User dynamic state prediction method and system based on space-time relative phase

Abstract

The invention discloses a user dynamic state prediction method and system based on a space-time relative phase, and belongs to the technical field of artificial intelligence and data mining. The method comprises the following steps: acquiring space-time coordinate data of a user terminal, and processing the space-time coordinate data into a standardized space-time feature vector according to a predefined rule; obtaining individual state data of a user, and quantifying the individual state data into a user state feature vector; splicing the spatio-temporal feature vector and the user state feature vector, inputting the spliced spatio-temporal feature vector and the user state feature vector into a pre-trained spatio-temporal phase mapping model, and calculating to obtain a relative phase value used for representing the dynamic interaction relationship between the user and the spatio-temporal environment; user state predictors and / or a set of machine executable behavior intervention instructions are generated based on the relative phase values. By introducing and quantizing the space-time environment parameters, the technical problems that an existing prediction method is single in data dimension and the model statically lags are solved.

Description

Technical Field

[0001] This invention relates to the fields of artificial intelligence, data mining, and computer system architecture, and in particular to a method and system for generating standardized control signals by quantifying the dynamic interaction between users and the spatiotemporal environment. Background Technology

[0002] When building adaptive intelligent systems, existing technologies often struggle to effectively quantify and represent the interaction between dynamic elements (such as users) and the spatiotemporal environment. This is mainly reflected in the following aspects: Data silos and decision-making fragmentation: Information of different modalities, such as user data, environmental data, and system resource data, is fragmented, forcing each application at the upper layer to build its own independent, complex mapping model from raw data to decisions. This approach often leads to redundant computing resources, complex system architecture, and the need to improve collaboration efficiency.

[0003] The lack of regulatory signals: The outputs of existing models (such as classification labels and regression scores) are "descriptive data" for human understanding, rather than "regulatory signals" for machine execution. This leads to a huge "semantic gap" between perception and action, requiring the system to have an additional, complex logical layer for "translation," which may introduce additional decision-making delays and affect the accuracy of control.

[0004] The contradiction between the static nature of the model and the dynamic nature of the environment: Models trained on static data cannot respond in real time to the rapidly changing dynamic system of "human-environment", resulting in rigid system behavior and an inability to make forward-looking and adaptive adjustments when the state undergoes critical transitions.

[0005] Therefore, there is an urgent need in this field for a technical solution that can transform multi-source heterogeneous data into a unified, machine-understandable, and dynamically controllable signal, so as to fundamentally simplify the system architecture, streamline the decision-making chain, and improve the adaptability and efficiency of the entire intelligent ecosystem. Summary of the Invention

[0006] (a) Purpose of the invention The purpose of this invention is to overcome the shortcomings of existing technologies and provide a method and system for predicting user dynamic states based on spatiotemporal relative phase. The core objective of this invention is not to make more accurate predictions, but rather to construct a universal method and system for generating control signals. Its output "relative phase value" is intended as a standardized unit of measurement to address technical problems such as lengthy system decision chains, huge resource consumption, and slow response caused by data modality heterogeneity and semantic gaps.

[0007] (II) Technical Solution To achieve the above-mentioned objectives, the present invention adopts the following technical solution: A method for predicting user dynamic state based on spatiotemporal relative phase, characterized in that it includes: Step S101: Obtain the spatiotemporal coordinate data of the user terminal, wherein the spatiotemporal coordinate data includes geographic latitude and longitude coordinates and corresponding UTC timestamps; Step S102: Based on predefined spatiotemporal parameterization rules, the spatiotemporal coordinate data is processed into a standardized spatiotemporal feature vector; the rules include converting the UTC timestamp into true solar time based on the geographic latitude and longitude coordinates, and extracting its temporal periodic features; Step S103: Obtain the user's individual state data and quantize the individual state data into a user state feature vector; Step S104: The standardized spatiotemporal feature vector is concatenated with the user state feature vector to form a fused feature vector, which is then input into a pre-trained spatiotemporal phase mapping model to calculate a standardized relative phase value that characterizes the dynamic interaction between the user and the spatiotemporal environment. Step S105: Based on the relative phase value, query the preset behavior-instruction mapping rule base, generate and output a set of machine-executable behavior intervention instructions.

[0008] A user dynamic state prediction system based on spatiotemporal relative phase, used to implement the above method, characterized in that it includes: The spatiotemporal data processing module is used to execute steps S101 and S102; The state data quantization module is used to execute step S103; The relation mapping calculation module is used to execute step S104; The result generation and output module is used to execute step S105.

[0009] (III) Beneficial Effects Compared with existing technologies, this invention has the following substantial features and technological advancements: It achieves "dimensional unification" of the control signal: This invention creatively outputs a standardized scalar of "relative phase value," using it as a universal control dimension within the digital ecosystem. This fundamentally eliminates semantic barriers between modules within the system, laying the foundation for building an efficient and collaborative technology stack.

[0010] This triggered an "efficiency revolution in system architecture": by providing unified control signals, diverse upper-layer applications (such as notifications, permissions, content, and resource scheduling) can reuse the same core perception and computing engine, avoiding the redundant construction of "one application, one model" and greatly reducing the overall system complexity and power consumption.

