An Interaction Adaptation Method and System Based on Personality Parameter Vectors
By generating structured personality parameter vectors and utilizing standardized interfaces and dynamic correction mechanisms, the problems of poor data compatibility and inconsistent interaction strategies caused by unstructured personality assessment results are solved, enabling personalized interaction adaptation and efficient expansion of the AI system.
Patent Information
- Application Number
- CN202611237033.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-08-14
- Publication Date
- 2026-11-17
AI Technical Summary
In existing technologies, personality assessment results are mostly unstructured data, which makes it difficult for downstream AI systems to parse. The interaction strategy parameter configuration lacks quantitative modeling and dynamic correction, and there is a lack of standardized data interfaces, resulting in high system coupling and difficulty in large-scale use.
By acquiring user selection results, a structured personality parameter vector is generated using a classification algorithm and output through a standardized interface. Combined with a dynamic correction mechanism and a mapping rule base, the interaction strategy is automatically adapted and updated in real time.
By converting personality recognition results into structured parameter vectors, the problem of poor data compatibility is solved, and the automatic adaptation and dynamic correction of interaction strategies are realized, reducing system maintenance costs and improving the system's versatility and scalability.
Smart Images

Figure CN123115998A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the fields of artificial intelligence and cognitive computing, specifically to an interaction adaptation method and system based on personality parameter vectors. Background Technology
[0002] With the development of artificial intelligence technology, digital human and intelligent agent interaction scenarios are becoming increasingly common, and personalized interaction based on user characteristics has become a research hotspot. Existing technologies typically use psychological scales to identify users and then apply the results to AI systems.
[0003] However, existing solutions suffer from the following technical shortcomings in practical engineering implementation: First, personality assessment results are mostly natural language reports or unstructured labels, which downstream AI systems cannot directly parse into executable control parameters, resulting in poor data compatibility. This necessitates the addition of a natural language processing module for conversion, increasing system computational consumption and response latency, and hindering large-scale deployment. Second, existing AI systems lack quantitative modeling for interaction strategy parameter configuration. Parameters are statically fixed and cannot be dynamically corrected based on interaction feedback data. This leads to the need to maintain multiple redundant configurations over long-term interactions, and inconsistencies in behavioral performance are likely to occur, increasing system maintenance costs. Finally, the lack of standardized data interfaces and the high coupling between the personality recognition module and the interaction execution module result in low data flow efficiency, limiting the technology's versatility and system scalability.
[0004] Therefore, there is an urgent need for a technical solution that can transform personality recognition results into structured parameter vectors and drive AI systems to achieve personalized interaction adaptation and dynamic correction. Summary of the Invention
[0005] This application provides an interaction adaptation method and system based on personality parameter vectors, which can solve the technical problems in the prior art, such as personality assessment results being mostly unstructured data that is difficult for downstream AI systems to parse, lack of quantitative modeling and dynamic correction of interaction strategy parameter configuration leading to inconsistent long-term interaction behavior, and lack of standardized data interfaces resulting in high system coupling and difficulty in large-scale calling.
[0006] In a first aspect, embodiments of this application provide an interaction adaptation method based on personality parameter vectors, comprising: obtaining a user's selection result for preset test questions; processing the selection result through a pre-stored classification algorithm model to generate data tags representing user characteristics, the data tags including basic personality type, temperament type, and developmental stage status; converting the data tags into a structured personality parameter vector; and outputting the personality parameter vector through a standardized interface; wherein the personality parameter vector includes interaction control parameters for driving a downstream AI system to perform interaction strategy adaptation.
[0007] In conjunction with the first aspect, in one implementation, the data structure of the personality parameter vector includes: a basic information field for storing user identifier and generation time; a personality type field for storing the basic personality type, temperament type, and developmental stage status and their confidence levels; and an interaction adaptation parameter field for storing the interaction control parameters; wherein the interaction control parameters include dialogue style parameters, emotional expression parameters, interaction rhythm parameters, and content preference parameters.
