information, and vice versa. Furthermore, frequent switching between interfaces suggests that the user is uncertain about obtaining effective information from each interface, and their purpose is less clear.

[0086] In this embodiment, first modal information is obtained, which includes at least two consecutively arranged images, each containing an image of the operator. Second modal information is obtained, which includes input operation information of each input device in the electronic device within a preset time period, at least two display interfaces displayed by the display device in the electronic device within the preset time period, and the display duration of each display interface. By combining the first and second modal information, the cognitive state of the operator can be analyzed from multiple dimensions, such as the operator's facial expressions, input operations, and control of related display content. Analyzing from multiple dimensions is more accurate and yields more detailed results compared to single-dimensional information analysis.

[0087] Figure 2 This is a flowchart illustrating the process of determining the cognitive state of an operator based on input information, as provided in this application embodiment. It may include steps 201 to 202, which are described in detail below.

[0088] 201. Based on the preset analysis model, analyze the first modal information in the input information to obtain the first cognitive state of the operator. The first cognitive state includes the first load cognitive state and the second load cognitive state. The cognitive load corresponding to the first load cognitive state is higher than the cognitive load corresponding to the second load cognitive state.

[0089] The first modal information can be used to determine the first cognitive state of the operator. The first cognitive state is a coarse classification and may include a first cognitive state with high cognitive load and a second cognitive state with low cognitive load.

[0090] In this first cognitive load state, the operator is under high cognitive load. High cognitive load refers to a state in which the demand on the brain's working memory and information processing resources exceeds the comfortable capacity when completing a task, resulting in attention strain, error-proneness, learning difficulties, or a decline in user experience.

[0091] In this second cognitive load state, the operator is in a low cognitive load state. Low cognitive load refers to a state in which the demand on the brain's working memory and information processing resources is low when completing a task, and the user can understand, operate, or make decisions without expending a lot of mental effort.

[0092] In one possible implementation, the pre-defined analysis model is a model used in the agent that can analyze the input information to obtain the cognitive state of the agent.

[0093] As an example, the first modal information is a video composed of multiple frames containing images of the operator. The preset analysis model analyzes the input video by analyzing physiological signals such as the operator's facial expressions in the video to obtain the first cognitive state.

[0094] In one possible implementation, the pre-defined analysis model can employ a deep learning model, which can analyze physiological signals such as facial expressions to determine the first cognitive state of the operator.

[0095] 202. Based on the fact that the first cognitive state is the first load cognitive state, and according to the second modal information in the input information, the second cognitive state of the operator is determined. The second cognitive state includes a first sub-state and a second sub-state. The first sub-state indicates that the behavior pattern of the operator meets the preset behavior pattern stability condition, and the second sub-state indicates that the behavior pattern of the operator does not meet the preset behavior pattern stability condition.

[0096] If the cognitive state of the operator is a high-load first cognitive state, it is further divided into a second cognitive state. This second cognitive state includes a first sub-state and a second sub-state, both of which characterize the operator's current thinking process. The first sub-state indicates that the operator's thinking is currently fluent, while the second sub-state indicates that the operator's thinking is currently not fluent. Specifically, the operator's behavior pattern is stable in the first sub-state and unstable in the second sub-state.

[0097] This behavioral pattern is derived from the analysis of second-modal information, which represents the operations performed by the operator on the electronic device and the corresponding information fed back by the electronic device.

[0098] As an example, the second modal information includes behavioral pattern information and contextual information. The behavioral pattern information is the operation information input through the input device, and the contextual information is the multiple display interfaces displayed on the display device in the electronic device and the display duration of each display interface. By analyzing the operation information to control the switching of display interfaces and the duration of stay on each display interface, it is possible to analyze and determine whether the current operator is in a high-intensity output state, which can be called flow, or in a state of mental stagnation due to encountering difficulties, which can be called struggle.

[0099] In one possible implementation, after the agent uses contextual information to determine the current display interfaces and the duration of each display interface, it can identify the interface with the longest duration, take a screenshot of the interface with the longest duration, and use the screenshot as input information for the agent so that it can be used for subsequent analysis.

[0100] In one possible implementation, in the first sub-state, the operator is in a flow state, where the user is in a high-intensity productive state, experiencing high workload but smooth thinking. In the second sub-state, the operator is in a stuck state, where the user encounters a problem and their thinking becomes stagnant, also manifesting as high workload.

[0101] The basis for determining whether the operator is in the first sub-state or the second sub-state is whether the operator's behavior pattern is stable. Stable behavior pattern can include a fixed input rhythm, a stable interface switching rhythm, or a stable operating object. When the operator's thinking is smooth, its behavior pattern is stable, and it can be determined that the operator is in a state of flow.

[0102] For example, the current operator is writing an article about the applications of robots in daily life. The operator switches between various interfaces, such as editing documents, referencing documents, and browsing web pages. These web pages include robot introduction interface 1, robot introduction interface 2, and a car interface. The operator switches between these interfaces using the mouse. If the operator frequently switches between interfaces within a preset time period, it can be determined that the current operator is in the second sub-state. If the operator switches between interfaces within the preset time period and stays on each interface or a particular interface for a relatively long time, allowing the operator to gather sufficient information from the interface, it can be determined that the current operator is in the first sub-state.

