User information management method based on internet of things big data

Abstract

The present application relates to the technical field of internet of things data processing and user behavior analysis management, and discloses a user information management method based on internet of things big data, comprising: obtaining device information used by a user from a plurality of internet of things devices; time stamping and time stamping synchronization of the device information to form an event sequence containing device identifiers and time stamps; analyzing the data interaction relationship between different internet of things devices in the event sequence through a time series association network algorithm, calculating the time series similarity between events, constructing an association matrix, and forming a graph structure representation; analyzing the user's behavior feedback from the graph structure representation, generating an instant reward, and according to the current state of device use and the generated instant reward, generating a personalized behavior management method for the user through a Q value optimization strategy. The problems of data heterogeneity, time desynchronization, lack of deep behavior analysis and personalized management in the prior art are solved.

Description

Technical Field

[0001] This invention relates to the technical field of Internet of Things (IoT) data processing and user behavior analysis and management, and in particular to a user information management method based on IoT big data. Background Technology

[0002] With the rapid development of the Internet of Things (IoT), the number of connected devices worldwide is growing exponentially, covering various fields from smart home devices and wearable devices to industrial sensors and autonomous vehicles. These devices generate massive amounts of data, collectively known as IoT big data. How to effectively manage and analyze this data has become a hot topic in current research and application. However, existing methods for managing user information on IoT devices have significant shortcomings in several aspects.

[0003] First, the diverse types and functions of IoT devices, along with their varying data formats and communication protocols, make it difficult to integrate and process multi-source heterogeneous data. Devices from different manufacturers may use different communication standards and data formats, and the metadata, attributes, and structures contained in the data also differ. This heterogeneity complicates the standardization and unified management of data, making it difficult to build a comprehensive user information model. Existing methods often fail to effectively integrate data from different devices, resulting in one-sided and incomplete user behavior information, which hinders comprehensive and accurate analysis of user behavior.

[0004] Secondly, the inconsistency of timestamps between devices is another problem that urgently needs to be solved. Since IoT devices may be distributed in different network environments, the timestamps recorded by each device may deviate due to limitations in device hardware performance, network latency, and different time synchronization mechanisms. This time synchronization problem will directly affect the accuracy of time series analysis and will not be able to correctly reconstruct the user's operation sequence and behavior trajectory on different devices. Traditional methods usually ignore the importance of time synchronization or rely on simple time correction, which cannot meet the requirements of high-precision time series analysis.

[0005] Furthermore, existing user behavior modeling methods are insufficient in mining the interaction relationships between users and multiple devices. Traditional user behavior analysis is mostly based on single device or single type of data, lacking a comprehensive understanding of user behavior across devices and scenarios. There is a lack of effective algorithms to mine the interaction relationships and user behavioral characteristics between different devices, and it is impossible to capture the comprehensive usage patterns of users on different devices. This results in limited capabilities in user profile construction, personalized recommendations, and intelligent decision support, and it is impossible to provide users with accurate personalized services.

[0006] Furthermore, with the growth in the number of IoT devices and the scale of data, traditional data processing methods face challenges in terms of efficiency and performance. Real-time processing and analysis of large-scale data requires efficient algorithms and powerful computing resources. However, existing methods have failed to fully consider the characteristics of IoT big data in terms of algorithm design and system architecture, such as high frequency, low latency and data streaming processing requirements. As a result, data processing latency is high, resource consumption is large, and it is difficult to meet the requirements of real-time performance and efficiency, which affects user experience and system responsiveness.

[0007] In summary, existing methods for managing user information on IoT devices have significant shortcomings, including difficulties in data fusion, inaccurate time synchronization, insufficient user behavior modeling, low processing efficiency, and lack of security and privacy protection. To address these issues, a method is needed that can efficiently manage and analyze user information across multiple IoT devices. Summary of the Invention

[0008] The purpose of this section is to outline some aspects of the embodiments of the present invention and to briefly introduce some preferred embodiments. Some simplifications or omissions may be made in this section, as well as in the abstract and title of the present application, to avoid obscuring the purpose of this section, the abstract and title of the invention. Such simplifications or omissions shall not be used to limit the scope of the present invention.

[0009] In view of the aforementioned existing problems, the present invention is proposed.