[0008] In conjunction with the first aspect, in one embodiment, the method further includes a dynamic correction step for the personality parameter vector: during multiple rounds of human-computer interaction, collecting user interaction feedback data; comparing the interaction feedback data with the predicted behavior pattern corresponding to the personality parameter vector, and calculating a deviation value; based on the deviation value, correcting the interaction control parameters in the personality parameter vector using a weighted algorithm including weight coefficients, and updating the personality parameter vector for subsequent interaction calls.
[0009] In conjunction with the first aspect, in one implementation, the weighting coefficients are determined according to the following rules: positively correlated with the number of interactions, the more interactions, the greater the contribution of new data to parameter correction; negatively correlated with the difference between the data generation time and the current time, the closer the interaction data is to the current time, the higher the weight; and positively correlated with the data confidence level, the higher the confidence level of the feedback data, the higher the weight.
[0010] In conjunction with the first aspect, in one embodiment, the method further includes an abnormal state detection step: real-time monitoring of changes in the personality parameter vector; triggering an abnormal state marker when a specific parameter combination in the personality parameter vector meets a preset threshold; and performing at least one of the following operations: outputting a warning signal, switching to a preset differentiated interaction mode, and adjusting the current interaction strategy.
[0011] In conjunction with the first aspect, in one implementation, converting the data tags into a structured personality parameter vector includes: accessing a pre-stored mapping rule library, and automatically matching corresponding interaction control parameters from the mapping rule library based on the basic personality types contained in the data tags; wherein the mapping rule library contains the association between data tags and interaction control parameters.
[0012] In conjunction with the first aspect, in one implementation, the association relationship is key-value pair data pre-stored in a database, where the key is the data tag and the value is an interactive control parameter configuration.
[0013] In conjunction with the first aspect, in one implementation, the standardized interface is a network protocol-based application programming interface (API) configured to receive user identification requests and return personality parameter vector data in a structured data format.
[0014] Secondly, embodiments of this application provide an interaction adaptation system based on personality parameter vectors, comprising: an assessment module for obtaining user selection results of preset test questions; a personality recognition module for processing the selection results through a pre-stored classification algorithm model to generate data tags representing user characteristics, the data tags including basic personality type, temperament type, and developmental stage status; a parameter conversion module for converting the data tags into structured personality parameter vectors; and an interface module for outputting the personality parameter vectors through a standardized interface; wherein the personality parameter vectors include interaction control parameters for driving downstream AI systems to perform interaction strategy adaptation.
[0015] In conjunction with the second aspect, in one embodiment, the system further includes a dynamic correction module, used to collect user interaction feedback data during multiple rounds of human-computer interaction, calculate deviation values, and correct and update the interaction control parameters in the personality parameter vector through a weighted algorithm containing weight coefficients.
[0016] In conjunction with the second aspect, in one embodiment, the system further includes an abnormal state detection module, which is used to monitor changes in the personality parameter vector in real time, and trigger an abnormal state marker and interaction strategy adjustment when a specific parameter combination meets a preset threshold.
[0017] Thirdly, embodiments of this application provide an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that the processor executes the program to implement the method as described in the first aspect and any embodiment of the first aspect.
[0018] The beneficial effects of the technical solution provided in this application include: by adopting the above-mentioned interaction adaptation method based on personality parameter vectors, this application can transform personality recognition results into structured, computable personality parameter vectors, and output them through a standardized interface, enabling downstream AI systems (including but not limited to large language model agents and digital humans) to directly call these parameters to achieve automated adaptation of interaction strategies. This solves the technical problem in the prior art that personality assessment results are mostly unstructured data, making it difficult for downstream AI systems to parse and resulting in poor data compatibility. At the same time, this application can update the personality parameter vectors in real time through a dynamic correction mechanism based on interaction feedback data, solving the technical problem that existing AI systems lack quantitative modeling and dynamic correction of interaction strategy parameter configuration, and that inconsistent behavior is prone to occur in long-term interactions. This achieves continuous optimization and zero drift of personality parameters. In addition, through the design of standardized interfaces and mapping rule bases, the coupling between the personality recognition module and the interaction execution module is reduced, achieving the technical effect of one-time recognition and reuse in all scenarios, and improving the versatility and scalability of the system. Attached Figure Description
[0019] Figure 1 This is a flowchart illustrating an embodiment of the interaction adaptation method based on personality parameter vectors in this application; Figure 2 This is a schematic diagram of the functional modules of an embodiment of the interaction adaptation system based on personality parameter vectors in this application. Detailed Implementation
[0020] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present application.