[0103] In one possible implementation, since the operator is in different cognitive states, its cognitive load and thinking state are different, and therefore its need for external prompts or assistance is different. Therefore, different service solutions can be provided for different cognitive states. The specific process of providing service solutions will be described in subsequent embodiments, and will not be detailed here.

[0104] In this embodiment, based on a preset analysis model, the system first analyzes and determines whether the operator is in a high-load cognitive state or a low-load cognitive state based on the first modal information. If the operator is in a high-load cognitive state, the system further analyzes and determines which sub-state the operator is in based on the second modal information. The operator's behavior patterns are different in different sub-states, thus achieving an accurate and detailed classification of the operator's cognitive state and providing a basis for providing different service solutions in the future.

[0105] Figure 3 This is a flowchart illustrating how, based on the first cognitive state being a first load cognitive state and according to the second modal information in the input information, the second cognitive state of the operator is determined according to the embodiments of this application. It may include steps 301 to 305, which are described in detail below.

[0106] 301. Analyze the second modal information to obtain the first parameter corresponding to the second modal information;

[0107] The second modal information includes operation information, display device interface, and display duration of each display interface. The operation information may include operation information for inputting characters and operation information for controlling interface switching.

[0108] Accordingly, for each item in the second modality information, the agent can process and obtain the parameter value for each item, which can be a score for that item. The first parameter corresponding to the second modality information can include the score for each item.

[0109] The agent is trained to process the input second modality information and obtain a score for each item.

[0110] As an example, the second modality information includes behavioral pattern information, contextual information, and screenshots determined using the contextual information. For each item, the agent obtains a corresponding parameter value. For instance, the behavioral pattern information obtains a parameter value, the contextual information obtains a parameter value, and the screenshot obtains a parameter value; these three parameter values ​​serve as the first parameter corresponding to the second modality information.

[0111] 302. Determine the second parameter corresponding to the first cognitive state of load;

[0112] A second parameter is set in advance for the first load cognitive state. When it is determined that the cognitive state of the operator is the first load cognitive state, the preset second parameter can be obtained.

[0113] 303. Determine the target parameters of the operating body based on the first and second parameters;

[0114] Using the first and second parameters, mathematical calculations are performed to obtain the target parameters corresponding to the operation body.

[0115] In one possible implementation, weights are set for the second modal information and the first load cognitive state, and the target parameters of the operator are obtained by weighted calculation.

[0116] This weight can be preset or adjusted in real time based on the agent's feedback on historical service plans during the interaction process.

[0117] 304. Based on the fact that the value of the target parameter is not greater than a preset threshold, the second cognitive state of the operator is determined to be the first sub-state;

[0118] 305. Based on the fact that the value of the target parameter is greater than the preset threshold, the second cognitive state of the operator is determined to be the second sub-state.

[0119] A preset threshold is set to determine whether the operator is in the first sub-state or the second sub-state. After obtaining the target parameter corresponding to the operator, the target parameter can be compared with the preset threshold. If the value of the target parameter is not greater than the preset threshold, the operator's cognitive state can be determined to be the first sub-state; otherwise, the operator's cognitive state is the second sub-state.

[0120] If the value of the target parameter is not greater than the preset threshold, it indicates that the operator is in a high cognitive load flow state. In this state, the operator is in a high-intensity output state and thinking is smooth. The cognitive state of the operator is determined to be the first sub-state. If the value of the target parameter is greater than the preset threshold, it indicates that the operator is in a high cognitive load stuck state. In this state, the operator is in a high-intensity cognition state, but is stuck in thinking due to encountering difficulties. The cognitive state of the operator is determined to be the second sub-state.

[0121] In subsequent steps, once the detailed cognitive states of the operator have been determined, corresponding service solutions can be provided for each cognitive state.

[0122] In this embodiment, by analyzing the second modal information, a first parameter corresponding to the second modal information is obtained, and a second parameter corresponding to the current first load cognitive state of the operator is determined. Based on the first and second parameters, the target parameter of the operator is determined. Using the relationship between the target parameter and a preset threshold, the second cognitive state of the operator is subdivided into a first sub-state and a second sub-state. If the value of the target parameter is not greater than the preset threshold, the second cognitive state of the operator is determined to be the first sub-state; otherwise, the second cognitive state of the operator is determined to be the second sub-state. This achieves the quantification and scoring of the second modal information and the first load cognitive state, obtains the corresponding target parameter through mathematical calculation, and then uses the relationship between the target parameter and the preset threshold to quantify and determine the refined classification of the operator's second cognitive state, providing a more accurate basis for providing different service solutions for different cognitive states.

[0123] Figure 4 This is a flowchart illustrating the process of analyzing second modal information to obtain the first parameter corresponding to the second modal information, as provided in the embodiments of this application. It may include steps 401 to 403, which are described in detail below.