[0010] To solve the above-mentioned technical problems, the present invention provides the following technical solution: obtaining device information used by a user from multiple Internet of Things devices, wherein the device information includes user identity information and device usage behavior information, and each device is identified by a unique device identifier;

[0011] The device information is timestamped and synchronized to form an event sequence containing device identifiers and timestamps;

[0012] The temporal correlation network algorithm is used to analyze the data interaction relationship between different IoT devices in the event sequence, calculate the temporal similarity between events, construct the correlation matrix, and form a graph structure representation.

[0013] The user's behavioral feedback is analyzed from the graph structure representation to generate instant rewards. Based on the current state of device usage and the generated instant rewards, a personalized management method is generated for the user through a Q-value optimization strategy.

[0014] As a preferred embodiment of the user information management method based on IoT big data described in this invention, the IoT devices used by users are bound to corresponding user IDs, and device information is collected through sensor technology and network communication to record device status and user operation behavior; wherein:

[0015] The user identity information includes username and user ID;

[0016] The device usage behavior information includes the device's on / off status, usage frequency, and operation method;

[0017] The device identifier is a unique identifier for each device, assigned by the device's hardware manufacturer or generated through a unique algorithm.

[0018] As a preferred embodiment of the user information management method based on IoT big data described in this invention, the device on / off status records the on / off state of the device at different time points, the usage frequency records the number of times the device is used or the active duration of the device within a certain time period, and the operation mode records the interaction mode between the user and the device.

[0019] As a preferred embodiment of the user information management method based on IoT big data described in this invention, the collected device information is used to form an initial dataset containing device identifiers, user identity information, and device behavior information, which is then stored in a structured manner. The mathematical expression of the initial dataset is as follows:

[0020] D i =(ID) device ,{Data behavior Data user})

[0021] Among them, D i For the data of the i-th device, ID device For device identifiers, Data behavior Data for device usage behavior. user For user identity data.

[0022] As a preferred embodiment of the user information management method based on Internet of Things big data described in this invention, the event sequence is obtained as follows:

[0023] The local time of each device is synchronized to a unified standard time using the PTP protocol.

[0024] Each device data record, after being updated with a timestamp, forms an event sequence containing a device identifier, a timestamp, and device behavior data. Each event record contains the device identifier, the timestamp after synchronization, and the device's behavior information.

[0025] The behavioral data of the device includes at least the device's on / off status, usage frequency, and operation mode.

[0026] As a preferred embodiment of the user information management method based on Internet of Things big data described in this invention, the mathematical expression formula for the event sequence is:

[0027] E t ={(ID) device timestamp t Data t )}

[0028] Among them, E t Let ID represent the sequence of events at time t. device A unique identifier for the device, timestamp t The synchronized timestamp represents the standard time when the device action occurred. (Data) t This refers to the device's behavioral data, which includes device behavior information.

[0029] As a preferred embodiment of the user information management method based on IoT big data described in this invention, the temporal similarity between two devices is calculated based on the timestamps and device behavior information between devices, and an event association matrix is ​​constructed. Each element of the event association matrix represents the similarity between two events. The higher the similarity between events, the stronger the correlation between the behaviors of the two devices.

[0030] The formula for calculating the temporal similarity is:

[0031]

[0032] ΔT=|Timestamp i -Timestamp j |

[0033] Where ΔT represents the time difference between device i and device j, and Adj i,j α is the element of the association matrix between device i and device j, α is a hyperparameter that adjusts the effect of time difference, and Behavior Similarity is the behavioral similarity, which is 1 if the behaviors are the same and 0 if the behaviors are different.

[0034] As a preferred embodiment of the user information management method based on IoT big data described in this invention, device data is fused and analyzed through graph convolution operations to construct a graph structure representation. In this representation, device identifiers are used as nodes of the graph, and temporal associations are used as edges to connect the nodes. The weight of each edge represents the temporal similarity between devices. The higher the similarity between each pair of devices, the greater the weight of the edge, indicating that the behavior is more similar.

[0035] As a preferred embodiment of the user information management method based on IoT big data described in this invention, the personalized management method includes controlling the device switch, switching device modes, and adjusting the configuration of the device bound to the user.