[0021] To make the objectives, technical solutions, and advantages of this application clearer, the embodiments of this application will be described in further detail below with reference to the accompanying drawings.
[0022] Firstly, embodiments of this application provide an interaction adaptation method based on personality parameter vectors. This method is applied to server-side or cloud computing nodes, aiming to solve the technical problem in the prior art that personality assessment results are mostly unstructured text, which downstream AI systems cannot directly parse into control parameters, making it difficult to automatically adapt interaction strategies.
[0023] In one embodiment, reference is made to Figure 1 , Figure 1 This is a flowchart illustrating the first embodiment of the interaction adaptation method based on personality parameter vectors in this application. Figure 1 As shown, the interaction adaptation method based on personality parameter vectors includes: Step S101: Obtain the user's selection results for the preset test questions.
[0024] In this step, the terminal device (such as a smartphone, PC, or VR device) displays a pre-set set of test questions through a graphical user interface. This pre-set set of test questions is stored in a database and contains several questions used to identify user characteristics. The user answers questions on the terminal, which collects the user's selection instructions and generates selection result data. This selection result data is transmitted to the server via a network security protocol (such as HTTPS). Here, the selection results are presented as structured data, such as a mapping set of question IDs and option IDs.
[0025] Step S102: Process the selection results using a pre-stored classification algorithm model to generate data labels representing user characteristics. The data labels include basic personality type, temperament type, and developmental stage status.
[0026] In this step, after receiving the selection result data, the server calls a pre-stored classification algorithm model for calculation. The classification algorithm model can be a rule-tree-based decision model, a trained machine learning classification model, or other algorithm models capable of classification. The model internally stores the mapping weights between question options and feature dimensions.
[0027] Specifically, the model performs weighted calculations on the selection results and outputs classification results across multiple dimensions. The data labels are computer-recognizable enumeration values or codes. For example, basic personality types can be identified as integer codes from Type_01 to Type_09; temperament subtypes can be identified as Boolean values or enumeration values of Open or Closed, where Open and Closed are two subtypes of the temperament dimension; and developmental stage states can be identified as quadrant codes from Quadrant_1 to Quadrant_4. This step transforms the user's responses into standardized data labels, avoiding the ambiguity of natural language descriptions.
[0028] Step S103: Convert the data tags into a structured personality parameter vector.
[0029] In this step, the server constructs a predefined data structure object (i.e., a personality parameter vector). This data structure contains multiple fields, each corresponding to a different dimension of the data label.
[0030] The specific conversion process includes: reading the values of basic personality type, temperament type, and developmental stage status from the data tags and filling them into the corresponding fields of the data structure. Furthermore, this step also includes retrieving the corresponding interaction control parameters (such as dialogue style ID, emotion coefficient value, etc.) from a preset configuration table based on the data tags, and filling the interaction control parameters into the interaction adaptation parameter field of the data structure.
[0031] The resulting personality parameter vector is typically a data packet in JavaScript Object Notation (JSON) or XML format, with clearly defined field types (such as String, Float, Enum) to facilitate direct machine parsing.
[0032] Step S104: Output the personality parameter vector through a standardized interface.
[0033] In this step, the server sends the personality parameter vector to the requester via an application programming interface (API). The standardized interface is defined as an interface based on a network protocol (such as HTTP / HTTPS, gRPC). The interface is configured to receive a request containing a user identifier and return the personality parameter vector.