[0011] It endows the system with the gene of "dynamic adaptation": by calculating and responding to changes in the "user-environment" interaction relationship in real time, the system has the ability to leap from "static rule execution" to "dynamic relationship regulation", which significantly improves its robustness and intelligence level in real complex environments.

[0012] (iv) Further explanation of core concepts The core output of this invention—the "relative phase value"—is designed based on dynamic systems theory. In this theory, "phase" is used to describe a specific stage in the periodic or quasi-periodic oscillator's motion cycle.

[0013] This invention views the "user-environment" as a coupled dynamic system. The user's cognitive and behavioral state can be regarded as an internal "oscillator," while the environmental context defined by spatiotemporal coordinates (such as the office environment on weekdays and the home environment at night) constitutes an external "reference oscillation trajectory."

[0014] The “relative phase value” is a continuous scalar that has been standardized (e.g., normalized to the [0,1] interval) and is used to quantify the real-time cooperative relationship or phase difference between the two dynamic systems.

[0015] Specifically, when this value approaches 1, it indicates that the user's behavioral rhythm, cognitive pattern, and the expected "reference trajectory" of the current spatiotemporal environment are in a highly synchronized and mutually beneficial state (e.g., maintaining focus within the work space). Conversely, when this value approaches 0 or drops sharply, it indicates that the user's state and the environmental reference trajectory are in a transitional state of being out of sync, disordered, or rapidly changing.

[0016] Therefore, the "relative phase value" of this invention is not a direct, static description of the user state or the environment state, but a high-order abstraction and real-time measurement of the energy level of the dynamic interaction between the two. This positioning enables it to become a universal and standardized control dimension. Attached Figure Description

[0017] Figure 1 This is a schematic diagram of the system module composition provided in an embodiment of the present invention; Figure 2 This is a schematic diagram of the method flow provided in an embodiment of the present invention. Detailed Implementation

[0018] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. First, a general description of the core training and construction process involved in this invention will be provided: 1. Training of the spatiotemporal phase mapping model: The pre-trained spatiotemporal phase mapping model is obtained through supervised learning. Specifically, a large number of historical data samples of users in different spatiotemporal scenarios are collected. Each sample includes spatiotemporal coordinate data and corresponding individual state data. Based on expert knowledge or through unsupervised learning methods such as clustering, these historical data samples are labeled with relational tags that reflect the degree of "user-environment" collaboration (e.g., qualitatively characterized as "efficient collaboration," "slight deviation," "serious loss of synchronization," etc.). Subsequently, these labeled historical data samples (with their fused feature vectors as input and relational tags as supervision targets) are used to train a selected machine learning model (such as gradient boosting decision trees, support vector machines, neural networks, etc.) to establish a robust nonlinear mapping model from multi-source heterogeneous data to standardized relative phase values.

[0019] 2. Regarding the construction of the behavior-instruction mapping rule base: The behavior-instruction mapping rule base is predefined based on domain knowledge (such as human-computer interaction principles and system optimization goals) or through empirical optimization via A / B testing. Its core is establishing a mapping relationship from different ranges and trends of relative phase values ​​to specific system atomic operations. For example, analysis determines that when the relative phase value remains in a high range, executing the operation of "allocating more computing resources" yields the best positive benefit; while when its value drops rapidly, triggering the operation of "upgrading the security authentication level" most effectively avoids potential risks. These determined mapping relationships are stored in a structured manner, forming the rule base.

[0020] Example 1: Identification and Adaptive Control of High-Efficiency Operating States The purpose of this embodiment is to demonstrate how the present invention can accurately quantify the dynamic interaction relationship between the user and the environment through the complete S101 to S105 method flow, and achieve adaptive and precise adaptation of system resources based on the generated relative phase value.

[0021] The specific implementation process is as follows: In step S101 (data acquisition), taking a user's fixed workstation scenario on a weekday afternoon as an example, the system obtains the precise geographic latitude and longitude coordinates and the corresponding UTC timestamp.

[0022] In step S102 (spatiotemporal characterization), the system converts the UTC timestamp into true solar time based on geographic latitude and longitude coordinates according to predefined spatiotemporal parameterization rules, extracts its periodic features, and generates a standardized spatiotemporal feature vector V_space.

[0023] In step S103 (state characterization), the system collects user behavior data imperceptibly through terminal sensors, including but not limited to application usage frequency, screen interaction data, and typing speed. The observed characteristics are: application activity is consistently concentrated on a few productivity tools, the interaction flow is stable, and the task switching rate is low. This data is quantified into a user state feature vector V_user.

[0024] In step S104 (relationship mapping calculation), V_space and V_user are concatenated to form a fused feature vector, which is then input into a pre-trained spatiotemporal phase mapping model (in this embodiment, a gradient boosting decision tree model is used). The model calculates a relative phase value P = 0.88. This value objectively represents the highly coordinated dynamic interaction between the user's behavior pattern and the spatiotemporal environmental context.