[0036] The beneficial effects of this invention are:

[0037] 1. It can accurately associate users with the various devices they use, avoiding data inaccuracies caused by device confusion or data duplication, and achieving the beneficial effect of providing a high-quality and reliable data foundation for subsequent data analysis and user behavior modeling;

[0038] 2. It can accurately reflect the interaction sequence and timing relationship between devices, avoiding data analysis deviations caused by inconsistent device timing, thus achieving the beneficial effects of improving the accuracy of timing analysis and ensuring data synchronization;

[0039] 3. It has achieved the modeling of complex interaction relationships between multiple devices, and has quantified temporal similarity into the form of matrices and graphs, making data relationships more intuitive and computable. This has achieved the beneficial effect of revealing hidden connections between devices and improving the depth of user behavior analysis.

[0040] 4. It has successfully introduced reinforcement learning into user behavior management. By analyzing actual user feedback, it continuously optimizes device management strategies, achieving the beneficial effects of improving user experience, optimizing resource utilization, and realizing personalized services. Attached Figure Description

[0041] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. Wherein:

[0042] Figure 1 This is a flowchart illustrating the user information management method based on Internet of Things big data as shown in this invention.

[0043] Figure 2 This is a schematic diagram comparing the accuracy of behavior recognition as shown in this invention;

[0044] Figure 3 This is a schematic diagram comparing energy consumption as shown in the present invention;

[0045] Figure 4 This is a schematic diagram showing the comparison of data processing efficiency as presented in this invention. Detailed Implementation

[0046] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments.

[0047] Based on the embodiments of this invention, all other embodiments obtained by those skilled in the art without inventive effort should fall within the scope of protection of this invention.

[0048] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0049] According to an embodiment of the present invention, in combination Figure 1 The flowchart shown illustrates a user information management method based on IoT big data, which specifically includes the following steps:

[0050] S1. Obtain device information used by the user from multiple IoT devices. This device information includes user identity information and device usage behavior information. Each device is identified by a unique device identifier. Note that the following points should be noted in this step:

[0051] Bind the IoT devices used by users to the corresponding user IDs. For example, user A uses device 1 and user B uses device 2.

[0052] The device identifier is a unique identifier for each device, assigned by the device's hardware manufacturer or generated by a unique algorithm. For example, the identifier for device 1 is 001, and the identifier for device 2 is 002.

[0053] User identity information includes username and user ID;

[0054] As an example, user A's username is "Alice" and user ID is 001, and user B's username is "Bob" and user ID is 002. The binding relationship is as follows: device 1 is bound to user A (ID 001), and device 2 is bound to user B (ID 002).

[0055] Collect equipment information and record equipment status and user operation behavior through sensor technology and network communication;

[0056] As an example, the behavior records of device 1 (used by user A) and device 2 (used by user B) during a certain period of time are as follows:

[0057] Device 1 (User A) Behavior Log:

[0058] Device on / off status: On at 08:00 on 2024-11-01, Off at 10:00 on 2024-11-01, On at 14:00 on 2024-11-01;

[0059] Usage frequency: Device 1 was turned on 3 times within November 1, 2024;

[0060] Operation method: Device 1 is turned on and off by user A by clicking a button;

[0061] Device 2 (User B) Behavior Log:

[0062] Device on / off status: On at 07:30 on 2024-11-01, Off at 12:30 on 2024-11-01;

[0063] Usage frequency: Device 2 was turned on twice within November 1, 2024;

[0064] Operation method: Device 2 is turned on and off via voice control;

[0065] The recorded device usage behavior information includes device on / off status, usage frequency, and operation mode. Device on / off status records the on / off status of the device at different points in time. Usage frequency records the number of times the device is used or the active duration of the device within a certain period of time. Operation mode records the interaction methods between the user and the device, such as button clicks, voice control, and automatic mode.

[0066] The collected device information is used to form an initial dataset containing device identifiers, user identity information, and device behavior information, which is then stored in a structured manner.