[0034] The requesting party is a downstream AI system, including but not limited to a large language model intelligent agent, a digital human driving engine, a humanoid robot control system, or an intelligent customer service system. After receiving the personality parameter vector, the downstream AI system parses the interaction control parameters within it and adjusts its content generation strategy (such as adjusting tone, response speed, content depth, etc.) according to the interaction control parameters.
[0035] In this embodiment, the user's test selection results are transformed into standardized data labels using a classification algorithm model, further encapsulated into a structured personality parameter vector, and finally output to the downstream AI system through a standardized API interface. This transforms the personality recognition results from human-readable natural language reports into machine-executable control parameters. This process solves the technical problem in related technologies where the data structures of personality assessment data and AI system control commands are incompatible, preventing direct driving of AI interaction strategies. Simultaneously, the standardized interface enables the generation of personality parameters once and their reuse across multiple scenarios, improving the efficiency and technical feasibility of personalized interaction adaptation in AI systems.
[0036] It should be noted that the execution entity for steps S101 to S104 above can be a single server or a distributed cloud computing cluster. The preset test questions can be dynamically updated, and the classification algorithm model supports version iteration. The authentication method for the standardized interface can adopt token authentication or open authorization (OAuth) authentication to ensure the security of data transmission.
[0037] In one specific embodiment, the specific data structure of the personality parameter vector is shown below: The personality parameter vector is in JSON format and includes fields for three-dimensional mental structure, personality type, developmental stage status, and interaction adaptation parameters. Additionally, as an optional implementation, a basic information field is also included.
[0038] The basic information fields include user ID (user_id), profile version (profile_version), and generation time (generated_at), which are used to identify the user's identity and profile version information.
[0039] The three-dimensional mental structure field is associated with the personality type field, and different personality types correspond to different configurations of three-dimensional mental structure parameters.
[0040] The three-dimensional mental structure fields include: Value orientation parameters are used to characterize the direction and level of a user's value pursuit. In a preferred implementation, these value orientation parameters include directional sub-parameters, intensity sub-parameters, and transcendence sub-parameters.
[0041] Cognitive architecture parameters are used to characterize a user's depth preferences during information processing. In a preferred implementation, these parameters include depth level parameters, tension coefficient parameters, flexibility parameters, abstraction parameters, and system recognition parameters. Thinking tension refers to the fundamental cognitive framework capability formed by integrating thinking dimensions with computing power. The tension coefficient parameter characterizes the comprehensive penetration and computing power of a user's thinking; a higher value indicates stronger coverage and analytical power. In a preferred implementation, the range of values for the tension coefficient parameter is related to the number of dimensions in the user's cognitive architecture.
[0042] Energy state parameters are used to characterize a user's response to internal and external stimuli. In a preferred embodiment, these energy state parameters include sensitivity parameters, stability parameters, and expression style parameters.
[0043] In one specific embodiment, the nine basic personality types are predefined based on configuration combinations of the aforementioned three-dimensional mental structure parameters (value orientation parameters, cognitive architecture parameters, and energy state parameters). Each basic personality type has a unique set of three-dimensional mental structure parameter configuration ranges. As an example, the parameter configurations are shown in Table 1 below: Table 1. Comparison of Enneagram Personality Types and Three-Dimensional Mental Structure
[0044] The numerical ranges in the table are exemplary data, and this application does not limit them. The tension coefficients in the table correspond to the degree of overt tension of an individual when expressing the same mind type, and are included in the cognitive framework parameters. All nine types are compatible with both open and introverted expression styles to reflect the temperament differences of the same mind type in different individuals.
[0045] The personality type field includes: basic personality type (heart_type), whose values include: Pioneer, Sharp-Edged, Experienced, Practical, Agile, Gamer, Craftsman, Flexible, and Self-Aware; and confidence. The growth stage status field includes: Closed Quadrant, Tempering Quadrant, Peak Quadrant, and Special Quadrant.