[0124] 401. Based on the second modal information containing the operator's operational behavior information, analyze the second modal information to obtain the operator's behavior pattern;

[0125] The second modal information includes the operator's operational behavior information, which is the input operation performed by the operator on the input device of the electronic device. Correspondingly, it may also include the response information of the electronic device in response to the input operation. The response information may include at least two display interfaces displayed by the display device in the electronic device and the display duration of each display interface, etc.

[0126] Accordingly, by analyzing the second modal information, the behavior pattern of the operator is obtained, which is the pattern in which the operator performs operation on the display content in the electronic device.

[0127] In one possible implementation, the agent can perform pattern analysis on operational behaviors to determine its behavioral patterns.

[0128] 402. Based on the fact that the behavior pattern meets the preset stability conditions of the behavior pattern, the first parameter is determined to be the first value;

[0129] The preset behavioral pattern stability condition is used to determine whether the behavior pattern of the operator is stable. Behavioral pattern stability means that the operator's input operation behavior and decision logic performed on the electronic device are highly consistent under different environmental conditions, input distribution, or time.

[0130] As an example, the stable condition for the preset behavior pattern can be that the frequency of input operations performed by the operator is stable within a preset time period, and the viewing duration in each display interface conforms to the viewing pattern and the input pattern.

[0131] Of course, the stability conditions of the preset behavior pattern may differ for different application scenarios. For example, the stability conditions of the preset behavior pattern used for service solutions in gaming scenarios and service solutions in office scenarios are different.

[0132] If the behavior pattern meets the preset behavior pattern stability condition, the behavior pattern of the operator is stable, and it can operate the electronic device with stable operation. Based on this, it can be determined that the current operator is in a high-intensity output state and its thinking is smooth. Accordingly, a lower first value can be set for the first parameter of the second modality information, and the target parameter of the operator is ultimately lower.

[0133] 403. Based on the fact that the behavior pattern does not meet the preset behavior pattern stability condition, the first parameter is determined to be the second value, and the second value is greater than the first value.

[0134] If the behavior pattern does not meet the preset behavior pattern stability condition, the behavior pattern of the operator is disordered, and it operates the electronic device with unstable operations. Based on this, it can be determined that the current operator is in a state of mental stagnation due to encountering difficulties. Accordingly, a higher second value can be set for the first parameter of the second modality information, and the target parameter of the operator will eventually be higher.

[0135] In this embodiment, if the second modal information contains the operator's operational behavior information, the operator's behavior pattern can be obtained by analyzing the second modal information. If the behavior pattern meets the preset behavior pattern stability condition, the first parameter of the second modal information is determined to be a lower first value; otherwise, the first parameter is determined to be a higher second value. By judging whether the operator's operational behavior pattern is stable, different values ​​are assigned to the first parameter of the second modal information, providing a basis for subsequently determining the value of the operator's target parameter. Moreover, assigning different values ​​to different behavior patterns of the operator enables a more detailed division of the operator's cognitive state after the target parameter of the operator is subsequently determined. This allows for the determination of the operator's current cognitive state by combining multiple dimensions such as the operator's physiological information and behavioral information, resulting in a more accurate and detailed determination of the operator's cognitive state.

[0136] Figure 5 This is a flowchart illustrating a service solution for an operator based on its cognitive state, provided in an embodiment of this application. It may include steps 501 to 502, which are described in detail below.

[0137] 501. Determine the type of service plan based on the cognitive state of the operator;

[0138] A pre-defined correspondence is established, which represents the type of service plan corresponding to each cognitive state of the operator.

[0139] Accordingly, after determining the cognitive state of the operator, the type of service solution corresponding to that cognitive state can be determined based on the correspondence.

[0140] For example, if there are three cognitive states, there can also be three types of service provision solutions.

[0141] In one possible implementation, the type of service provision is determined based on the cognitive state of the operator, including:

[0142] 5011. Based on the fact that the cognitive state of the determined operator is the second load cognitive state, the first type of service plan is determined according to the second load cognitive state;

[0143] The second cognitive load corresponds to a lower cognitive load for the operator. At this time, the operator's demand for working memory and information processing resources is low, and the operator has the energy to understand a large amount of information in detail. Accordingly, the service plan is determined to be of the first type.

[0144] In one possible implementation, the first type of service solution can be a detailed solution, and correspondingly, the generated service solution provides an active and detailed solution for the operator.

[0145] 5012. Based on the fact that the cognitive state of the operator is the first sub-state, it is determined to provide a second type of service scheme. The amount of information provided by the second type of service scheme is less than the amount of information provided by the first type of service scheme.

[0146] The first sub-state corresponds to the operator being under high cognitive load and currently in a high-intensity output state. In this case, in order to protect the operator's smooth thinking from being disturbed, the most taboo thing is to be disturbed. Accordingly, the service solution is determined to be the second type to protect the operator's flow state.

[0147] In one possible implementation, the second type of service scheme can be a silent service scheme.

[0148] As an example, if a user's inbox receives an email, the second type of service provided by the intelligent agent can directly control the inbox to send a reply email in the background.