[0067] As an example, each device information data record contains a device identifier (Device ID), user identity information (such as user ID, username, device association information), and device behavior information (such as on / off status, usage frequency, and operation method). The mathematical expression of the initial dataset formed by these elements is as follows:

[0068] D i =(ID) device ,{Data behavior Data user})

[0069] Among them, D i For the data of the i-th device, ID device For device identifiers, Data behavior Data for device usage behavior. user For user identity data;

[0070] For example, data from device 1 (user A):

[0071] D1 = (ID) device =001,{Data behavior ={state={2024-11-01 08:00:00:ON,2024-11-01 10:00:00:OFF,2024-11-01 14:00:00:ON},frequency=3,operation=buttonclick},Data user ={uesrID=001,username=Alice,deviceAssociation={device1}}})

[0072] Device 2 (User B) data:

[0073] D2 = (ID) device =002,{Data behavior ={state={2024-11-01 07:30:00:ON,2024-11-01 12:30:00:OFF},frequency=2,operation=voicecontrol},Datau ser ={userID=002,username=Bob,deviceAssociation={device2}}})

[0074] By using the data structure described above to store and query user and device behavior information, each record includes the device identifier, device behavior data, and user identity information, forming a complete structured dataset. Its mathematical expression is as follows:

[0075] D = {D1,D2,…,D} n}

[0076] Among them, each D n These are data records containing device behavior data and user identity information, ensuring that user and device interactions are recorded in detail, facilitating subsequent analysis and the generation of personalized management strategies.

[0077] S2. Each device generates a local timestamp when collecting data and synchronizes the time of all devices to a unified standard time via the PTP protocol. Devices adjust their local time by exchanging synchronization messages. A synchronized timestamp is added to each device data record, forming an event sequence containing the device identifier and timestamp. It is important to note the following in this step:

[0078] When IoT devices collect data, each device generates a local timestamp to indicate the specific time when the device collected the data. Since there is a clock deviation between devices, this embodiment of the invention uses the PTP protocol (Precise Time Protocol) to synchronize the local time of each device to a unified standard time (e.g., UTC).

[0079] For example, suppose device 1, device 2, and device 3 collect data at different time points, and their local timestamps are as follows:

[0080] Device 1: Local timestamp = 2024-11-01 08:00:00 (Device 1's local time);

[0081] Device 2: Local timestamp = 2024-11-01 08:00:10 (Device 2's local time);

[0082] Device 3: Local timestamp = 2024-11-01 08:00:20 (Device 3's local time);

[0083] Through the PTP protocol, devices exchange synchronization messages and adjust their respective clocks to synchronize the local time of all devices to a unified standard time (e.g., UTC time). The synchronized standard time is:

[0084] Standard Time (UTC) = 2024-11-01 08:00:00

[0085] Device 1, Device 2, and Device 3 will adjust their local time according to the PTP protocol to ensure that their timestamps are aligned as follows:

[0086] Device 1: Synchronized timestamp = 2024-11-01 08:00:00 (UTC);

[0087] Device 2: Synchronized timestamp = 2024-11-01 08:00:00 (UTC);

[0088] Device 3: Synchronized timestamp = 2024-11-01 08:00:00 (UTC);

[0089] Each device data record, after being updated with a timestamp, forms an event sequence containing a device identifier, a timestamp, and device behavior data. Each event record contains the device identifier, the updated timestamp, and the device's behavior data (such as device on / off status, usage frequency, and operation mode).

[0090] As an example, the behavioral data collected by Device 1, Device 2, and Device 3 at the standard time after synchronization are as follows:

[0091] Device 1: Device identifier is 001, and the synchronized timestamp is 2024-11-01 08:00:00 (UTC). Device behavior data includes device activation and button click operations.

[0092] Device 2: Device identifier is 002, and the synchronized timestamp is 2024-11-01 08:00:00 (UTC). Device behavior data includes device shutdown and voice control operations.

[0093] Device 3: Device identifier is 003, and the synchronized timestamp is 2024-11-01 08:00:00 (UTC). Device behavior data includes device startup and automatic mode.

[0094] Each device record contains a device identifier, a synchronized timestamp, and device behavior data. Therefore, the mathematical formula for representing the event sequence is:

[0095] E t ={(ID) device ,timestam p t,Data t )}

[0096] Among them, E t Let ID represent the sequence of events at time t. device A unique identifier for the device, timestamp t The synchronized timestamp represents the standard time when the device action occurred. (Data) t This refers to the behavioral data of the device, including device behavior information (such as device status and operation mode);

[0097] In an optional implementation, the device behavior data collected at time t=1 is defined as follows:

[0098] Event log for device 1:

[0099] E1 = {(ID device =001, timestamp1 = 2024-11-01 08:00:00UTC, Data1 = {state = ON, operation = buttonclick})}