[0046] The interaction adaptation parameter field (interaction_params) includes: dialogue_style, emotional_style, response_pace, content_preference, depth_preference, initiative_level, and emotional_intensity.
[0047] The temperament subtype is either open or introverted. As a preferred implementation, all basic personality types are compatible with both open and introverted temperament subtypes to reflect the temperament differences of the same basic personality type in different individuals.
[0048] In a preferred embodiment, the cognitive architecture parameters in the three-dimensional mental structure field further include driving force directionality markers, used to characterize the energy projection direction of user behavior. Specifically, the driving force directionality markers include a first directionality and a second directionality; wherein, the first directionality represents energy projected outward, towards the system and society; and the second directionality represents energy converging inward, towards the self and the individual.
[0049] By defining the standardized field structure and data types mentioned above, the technical problems of inconsistent personality data formats and difficulties in downstream system parsing in related technologies are solved. This allows downstream AI systems to directly read parameter values without natural language processing, improving the efficiency of data interaction and the accuracy of machine parsing. Simultaneously, through the quantitative configuration of the three-dimensional mental structure parameters, quantitative modeling of the underlying driving forces of personality is achieved, enabling AI systems to understand the personality differences of different users based on a unified parameter system, rather than relying on superficial behavioral labels.
[0050] Further, in one embodiment, converting data tags into structured personality parameter vectors includes: accessing a pre-stored mapping rule base; automatically matching corresponding interaction control parameters from the mapping rule base based on the basic personality types contained in the data tags; wherein the mapping rule base contains the association relationship between data tags and interaction control parameters. The association relationship is key-value pair data pre-stored in a database, where the key is the data tag and the value is the interaction control parameter configuration.
[0051] In this embodiment, the mapping rule base is stored in the server's non-volatile storage medium (such as a hard disk or solid-state drive) and loaded into the memory cache when the system starts. The key-value pair data is specifically represented as a hash table or a relational database table. When the data tag is received, the system uses the data tag as the index key to query the mapping rule base and obtain the corresponding value, i.e., the interaction control parameter configuration. For example, when the key is a data tag representing a basic personality type (such as Type_01), the corresponding value is "Conversation style = Strategic, Interaction pace = Slow".
[0052] By utilizing a pre-stored mapping rule base and key-value pair matching mechanism, the technical challenge of lacking a standardized mapping mechanism between personality recognition results and AI interaction strategies is resolved. Automated matching reduces manual intervention, enables automated configuration of interaction strategies, and lowers system operation and maintenance costs.
[0053] As a preferred implementation, the mapping rule base stores examples of specific mapping relationships between basic personality types and interaction control parameters, as shown in Table 2 below: Table 2. Specific Mapping Relationship between Basic Personality Types and Interaction Control Parameters
[0054] The aforementioned mapping rule base solves the technical problem of the lack of a standardized mapping mechanism between personality recognition results and AI interaction strategies, enabling the automated and standardized implementation of recognition results, allowing downstream AI systems to obtain personality adaptation capabilities without additional development.
[0055] Furthermore, in one embodiment, the method further includes a dynamic correction step for the personality parameter vector: during multiple rounds of human-computer interaction, user interaction feedback data is collected; the interaction feedback data is compared with the predicted behavior pattern corresponding to the personality parameter vector, and a deviation value is calculated; based on the deviation value, the interaction control parameters in the personality parameter vector are corrected using a weighted algorithm containing weight coefficients, and the personality parameter vector is updated for subsequent interaction calls.
[0056] In this embodiment, the interactive feedback data includes, but is not limited to: dialogue text, user response time, emotional expression characteristics, frequency of active input, topic selection preferences, and explicit evaluation data of the AI output content. Specifically, during data collection, behavioral log data such as user response timestamps, input text length, and frequency of actively initiated dialogues can be recorded.