[0149] 5013. Based on the fact that the cognitive state of the operator is the second sub-state, it is determined to provide a third type of service scheme. The amount of information provided by the third type of service scheme is greater than the amount of information provided by the second type and less than the amount of information provided by the first type.

[0150] The first sub-state corresponds to an operator being under high cognitive load, and the operator is currently in a state of mental stagnation due to encountering difficulties. In this case, in order to provide timely and effective proactive services to the operator, precise services are required. Accordingly, the service plan is determined to be of the third type.

[0151] In one possible implementation, the third type of service solution could be a solution that provides precise services.

[0152] As an example, if a user's inbox receives an email, the third type of service offered by the intelligent agent is to notify the user that an email has been received, which can include information about the received email and a redirect link.

[0153] 502. Based on the type of service solution provided, determine the service solution and provide a service solution for the operating entity.

[0154] After determining the type of service plan, and combining it with the content presented in the electronic device, the necessary proactive service plan is provided to the operator.

[0155] In this first type of service solution, a detailed explanation can be provided. Accordingly, the generated service solution offers a proactive and detailed explanation for the operator. This detailed explanation may include the reasoning process and the detailed operational steps that the operator can perform.

[0156] As an example, if a user's inbox receives an email, the first type of service provided by the intelligent agent is to notify the user that an email has been received, provide the content of the email, prompt whether to be redirected to the email interface to reply, and provide a redirect link and reply template, etc.

[0157] Among them, the second type of service solution can be a silent service solution, which means that no prompt information is actively generated, and a direct handling service solution can be adopted.

[0158] As an example, if a user's inbox receives an email, the second type of service provided by the intelligent agent can directly control the inbox to send a reply email in the background.

[0159] The third type of service solution can be a precise service solution. Correspondingly, the generated service solution is a direct answer, such as providing a brief explanation for the operator. This brief explanation can include the operational steps that the operator can perform, without needing to explain the rationale behind each step in detail.

[0160] As an example, if a user's inbox receives an email, the third type of service provided by the intelligent agent is to notify the user that an email has been received, which can include a notification and a redirect link.

[0161] In this embodiment, based on the determination that the cognitive state of the operator is a second load cognitive state, a first type of service scheme is determined according to the second load cognitive state; based on the operator's cognitive state being a first sub-state, a second type of service scheme is determined to be provided, where the amount of information provided by the second type of service scheme is less than the amount of information provided by the first type of service scheme; based on the operator's cognitive state being a second sub-state, a third type of service is determined to be provided, where the amount of information provided by the third type of service scheme is greater than the amount of information provided by the second type of service scheme but less than the amount of information provided by the first type of service scheme, so as to achieve providing service schemes with different amounts of information for different cognitive states of the operator.

[0162] Figure 6 This is a flowchart illustrating the process of determining a service plan based on the type of service plan provided in this application embodiment, and providing a service plan for the operating entity. It may include steps 601 to 603, which are described in detail below.

[0163] 601. Based on the fact that the service plan is of type 3, generate at least two alternative service plans;

[0164] If the service plan is of type three, a precise service plan needs to be provided for the operator.

[0165] In one possible implementation, multiple alternative service plans can be generated based on the principle of providing a precise service plan.

[0166] As an example, after determining that the service plan is type three, two alternative service plans are generated for an email received in the mailbox, including: Alternative service plan 1 is to give a notification that "email from A has been received" and provide a jump link; Alternative service plan 2 is to give a notification that "email from A has been received" and a brief email content and provide a jump link.

[0167] As an example, after determining that the service plan is type three, if a user wants to modify a program code, three alternative service plans are generated: Alternative service plan 1 provides an analysis method and details how to modify it; Alternative service plan 2 points out the parts of the code that need to be modified; Alternative service plan 3 makes the changes directly without giving the user feedback.

[0168] 602. Based on at least two decision dimensions, determine the expected cognitive benefits of each alternative service plan. The expected cognitive benefits represent the degree to which the operator accepts the alternative service plan.

[0169] This decision-making dimension is used as a reference for deciding on alternative service options. Based on this dimension, an expected cognitive benefit function for each service option can be constructed. This function is then used to calculate the expected cognitive benefit for each alternative service option, allowing for an overall evaluation of the alternative service options based on the expected cognitive benefits.

[0170] In one possible implementation, three decision dimensions can be used: interaction cost, offloading benefit, and reliability. The values ​​of each decision dimension can be generated together with the agent during the process of generating alternative service solutions. The values ​​of interaction cost, offloading benefit, and reliability of each alternative service solution are used as parameters of the alternative service solution.

[0171] The interaction cost can be calculated by designing tasks with different interaction costs through user experiments, and then combining the user's task completion rate based on guidance with the acceptance rate of proactive services. This interaction cost can be represented numerically. The offloading benefit represents the proportion of the current problem solved by the proactive service proposed by the agent. This offloading benefit can also be represented numerically. The reliability can be calculated by combining model confidence and user historical preferences to determine the likelihood that the service will be accepted by the user and solve the problem. This reliability can be quantified numerically.