[0025] In step S105 (instruction generation and output), the system queries a preset behavior-instruction mapping rule base based on the relative phase value. Subsequently, it generates and outputs a set of machine-executable behavior intervention instructions, configured to be directly parsed and executed by the operating system kernel or resource manager. Specifically, the generated instruction set includes: a first instruction configured to delay the delivery of system notifications with priority [low, medium] for 30 minutes; and a second instruction configured to dynamically increase the CPU resource quota of the current foreground process by 15%.

[0026] The technical advantage of this embodiment lies in demonstrating a complete technical closed loop from multi-source data perception to the generation of directly executable instructions. Real-world testing shows that this solution can effectively identify and protect users' deep working states. In such scenarios, the overall system power consumption is significantly reduced by avoiding redundant calculations and context switching, and the average CPU utilization rate decreases by more than 20%.

[0027] Example 2: Empowering the Token Economy as a Value Measurement Measure The purpose of this embodiment is to demonstrate that the relative phase value output by the present invention can serve as a reliable measure of the intrinsic value of user data contribution, providing infrastructure for a sustainable digital economy ecosystem.

[0028] The specific implementation process is as follows: In the digital ecosystem, when a user generates native user data (such as behavioral sequences) or user-generated content, the system records the relative phase value P corresponding to the moment the data is generated. The level of this P value reliably reflects the cognitive context value at the time the user's contribution occurs (for example, data contributions made in a highly focused state with P=0.88 are far more valuable than data generated in a distracted state with P=0.30).

[0029] The relative phase value P is provided to a data contribution quantification system, which uses its contribution evaluation engine as the core input parameter to accurately and fraud-resistantly quantify user contributions.

[0030] The technical effect of this embodiment is that it realizes the value measurement leap of data contribution from "quantity" to "quality", so that token incentives can accurately target high-quality contributions that have long-term positive externalities to ecological health, avoid the extensiveness and incentive mismatch of traditional points system, and lay the foundation for building a sustainable digital economy closed loop.

[0031] It should be emphasized that the above embodiments are merely typical application scenarios listed to clearly illustrate the technical solutions of the present invention, and the scope of protection of the present invention is by no means limited to these embodiments. Any solution based on the core idea of ​​the present invention, namely "generating relative phase values ​​for system regulation by fusing environmental and native data", regardless of its specific application field, falls within the scope of protection sought by the present invention.

Claims

1. A method for predicting user dynamic state based on spatiotemporal relative phase, characterized in that, include: Step S101: Obtain the spatiotemporal coordinate data of the user terminal, wherein the spatiotemporal coordinate data includes geographic latitude and longitude coordinates and corresponding UTC timestamps; Step S102: Based on predefined spatiotemporal parameterization rules, the spatiotemporal coordinate data is processed into a standardized spatiotemporal feature vector; the rules include converting the UTC timestamp into true solar time based on the geographic latitude and longitude coordinates, and extracting its temporal periodic features; Step S103: Obtain the user's individual state data and quantize the individual state data into a user state feature vector; Step S104: The standardized spatiotemporal feature vector is concatenated with the user state feature vector to form a fused feature vector, which is then input into a pre-trained spatiotemporal phase mapping model to calculate a standardized relative phase value that characterizes the dynamic interaction between the user and the spatiotemporal environment. Step S105: Based on the relative phase value, query the preset behavior-instruction mapping rule base, generate and output a set of machine-executable behavior intervention instructions.

2. The method according to claim 1, characterized in that, The individual state data mentioned in step S103 includes emotion rating data actively input by the user through the terminal interface.

3. The method according to claim 1, characterized in that, The individual status data mentioned in step S103 includes user behavior data collected imperceptibly through terminal sensors or applications.

4. The method according to claim 3, characterized in that, The user behavior data includes one or more of the following: application usage frequency, screen interaction data, typing speed, and movement trajectory.

5. The method according to claim 1, characterized in that, The pre-trained spatiotemporal phase mapping model mentioned in step S104 is a machine learning model based on gradient boosting decision tree, support vector machine or neural network.

6. The method according to claim 1, characterized in that, The behavioral intervention instruction set described in step S105 is logically represented as a sequence of one or more atomic operation instructions; wherein each atomic operation instruction is uniquely defined by its instruction type and instruction parameters.

7. The method according to claim 1, characterized in that, The behavioral intervention instruction set is configured as a series of atomic operations that are directly parsed and executed by the operating system kernel, resource manager, or application.

8. A system control method based on relative phase values, characterized in that, include: Obtain a relative phase value; The relative phase value is calculated by the method described in any one of claims 1 to 7; The relative phase value is provided as an input signal to a target system; The target system is triggered to perform a preset operation based on the relative phase value.

9. The system control method according to claim 8, characterized in that, The target system is at least one of a notification management system, a computing resource scheduling system, a data privacy permission system, a digital content generation system, or a token economy incentive system.

10. A user dynamic state prediction system based on spatiotemporal relative phase, used to implement the method as described in any one of claims 1 to 7, characterized in that, include: The spatiotemporal data processing module is used to execute steps S101 and S102; The state data quantization module is used to execute step S103; The relation mapping calculation module is used to execute step S104; The result generation and output module is used to execute step S105.

11. 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 7.

12. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the method as described in any one of claims 1 to 7.

Images