[0057] The predicted behavior pattern is the range of expected behavior data based on the current personality parameter vector. The deviation value is the absolute value or mean square error of the difference between the actual behavior data and the expected behavior data. The weighting algorithm is specifically a linear weighted or exponentially weighted moving average (EWMA) algorithm. The system dynamically calculates the weight coefficients based on the current interaction round count, the difference between the data generation time and the current time, and the confidence score of the data source. The weight coefficients are determined according to the following rules: positively correlated with the number of interactions (the more interactions, the greater the contribution of new data to parameter correction); negatively correlated with the difference between the data generation time and the current time (interaction data closer to the current time has a higher weight); and positively correlated with data confidence (feedback data with higher confidence has a higher weight).
[0058] The system uses the weight coefficients to iteratively update the interaction control parameters. The updated personality parameter vector is written to the cache or database, overwriting the old version parameters.
[0059] The above-mentioned dynamic correction method solves the technical problem of the accumulation of deviation between the initial personality setting and the actual interaction, and realizes the dynamic tracking and continuous optimization of personality parameters. That is, as the number of interactions increases, the AI system's perception of the user's personality continues to approach the real state, effectively suppressing the personality drift phenomenon in long-term interaction and ensuring the long-term effectiveness of the interaction strategy.
[0060] Furthermore, in one embodiment, the method further includes an abnormal state detection step: real-time monitoring of changes in the personality parameter vector; triggering an abnormal state marker when a specific parameter combination in the personality parameter vector meets a preset threshold; and performing at least one of the following operations: outputting a warning signal, switching to a preset differentiated interaction mode, and adjusting the current interaction strategy.
[0061] Furthermore, the preset exception triggering conditions include, but are not limited to, the following rules: Rule 1: When the driving force orientation of the cognitive architecture parameters in the three-dimensional mental structure field is marked as the second orientation, if the deviation between the depth level parameter and the target level parameter in the cognitive architecture parameters exceeds a preset threshold, an anomaly is triggered. Rule 2: When the state parameters of the growth stage change by more than a preset amount within a preset number of consecutive times (e.g., falling from the "peak quadrant" to the "closed quadrant"), an anomaly is triggered. Rule 3: When the combined change of several driving force parameters exceeds a preset threshold within a preset time window, an anomaly is triggered.
[0062] In this embodiment, the real-time monitoring is implemented through a background daemon process or a scheduled task. The specific parameter combination includes, for example, multiple jumps in the "growth stage state" within a short period, or "emotional expression parameters" exceeding the normal value range. The preset threshold is stored in a configuration file. When the detected parameter change exceeds the preset threshold, the system generates an abnormal state marker and sends a warning signal to the administrator via a message queue, or automatically switches the interaction mode of the downstream AI system to a preset differentiated interaction mode.
[0063] The aforementioned abnormal state detection methods enable early identification and automated response to changes in user personality states, resolving the technical challenges of failing to identify systemic risks early and the inability of interaction strategies to adapt to unforeseen circumstances. Automated early warning and strategy switching enhance the security and proactive care capabilities of the AI system, reducing systemic risks arising from abnormal interactions.
[0064] Furthermore, in one embodiment, the standardized interface is a network protocol-based application programming interface (API) configured to receive user identification requests and return personality parameter vector data in a structured data format.
[0065] Preferably, the specific definition and response data example of the standardized interface are as follows: The API request method is GET, the request path is / api / v1 / user / persona / {user_id}, and the request header contains authentication information Authorization:Bearer{api_key}.
[0066] The interface response data is in JSON format and includes a status code (code) and a data payload (data). The data payload specifically includes the following fields: user_id: User identifier; profile_version: Configuration file version; generated_at: Generation time; three_dim_mind: Three-dimensional mental structure parameters, including value_orientation, cognitive_architecture, and energy_state; heart_type: Basic personality type, including type primary and confidence level; temperament: Inner temperament classification, including type and confidence; growth_quadrant: growth stage status, including type and confidence; interaction_params: Interaction adaptation parameters, including dialogue_style, emotional_style, response_pace, content_preference, depth_preference, initiative_level, and emotional_intensity.