[0172] As an example, using three decision dimensions—interaction cost, unloading benefit, and reliability—the corresponding expected cognitive benefit function is expressed by the formula: Expected cognitive benefit = (unloading benefit × reliability) - interaction cost.

[0173] Accordingly, using the above formula, the expected cognitive benefits of each alternative service plan can be calculated.

[0174] The expected cognitive benefit can be the degree to which the agent expects to approve of the alternative service plan. The higher the degree of expectation, the more the agent believes that the agent approves of the alternative service plan. Conversely, the lower the degree of expectation, the less the agent believes that the agent approves of the alternative service plan.

[0175] 603. Based on the expected cognitive benefits, determine the target service plan from at least two alternative service plans, wherein the expected cognitive benefits of the target service plan are greater than the expected cognitive benefits of any non-target service plan.

[0176] Since the expected cognitive benefit is the degree to which the agent believes the operator approves of the alternative service plan, in order to provide the agent with a service plan that fits the agent's needs, the one with the greatest expected cognitive benefit can be selected as the target service plan, so as to provide the operator with a proactive service plan.

[0177] Accordingly, by comparing the expected cognitive benefits of each alternative service plan, the one with the highest expected cognitive benefit is selected as the target service plan, in order to ensure that proactive service is a cognitive profit for the operator.

[0178] In one possible implementation, for cognitive states with high cognitive load, a low-cost service solution is preferred, which is typically a direct answer or an automated operation.

[0179] In one possible implementation, after determining the expected cognitive benefits of each alternative service plan, it can be first determined whether the expected cognitive benefits of each alternative service plan are all greater than a preset threshold. This preset threshold is used to define whether the expected cognitive benefits are positive. If there are alternative service plans whose expected cognitive benefits are greater than the preset threshold, the one with the largest expected cognitive benefit is selected as the target service plan. If the expected cognitive benefits of each alternative service plan are less than the preset threshold, indicating no positive benefits, then no alternative plan can be adopted, and no active feedback can be initiated.

[0180] In one possible implementation, since the intelligent agent is not the operator itself, the service solution it proactively provides may or may not match the operator's needs. Whether it matches the operator's needs can be reflected in the operator's feedback operation on the service solution, which can reflect the operator's preference for the amount of information provided by the service solution.

[0181] Accordingly, in order to further improve the effectiveness of the intelligent agent in providing proactive services, the target service scheme can also be selected by combining the characteristic information of the operator.

[0182] In one possible implementation, step 603 includes:

[0183] 6031. Obtain the feature information of the operator. The feature information represents the operator's preference for the amount of information provided by the service information. The feature information is obtained based on the analysis of the operator's historical input information.

[0184] In the process of selecting alternative service solutions, the characteristic information of the operator is taken into account to provide a targeted service solution for the operator of the current electronic device.

[0185] Because different people have different preferences for the amount of information provided by the service plan in the second sub-state. For example, some users want more detailed operation instructions, while others prefer a brief explanation and operation links.

[0186] Accordingly, the characteristic information of the operator is obtained. This characteristic information can be generated by the operator's feedback on historical service plans, and this feedback is the operator's historical input information.

[0187] As an example, the same service plan is provided to both user A and user B. This service plan provides an operation link. User A's historical input information is that they clicked the operation link, while user B's historical input information is that they asked for the operation steps corresponding to the operation link.

[0188] In one possible implementation, accounts logged in by intelligent agents or electronic devices can be used, with different accounts serving as different operating entities, and the characteristic information of the operating entity recorded for each account.

[0189] Correspondingly, when using an intelligent agent to provide services to an operator, the current operator can be identified based on the account logged in by the intelligent agent or electronic device, and then the characteristic information of the current operator can be determined.

[0190] In one possible implementation, the historical input information and historical service plans provided by the operator can be analyzed according to a set period to determine the characteristic information of the operator.

[0191] 6032. Based on expected cognitive benefits and feature information, determine the target service plan from at least two alternative service plans. The degree of matching between the target service plan and the feature information and the expected cognitive benefits are better than the degree of matching between the non-target service plan and the feature information and the expected cognitive benefits.

[0192] After determining the expected cognitive benefits of each alternative, the target service plan can be selected by combining the characteristics of the operator with the expected cognitive benefits.

[0193] In one possible implementation, the target service solution can be selected based on the degree of matching between the service solution and the feature information, as well as the expected cognitive benefits. The degree of matching between the candidate service solution and the feature information of the operator, and the expected cognitive benefits of the candidate service solution, can be calculated separately. Among multiple candidate service solutions, the optimal one is selected.

[0194] In one possible implementation, the degree of matching can be quantified by weighting the degree of matching and the expected cognitive benefits to obtain the final score of each alternative service plan, and then selecting the one with the highest final score as the target service plan.