[0100] Event log for device 2:

[0101] E2 = {(ID device =002, timestamp2 = 2024-11-01 08:00:00UTC, Data2 = {state = OFF, operation = voicecontrol})}

[0102] Event log for device 3:

[0103] E3 = {(ID device =003, timestamp3=2024-11-01 08:00:00UTC, Data3={state=ON, operation=automode})}

[0104] Ultimately, the event logs from all devices will be aggregated into a structured event sequence, sorted and stored according to device identifiers and timestamps for subsequent querying and processing;

[0105] The summarized event sequence is as follows:

[0106] E = {E1, E2, E3}

[0107] Among them, E1, E2, and E3 represent the event records of each device at different points in time. Each event record contains a device identifier, a synchronized timestamp, and device behavior data.

[0108] Furthermore, the event sequence E is stored in the database for subsequent analysis, processing, and optimization. The event record storage format is shown in Table 1.

[0109] Table 1. Event Log Table

[0110] Device Identifier Timestamp (UTC) Device behavior information 001 2024-11-01 08:00:00UTC Turn on the device by clicking the button. 002 2024-11-01 08:00:00UTC Turn off the device; voice control operation. 003 2024-11-01 08:00:00UTC Turn on the device, automatic mode

[0111] S3. Analyze the data interaction relationships between different IoT devices in the event sequence using a temporal association network algorithm, calculate the temporal similarity between events, construct an association matrix, and form a graph structure representation. Note that the following points should be noted in this step:

[0112] Based on timestamps and device behavior data between devices, the temporal similarity between two devices is calculated, and an event association matrix is ​​constructed. Each element of the matrix represents the similarity between two events. The higher the similarity between events, the stronger the correlation between the behaviors of the two devices.

[0113] As an example, three devices i, j, and k are defined, and their behavior at different time points is shown in Table 2:

[0114] Table 2. Equipment Usage Record Form

[0115] equipment Timestamp Behavior i 2024-11-01 08:00:00 Turn on the device j 2024-11-01 08:00:05 Turn on the device k 2024-11-01 08:00:10 Turn off the device

[0116] The formula for calculating temporal similarity is:

[0117]

[0118] ΔT=|Timestamp i -Timestamp j |

[0119] Where ΔT represents the time difference between device i and device j, and Adj i,j α is the element of the association matrix between device i and device j, α is a hyperparameter that adjusts the effect of time difference, and BehaviorSimilarity is the behavioral similarity, which is 1 if the behaviors are the same and 0 if the behaviors are different.

[0120] For example, the calculation process for device i and device j is as follows:

[0121] Time difference:

[0122] ΔT ij =|2024-11-01 08:00:00-2024-11-01 08:00:05| = 5 seconds

[0123] Behavioral similarity: Since the devices that are turned on are the same, the behavioral similarity is 1.

[0124] Temporal similarity:

[0125]

[0126] For example, the calculation process for device i and device k is as follows:

[0127] Time difference:

[0128] ΔT ik =|2024-11-01 08:00:00-2024-11-01 08:00:10| = 10 seconds

[0129] Behavioral similarity: Turning the device on and turning it off are different, so the behavioral similarity is 0;

[0130] Temporal similarity:

[0131]

[0132] For example, the calculation process for device j and device k is as follows:

[0133] Time difference:

[0134] ΔT jk =|2024-11-01 08:00:05-2024-11-01 08:00:10|=5 seconds

[0135] Behavioral similarity: Turning the device on and turning it off are different, so the behavioral similarity is 0;

[0136] Temporal similarity:

[0137]

[0138] Specifically, based on the above calculation results of temporal similarity, an event association matrix Adj is constructed:

[0139]

[0140] Each element Adj in the matrix i,j This represents the temporal similarity between device i and device j, for example, Adj i,j =0.952, indicating that device i and device j have a high degree of behavioral similarity, while Adj i,k =0 indicates that device i and device k have no similar behavior;

[0141] Furthermore, device data is fused and analyzed through graph convolution operations to construct a graph structure representation, where device identifiers are used as nodes in the graph; for example, device i is defined as ID. i Device j is ID j Device k is ID k Then, the temporal correlation is used as the edge connection node of the graph, and the weight of each edge represents the temporal similarity between devices. For example, from ID i To ID j The edge with a weight of 0.952 starts from ID. i To ID k The edge with weight 0 (no connection) starts from ID. j To ID k If an edge has a weight of 0 (no edge connection), then the final adjacency matrix Adj is:

[0142]

[0143] This matrix not only represents the connection relationship between devices, but also contains similarity information of device behavior. The higher the similarity of each pair of devices, the greater the weight of the edge, indicating that their behavior is more similar.