[0067] In this embodiment, the network protocol is Hypertext Transfer Protocol (HTTP / HTTPS), and the application programming interface (API) is a RESTful API. The user identification request is an HTTP GET or POST request, and the request header contains an Authorization Token. The structured data format is JSON. The interface response body contains a status code and a data payload, which is the personality parameter vector. The downstream AI system obtains the interaction control parameters by parsing the JSON data payload.
[0068] Through the aforementioned standardized API interface, downstream AI systems (including but not limited to large language model agents, digital humans, humanoid robots, and intelligent customer service systems) can obtain user personality parameter vectors through a unified interface and automatically adapt interaction strategies based on these parameter vectors. The standardized API interface based on network protocols and the JSON data format solves the technical problems of scalable application of personality recognition results and difficulties in cross-system calls in related technologies. The standardized interface definition decouples the personality recognition module from the interaction execution module, supports modular deployment across platforms and systems, and enables efficient reuse of personality parameters.
[0069] Secondly, embodiments of this application also provide an interaction adaptation system based on personality parameter vectors.
[0070] In one embodiment, reference is made to Figure 2 , Figure 2 This is a schematic diagram of the functional modules of an embodiment of the interaction adaptation system based on personality parameter vectors according to this application. Figure 2 As shown, the interactive adaptation system based on personality parameter vectors includes assessment module 1, personality recognition module 2, parameter conversion module 3, and interface module 4.
[0071] The assessment module 1 is used to obtain the user's selection results for preset test questions; the personality recognition module 2 is used to process the selection results through a pre-stored classification algorithm model to generate data labels representing user characteristics, including basic personality type, temperament type and developmental stage status; the parameter conversion module 3 is used to convert the data labels into a structured personality parameter vector; the interface module 4 is used to output the personality parameter vector through a standardized interface so that the downstream AI system can adjust the interaction strategy based on the interaction control parameters.
[0072] The functional implementation of each module in the above-mentioned interaction adaptation system based on personality parameter vectors corresponds to each step in the above-mentioned interaction adaptation method embodiment based on personality parameter vectors. Their functions and implementation processes will not be described in detail here.
[0073] Thirdly, embodiments of this application provide 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 aforementioned interaction adaptation method based on personality parameter vectors.
[0074] It should be noted that the sequence numbers of the embodiments in this application are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.
[0075] The terms "comprising" and "having," and any variations thereof, in the specification, claims, and accompanying drawings of this application are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or apparatus that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to such process, method, product, or apparatus. The terms "first," "second," and "third," etc., are used to distinguish different objects, etc., and do not indicate a sequence, nor do they limit "first," "second," and "third" to different types.
[0076] In the description of the embodiments of this application, terms such as "exemplary," "for example," or "for instance" are used to indicate examples, illustrations, or explanations. Any embodiment or design described as "exemplary," "for example," or "for instance" in the embodiments of this application should not be construed as being more preferred or advantageous than other embodiments or designs. Specifically, the use of terms such as "exemplary," "for example," or "for instance" is intended to present the relevant concepts in a concrete manner.
[0077] In the description of the embodiments of this application, unless otherwise stated, " / " means "or". For example, A / B can mean A or B. The "and / or" in the text is merely a description of the relationship between related objects, indicating that there can be three relationships. For example, A and / or B can mean: A exists alone, A and B exist simultaneously, and B exists alone. In addition, in the description of the embodiments of this application, "multiple" means two or more.
[0078] In some processes described in the embodiments of this application, multiple operations or steps are included in a specific order. However, it should be understood that these operations or steps may not be executed in the order they appear in the embodiments of this application, or they may be executed in parallel. The sequence number of the operation is only used to distinguish different operations, and the sequence number itself does not represent any execution order. In addition, these processes may include more or fewer operations, and these operations or steps may be executed sequentially or in parallel, and these operations or steps may be combined.