[0195] In this embodiment, based on the determined service plan being of the third type, at least two alternative service plans are generated. According to at least two decision dimensions, the expected cognitive benefit of each alternative service plan is determined, whereby the expected cognitive benefit characterizes the degree to which the operator accepts the alternative service plan. Based on the expected cognitive benefit, a target service plan is determined from the at least two alternative service plans, where the expected cognitive benefit of the target service plan is greater than the expected cognitive benefit of any non-target service plan. The expected cognitive benefit of each alternative service plan is determined according to multiple decision dimensions, and the one with the largest expected cognitive benefit is selected as the target service plan. The expected cognitive benefit quantifies the degree to which the intelligent agent predicts the operator's acceptance of the alternative service plan. This quantified value is used to select the target service plan. The basis for selecting the target service plan is intuitive and interpretable, improving the accuracy of providing proactive services to the operator.

[0196] Figure 7 This is a flowchart illustrating a service recommendation method provided in this application embodiment, applied to an intelligent agent. The flowchart includes the following steps:

[0197] 701. Obtain the first cognitive state of the operator, and the first cognitive state is the first load cognitive state;

[0198] The first cognitive state can be determined by the deep learning model based on the image analysis of the agent, and the first cognitive state in the determination result is the first load cognitive state, which is then input into the agent as input information.

[0199] 702. Multimodal information comprehensive analysis;

[0200] This multimodal information includes input operation information and contextual information. Combining this multimodal information allows for a more detailed classification of the first cognitive state.

[0201] 703. Determine whether the behavior pattern of the operator is stable;

[0202] Based on the input operation information and context information in the multimodal information, analyze whether the behavior pattern of the operator is stable. If the behavior pattern is stable, proceed to step 704; otherwise, proceed to step 706.

[0203] 704. The state of the operating body is determined to be flow;

[0204] The operator's behavior pattern is stable and it is currently in a state of high cognitive load, indicating that the operator's state is a flow state.

[0205] 705. The service plan is to remain silent in order to protect flow.

[0206] 706. The state of the operating body is determined to be jammed;

[0207] The behavior of this operator is disordered (unstable) and it is currently in a state of high cognitive load, indicating that the operator is stuck.

[0208] 707. Generate multiple alternative service plans;

[0209] Since the operating entity is stuck, it is necessary to provide proactive services to the operating entity, identify multiple alternative service options, and provide prompt information to the operating entity for each alternative service option.

[0210] 708. Rank the alternative service options according to the expected perceived benefits;

[0211] 709. Determine whether the expected cognitive benefits of each alternative service plan are greater than the preset threshold;

[0212] If not, proceed to step 710; if yes, proceed to step 711.

[0213] The preset threshold can be 0, and by passing through the preset threshold, it can be determined whether there is a positive expected cognitive benefit.

[0214] 710. Confirmed that we will not disturb you for the time being;

[0215] "Do not disturb for the time being" means that no service will be provided.

[0216] If the expected cognitive benefits of each alternative service option are not greater than the preset threshold, it indicates that the expected cognitive benefits of each alternative service option are low, and no alternative service option can provide a good result. In this case, no service will be provided to avoid negative impact on the operation.

[0217] 711. Choose the service plan with the highest expected cognitive benefit;

[0218] If the expected cognitive benefits of each alternative service plan are greater than the preset threshold, it indicates that the expected cognitive benefits of each alternative service plan are relatively high, and the one with the highest expected cognitive benefits is selected as the service plan to be adopted.

[0219] 712. Determining the service plan is to provide precise services.

[0220] The service solution with the highest expected cognitive benefit is the one that can provide precise services to the operator, accurately indicating the operator's current stuck state to resolve the current stuck problem.

[0221] Figure 8 This is a schematic diagram of the structure of an intelligent agent provided in an embodiment of this application. The intelligent agent 800 includes a planning brain 801 and an execution brain 802. The division of the two brains in the intelligent agent is based on the functions implemented by the intelligent agent. In the application scenario of this intelligent agent, the operator is the user of the electronic device.

[0222] This multimodal information is input as input information into the planning brain 801 and the execution brain 802.

[0223] The execution brain 802 calculates the user's state score by weighting multimodal information according to a weighted formula. This state score scores each data point in the multimodal information to obtain corresponding parameters. These parameters are then weighted to calculate the user's target parameter (state score). If the state score does not exceed the threshold, it can be determined that the user is in a flow state, and no proactive service is needed. Multimodal information is then collected to determine whether to provide proactive service. If the state score exceeds the threshold, it can be determined that the user is stuck, and proactive service needs to be provided. When providing proactive service, alternative service plans can be generated. The expected cognitive benefits of each alternative service plan are determined from three decision dimensions: offloading benefits, interaction costs, and reliability. Based on guidance information, the service to be provided is selected from the alternative service plans.

[0224] In the planning brain 801, user feedback and received multimodal information are analyzed to determine parameter values. These parameters may include weights in a weighted formula, scoring thresholds, etc. Subsequently, the execution brain 802 calculates state scores for the multimodal information according to the weighted formula and determines whether the current user is in a flow state or stuck based on the updated scoring thresholds. The planning brain 801 can also analyze the provided dialogue history at set intervals to determine user characteristic information. This characteristic information represents the operator's preference for the amount of service information provided, etc., and is used to update guidance information. This updated guidance information allows the selection of services that better match user preferences from alternative options, improving the effectiveness of proactive service provision.