[0144] This step also requires detailed explanation: the comprehensive feature representation of each device is a high-dimensional vector containing the device's behavioral features and its fusion information with the behaviors of neighboring devices. These features are learned from temporal similarity and behavioral data between the device and its neighbors through graph convolutional operations (GCN), where:

[0145] Node representation (device feature vector): The representation of each node (device) obtained through graph convolution is a vector containing the temporal behavior data of that device and its relationship with neighboring devices. For example, the feature representation of device i can be represented by a vector h.i express:

[0146] h i =[Behavioral characteristics, influence of temporal similarity of neighboring devices]

[0147] Integrating neighbor information: The feature representation of a device not only considers its own behavior, but also integrates the feature information of neighboring devices. This process is achieved through the weighted neighbor aggregation operation of each layer in the graph convolutional network.

[0148] Hierarchical information: After multi-layer graph convolution, the node representation will aggregate neighbor information from different levels. For example, the device feature representation of the k-th layer will contain information from the neighbors of the k-th layer, thus reflecting deeper temporal and behavioral features.

[0149] As an example, a comprehensive characteristic representation of a device can be obtained in the following ways:

[0150] Assuming the first layer of the graph convolutional network uses a simplified weighted summation aggregation method, and the initial values ​​of each device's features are only its own behavioral data (such as device status and usage frequency), the graph convolutional operation will adjust the representation of each device according to the features of its neighboring devices;

[0151] We define three devices i, j, and k, whose behavioral data is already included in the event association matrix, as shown in Table 3:

[0152] Table 3. Event Association Record Table

[0153]

[0154]

[0155] The initial feature matrix is ​​the initial behavioral feature vector for each device. Let the initial feature matrix be X, where each row is the feature vector of the device, and its mathematical expression is:

[0156]

[0157] The feature vector of device i is [1,0], which means that the state of device i is "device turned on" and the usage frequency is 1.

[0158] The feature vector of device j is [1,0], which means that the state of device j is "device enabled" and the usage frequency is 1.

[0159] The feature vector of device k is [0,1], which means that the state of device k is "device off" and the usage frequency is 1.

[0160] The graph convolution calculation is based on the adjacency matrix Adj and the initial feature matrix X, performing the first layer of graph convolution operation:

[0161]

[0162] in, W is the representation of device i after the first convolution, N(i) is the set of neighbor nodes of device i, and W is the representation of device i after the first convolution. (1) This is the weight matrix of the first layer, b (1) It is the bias term of the first level;

[0163] The weight matrix and bias terms are defined as follows:

[0164]

[0165] b (1) =[0,0]

[0166] The neighbor node of device i is j, so N(i) = {j}. The similarity between device i and device j is 0.952. According to the graph convolution formula, the first convolution result of device i is:

[0167]

[0168]

[0169] Therefore, the first layer of features of device i is represented as [0.622, 0.622];

[0170] It is easy to understand that the new feature representations of device j and device k are also calculated through the convolution operation shown in the figure above. In this embodiment, no further examples or explanations will be given. The feature representation of device j will be updated because of its connection with device i, and the feature representation of device k will be updated because of its connection with i and j.

[0171] The second layer of graph convolution further integrates information from neighboring devices into the device's feature representation. Through this second layer, the final feature representation of the device includes more fused features of temporal similarity between devices, enhancing the comprehensive behavioral features of each device and making the similarity between devices more obvious. After two layers of graph convolution, the comprehensive feature representations of devices i, j, and k are as follows:

[0172]

[0173] S4. Analyze user behavior feedback (such as device usage frequency) from the graph structure representation, generate instant rewards, update the Q-value based on the current device usage status and the generated instant rewards, and generate personalized behavior management methods for users through Q-value optimization strategies. Note that the following points should be noted in this step:

[0174] The Q-learning algorithm is used to analyze user behavior feedback from a graph structure representation. By calculating the state space, action space, and reward function, an immediate reward is generated.