[0079] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better 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 storage medium (such as ROM / RAM, magnetic disk, optical disk) as described above, and includes several instructions to cause a terminal device to execute the methods described in the various embodiments of this application.
[0080] The above are merely preferred embodiments of this application and do not limit the patent scope of this application. Any equivalent structural or procedural transformations made using the content of this application's specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of this application.
Claims
1. An interaction adaptation method based on personality parameter vectors, characterized in that, include: Obtain the user's selection results for preset test questions; The selection results are processed by a pre-stored classification algorithm model to generate data labels that represent user characteristics. The data labels include basic personality type, temperament type and growth stage status. The data tags are converted into a structured personality parameter vector, which contains interaction control parameters for driving the downstream AI system to adapt interaction strategies. The personality parameter vector is output through a standardized interface so that downstream AI systems can adjust their interaction strategies based on the interaction control parameters.
2. The method according to claim 1, characterized in that, The data structure of the personality parameter vector includes: The basic information field is used to store the user identifier and the generation time; The personality type field is used to store the basic personality type, temperament type, and developmental stage status and their confidence level. The interaction adaptation parameter field is used to store the interaction control parameters; The interaction control parameters include dialogue style parameters, emotion expression parameters, interaction rhythm parameters, and content preference parameters.
3. The method according to claim 1, characterized in that, The method also includes a dynamic correction step for the personality parameter vector: During multiple rounds of human-computer interaction, user interaction feedback data is collected; The interactive feedback data is compared with the predicted behavior pattern corresponding to the personality parameter vector, and the deviation value is calculated. Based on the deviation value, the interaction control parameters in the personality parameter vector are corrected by a weighted algorithm that includes weight coefficients, and the personality parameter vector is updated for subsequent interaction calls.
4. The method according to claim 3, characterized in that, The rules for determining the weighting coefficients satisfy the following: It is positively correlated with the number of interactions; the more interactions, the greater the contribution of new data to parameter correction. It is negatively correlated with the difference between the data generation time and the current time, and the interaction data that is closer to the current time has a higher weight; It is positively correlated with the confidence level of the data; the higher the confidence level, the higher the weight of the feedback data.
5. The method according to claim 1, characterized in that, The method also includes an abnormal state detection step: Real-time monitoring of changes in the personality parameter vector; When a specific combination of parameters in the personality parameter vector meets a preset threshold, an abnormal state marker is triggered. Perform at least one of the following operations: output a warning signal, switch to a preset differentiated interaction mode, or adjust the current interaction strategy.
6. The method according to claim 1, characterized in that, The process of converting the data tags into a structured personality parameter vector includes: Access the pre-stored mapping rule base, and automatically match the corresponding interaction control parameters from the mapping rule base based on the basic personality type contained in the data tags; The mapping rule base contains the association between data tags and interactive control parameters.
7. The method according to claim 6, characterized in that, The association relationship is key-value pair data pre-stored in the database, where the key is the data tag and the value is the interactive control parameter configuration.
8. The method according to claim 1, characterized in that, The standardized interface is a network protocol-based application programming interface (API) configured to receive user identification requests and return personality parameter vector data in a structured data format.
9. An interactive adaptation system based on personality parameter vectors, characterized in that, include: The assessment module is used to obtain the user's selection results for preset test questions; The personality recognition module is used to process the selection results through a pre-stored classification algorithm model to generate data tags that represent user characteristics. The data tags include basic personality type, temperament type and developmental stage status. The parameter conversion module is used to convert the data tags into a structured personality parameter vector; The interface module is used to output the personality parameter vector through a standardized interface, so that the downstream AI system can adjust the interaction strategy based on the interaction control parameters; The personality parameter vector includes interaction control parameters used to drive the downstream AI system to adapt interaction strategies.
10. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the method as described in any one of claims 1 to 8.