[0225] The above describes a service scheme recommendation method provided by the embodiments of this application. The following will describe an electronic device that performs the above service scheme recommendation method.

[0226] Please see Figure 9 , Figure 9 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. For example... Figure 9 As shown, the electronic device 900 includes: at least two interfaces 901 and a processor 902;

[0227] The at least two interfaces 901 are respectively used to receive input information of at least two modes, wherein the input information represents the cognitive state of the operator of the electronic device;

[0228] The processor 902 is used to determine the cognitive state of the operator based on the input information; and to provide a service plan for the operator based on the cognitive state. Different cognitive states correspond to different service plans, and different service plans provide different amounts of information.

[0229] The processor is a hardware device used to carry the intelligent agent. The processor can be a CPU (central processing unit), an NPU (neural processing unit), etc. This application does not limit the specific form of the processor.

[0230] It should be noted that the specific steps and explanations for the processor's execution function can be found in the explanations in the foregoing method embodiments, and will not be repeated here.

[0231] In this embodiment, the electronic device includes: at least two interfaces, each used to receive input information in at least two modalities, the input information representing the cognitive state of the electronic device's operator; a processor, used to determine the operator's cognitive state based on the input information; and to provide a service plan for the operator based on the operator's cognitive state, with different service plans corresponding to different cognitive states, and different service plans providing different amounts of information. By using multimodal input information, the operator's cognitive state can be determined more accurately, and different information service plans can be provided to the operator for different cognitive states to match the user's need for understanding external information, thereby improving the effectiveness of user assistance and support.

[0232] This application also provides a computer program product including computer-readable instructions, which, when executed on an electronic device, cause the electronic device to implement any of the service scheme recommendation methods provided in this application.

[0233] This application also provides an electronic device, including at least one processor and a memory connected to the processor, wherein:

[0234] Memory is used to store computer programs;

[0235] The processor is used to execute computer programs to enable electronic devices to implement the service scheme recommended method of the first aspect or any implementation of the first aspect described above.

[0236] This application also provides a computer-readable storage medium that carries one or more computer programs. When the one or more computer programs are executed by an electronic device, the electronic device can implement any of the service scheme recommended methods provided in this application.

[0237] It is understood that before using the technical solutions disclosed in the various embodiments of the present invention, users should be informed of the types, scope of use, and usage scenarios of the personal information involved in the present invention and their authorization should be obtained in accordance with relevant laws and regulations through appropriate means.

[0238] For example, upon receiving a user's active request, a prompt message is sent to the user to explicitly inform them that the requested operation will require the acquisition and use of the user's personal information. This allows the user to independently choose whether to provide personal information to the software or hardware, such as the electronic device, application program, server, or storage medium executing the operation of this invention, based on the prompt message.

[0239] As an optional but non-limiting implementation, in response to a user's active request, sending a prompt message to the user can be done via a pop-up window, where the prompt message can be presented in text format. Furthermore, the pop-up window can also include a selection control allowing the user to choose "agree" or "disagree" to provide personal information to the electronic device.

[0240] It is understood that the above notification and user authorization process is merely illustrative and does not constitute a limitation on the implementation of the present invention. Other methods that comply with relevant laws and regulations may also be applied to the implementation of the present invention.

[0241] It is understood that the data involved in this technical solution (including but not limited to the data itself, the acquisition or use of the data) shall comply with the requirements of relevant laws, regulations and related provisions.

[0242] It should also be noted that the device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and 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. In addition, in the device embodiment drawings provided in this application, the connection relationship between modules indicates that they have a communication connection, which can be implemented as one or more communication buses or signal lines.

[0243] Through the above description of the embodiments, those skilled in the art can clearly understand that this application can be implemented by means of software plus necessary general-purpose hardware, or it can be implemented by special-purpose hardware including application-specific integrated circuits, special-purpose CPUs, special-purpose memory, special-purpose components, etc. Generally, any function performed by a computer program can be easily implemented by corresponding hardware, and the specific hardware structure used to implement the same function can also be diverse, such as analog circuits, digital circuits, or special-purpose circuits. However, for this application, software program implementation is more often the preferred implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a readable storage medium, such as a computer floppy disk, USB flash drive, mobile hard disk, ROM, RAM, magnetic disk, or optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, training equipment, or network device, etc.) to execute the methods described in the various embodiments of this application.

[0244] In the above embodiments, implementation can be achieved, in whole or in part, through software, hardware, firmware, or any combination thereof. When implemented in software, it can be implemented, in whole or in part, as a computer program product.

[0245] The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions may be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions may be transmitted from one website, computer, training device, or data center to another website, computer, training device, or data center via wired (e.g., coaxial cable, fiber optic, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium may be any available medium that a computer can store or a data storage device such as a training device or data center that integrates one or more available media. The available media may be magnetic media (e.g., floppy disks, hard disks, magnetic tapes), optical media (e.g., DVDs), or semiconductor media (e.g., solid-state drives (SSDs)).