[0175] State space is a set describing a user's current device usage behavior. In this embodiment of the invention, state space includes the current state of device usage, the user's operation history, and the device's working state. Its mathematical expression formula is:

[0176] S = {s1, s2, ..., s} n}

[0177] Where each state s i This includes equipment status, equipment usage frequency, and equipment historical behavior;

[0178] Device status: Whether the device is turned on (0 or 1), and the current mode of the device (e.g., energy-saving mode, normal mode, standby mode);

[0179] Equipment usage frequency: For example, the number of times the equipment has been used or the duration of continuous use;

[0180] Device history behavior: The user's previous behavior sequence, including whether device adjustments or optimizations were performed;

[0181] In this embodiment of the invention, the action space represents the operations that a user can perform on the device. Each action represents an adjustment by the user to the device's behavior, such as turning the device on or off, switching device modes, or adjusting configurations. Its mathematical expression is as follows:

[0182] A = {a1, a2, ..., a} m}

[0183] Among them, action a i This includes device on / off, device mode switching, and configuration adjustment;

[0184] Equipment On / Off: Controls whether the equipment is turned on or off;

[0185] Device mode switching: Switch to energy-saving mode, low-power mode or high-efficiency mode according to the current working status of the device;

[0186] Configuration adjustment: Adjust the device's operating configuration or parameters, such as temperature settings and brightness adjustment;

[0187] In this embodiment of the invention, the reward function is used to evaluate the effect of each action and provide feedback for subsequent optimization strategies. Specifically, the rules of the reward function are as follows:

[0188] Positive behaviors: Give higher rewards when user actions lead to more efficient device operation (e.g., switching to power-saving mode);

[0189] Negative behavior: When a user's actions cause the device to waste resources or fail to achieve the expected goal, a lower reward should be given;

[0190] Neutral behavior: When the operation does not significantly change the equipment efficiency, the reward is 0;

[0191] The mathematical expression for the reward function is:

[0192]

[0193] Suppose that each state-action pair (s, a) has an immediate reward value R(s, a), which reflects the benefit of the action;

[0194] It should be noted that in the Q-learning algorithm, the Q-value represents the cumulative reward obtained after performing a certain action in a certain state. The Q-value is updated using the following formula:

[0195]

[0196] Where Q(s,a) represents the Q-value of performing action a in state s, α is the learning rate, which controls the weight of updating old information with new information, R(s,a) is the immediate reward, representing the reward obtained after taking action a, and γ is the discount factor, representing the degree of discount on future rewards. In the new state s ′ Below, all possible actions a ′ The maximum value of Q is used to reflect the maximum expected value of future rewards;

[0197] Based on the Q-learning process, the system continuously updates the Q-value and selects the optimal action using the following strategies:

[0198] Greedy Strategy: Select the action with the highest Q value, i.e.:

[0199]

[0200] ε-greedy strategy: To balance exploration and exploitation, the system selects a random action with probability ∈ and selects the action with the largest Q value with probability 1-∈, that is:

[0201] Random action with probability∈,

[0202]

[0203] By continuously optimizing the Q value through the Q-learning algorithm, personalized behavior management strategies are generated. After training is completed, the system will automatically adjust the device management strategy according to the user's behavior patterns to ensure that the user's needs are effectively met.

[0204] Specifically, the generated behavior management strategies include:

[0205] Device On / Off: The device will automatically turn off when its energy consumption is too high or when the user stops using it, and will automatically turn on when the device needs to be used.

[0206] Device mode switching: If the device is currently operating in high-energy-consumption mode and the user does not perform frequent operations, the system will automatically switch to energy-saving mode;

[0207] Energy saving and efficiency optimization: The system will automatically adjust the usage time, operating mode and energy consumption of the equipment based on historical usage data in order to maximize the efficiency of resource utilization;

[0208] For example, if the Q value is related to a certain state s i The following action a j If the reward is high (such as an efficient energy-saving mode), the system will automatically select the action in similar subsequent states, continuously optimizing the user's device usage strategy;

[0209] In this embodiment of the invention, the device behavior features extracted by the graph convolutional network (GCN) are combined with the Q-learning algorithm to generate personalized device behavior management strategies. Through the Q-learning learning process, the system can automatically optimize the device's on / off control, mode switching, and configuration adjustment operations to improve energy efficiency.