Claims

1. A service plan recommendation method, applied to electronic devices, comprising: Obtain input information in at least two modalities, wherein the input information characterizes the cognitive state of the operator of the electronic device; The cognitive state of the operator is determined based on the input information; Based on the cognitive state of the operator, a service plan is provided to the operator. Different cognitive states correspond to different service plans, and different service plans provide different amounts of information.

2. The service plan recommendation method according to claim 1, wherein determining the cognitive state of the operator based on the input information includes: Based on the preset analysis model, the first modal information in the input information is analyzed to obtain the first cognitive state of the operator. The first cognitive state includes a first load cognitive state and a second load cognitive state. The cognitive load corresponding to the first load cognitive state is higher than the cognitive load corresponding to the second load cognitive state. Based on the first cognitive state being a first load cognitive state, and according to the second modal information in the input information, the second cognitive state of the operator is determined. The second cognitive state includes a first sub-state and a second sub-state. The first sub-state indicates that the behavior pattern of the operator satisfies the preset behavior pattern stability condition, and the second sub-state indicates that the behavior pattern of the operator does not satisfy the preset behavior pattern stability condition.

3. The service scheme recommendation method according to claim 2, based on the first cognitive state being a first load cognitive state, and according to the second modal information in the input information, determining the second cognitive state of the operator, includes: Analyze the second modal information to obtain the first parameter corresponding to the second modal information; Determine the second parameter corresponding to the first load cognitive state; Based on the first parameter and the second parameter, the target parameters of the operating body are determined; Based on the fact that the value of the target parameter is not greater than a preset threshold, the second cognitive state of the operator is determined to be the first sub-state; Based on the fact that the value of the target parameter is greater than a preset threshold, the second cognitive state of the operator is determined to be the second sub-state.

4. The service scheme recommendation method according to claim 3, wherein analyzing the second modality information to obtain the first parameter corresponding to the second modality information includes: Based on the second modal information containing the operational behavior information of the operator, the second modal information is analyzed to obtain the behavior pattern of the operator; Based on the fact that the behavior pattern satisfies the preset behavior pattern stability condition, the first parameter is determined to be a first value; Based on the fact that the behavior pattern does not meet the preset behavior pattern stability condition, the first parameter is determined to be a second value, and the second value is greater than the first value.

5. The service solution recommendation method according to claim 2, wherein providing a service solution to the operator based on the operator's cognitive state includes: Based on the cognitive state of the operator, the type of service provision plan is determined; Based on the type of service plan provided, a service plan is determined, and a service plan is provided for the operating body.

6. The service plan recommendation method according to claim 5, wherein determining the type of service plan based on the cognitive state of the operator includes: Based on the determination that the cognitive state of the operator is the second load cognitive state, a first type of service plan is determined according to the second load cognitive state; Based on the fact that the cognitive state of the operator is the first sub-state, a second type of service scheme is determined to be provided. The amount of information provided by the second type of service scheme is less than the amount of information provided by the first type of service scheme. Based on the fact that the cognitive state of the operator is the second sub-state, it is determined to provide a third type of service. The amount of information provided by the third type of service is greater than the amount of information provided by the second type and less than the amount of information provided by the first type.

7. The service plan recommendation method according to claim 6, wherein a service plan is determined based on the type of service plan provided, and a service plan is provided to the operating entity, comprising: Based on the fact that the service plan is of the third type, at least two alternative service plans are generated; Based on at least two decision dimensions, the expected cognitive benefit of each alternative service plan is determined, wherein the expected cognitive benefit represents the degree to which the operator recognizes the expected value of the alternative service plan; Based on the expected cognitive benefits, a target service plan is determined from at least two alternative service plans, wherein the expected cognitive benefits of the target service plan are greater than the expected cognitive benefits of any non-target service plan.

8. The service plan recommendation method according to claim 7, wherein determining the target service plan from at least two alternative service plans based on the expected cognitive benefit includes: The feature information of the operator is obtained, which represents the operator's preference for the amount of information provided by the service information. The feature information is obtained based on the analysis of the operator's historical input information. Based on the expected cognitive benefits and the feature information, a target service solution is determined from at least two alternative service solutions. The degree of matching between the target service solution and the feature information, as well as the expected cognitive benefits, are better than the degree of matching between the non-target service solutions and the feature information, as well as the expected cognitive benefits.

9. The service plan recommendation method according to claim 1, wherein obtaining input information in at least two modalities includes: Obtain first modal information, which includes at least two consecutively arranged images, each containing an image of the operator; The second modal information is obtained, which includes input operation information of each input device in the electronic device within a preset time period, at least two display interfaces displayed by the display device in the electronic device within the preset time period, and the display duration of each display interface.

10. An electronic device, comprising: At least two interfaces are provided for receiving input information in at least two modalities, wherein the input information characterizes the cognitive state of the operator of the electronic device. The processor is configured to determine the cognitive state of the operator based on the input information; and to provide a service plan for the operator based on the cognitive state, wherein different cognitive states correspond to different service plans and different service plans provide different amounts of information.

Images