[0210] For example, suppose there are two devices (device A and device B). The behavioral characteristics of these devices are collected and represented as a graph structure. The behavior of each device includes: the device's on / off state, usage frequency, operating mode, and the user's historical operations, as shown in Table 4.

[0211] Table 4. Equipment Behavior Table

[0212]

[0213] The temporal similarity between device A and device B is calculated using device behavior data. Assuming timestamps and device behavior data are used to measure the temporal similarity between the two devices, the behavioral similarity between device A and device B can be calculated using the following formula:

[0214] Adj A,B =f(Timestamp) A Timestamp B BehaviorA Behavior B )

[0215] The temporal similarity between device A and device B, calculated using the above formula, is 0.85 (indicating that their behavior and timestamps are similar).

[0216] In the graph structure, nodes represent devices, and devices A and B correspond to nodes in the graph. Edges represent the temporal similarity between devices, and the weight of an edge is 0.85, indicating that devices A and B have high similarity in behavior.

[0217] For example, the graph structure is represented as follows:

[0218] Node 1 (Device A): Includes device on / off status, usage frequency, current mode, and historical behavior characteristics;

[0219] Node 2 (Device B): Includes device on / off status, usage frequency, current mode, and historical behavior characteristics;

[0220] Edge (0.85): Similarity edge between device A and device B, indicating that the behavioral characteristics of the two devices are highly similar;

[0221] In the Q-learning algorithm, the state space (S) defines all possible states of the device, each state s i Includes the following features:

[0222] Equipment status: Whether the equipment is turned on (0 or 1), and the equipment's operating mode (e.g., energy-saving mode, normal mode);

[0223] Equipment usage frequency: How often the equipment is used, such as how many times it is used per day;

[0224] Equipment operating time: The duration of equipment operation in the current mode;

[0225] Device history behavior: The device's previous operation history (e.g., whether the user has adjusted the temperature or switched modes);

[0226] Similarity features in graph structures: Temporal similarity features between devices extracted through graph convolutional networks;

[0227] For example, define the state of device A (s1): device A is currently on, used 5 times / day, working in normal mode, and its historical behavior is that the temperature has been adjusted. Define the state of device B (s2): device B is currently off, used 2 times / day, working in energy-saving mode, and its historical behavior is that the mode has been switched. State space: S = {s1, s2}.

[0228] Action space (A) defines the operations that users can perform on the device. Action space includes turning the device on / off (controlling the device to turn on or off), mode switching (switching to energy-saving mode, low-power mode or high-efficiency mode according to the current working state of the device), and configuration adjustment (adjusting the temperature and brightness configuration of the device).

[0229] For example, if action a1 is defined as turning the device on / off, action a2 as switching the device mode (e.g., switching from normal mode to energy-saving mode), and action a3 as adjusting the device configuration (e.g., adjusting the temperature setting), then the action space is:

[0230] A = {a1: turn device on / off, a2: switch device mode, a3: adjust device configuration};

[0231] For example, the reward function R(s,a) is used to evaluate the effect of each action. When switching to energy-saving mode, the reward is +1; when switching to high-energy mode, the reward is -1; and when adjusting the temperature does not significantly change the energy efficiency, the reward is 0.

[0232] Define the Q-value update parameters as follows: learning rate α = 0.1, discount factor γ = 0.9, current state s1 (device A is currently on, in normal mode), current action a2 (switch device mode to power saving mode), instantaneous reward R(s1,a2) = +1, and maximum Q value. (Assuming device B currently does not have a significant Q value), the Q value update process is as follows:

[0233] Q(s1,a2)←Q(s1,a2)+0.1[1+0.9×0-0]=Q(s1,a2)+0.1

[0234] When Q(s1,a2)=0, then after the update:

[0235] Q(s1,a2)=0+0.1=0.1

[0236] Through the Q-learning update process described above, the system will select the optimal action based on the Q value. For example, in the subsequent state, the system will select a2 (switching to energy-saving mode) as the optimal action because Q(s1,a2) is the largest.

[0237] When the system selects "switch to power saving mode" as the optimal action, it means that the current state of the device (e.g., high power consumption mode, less operation or infrequent use) does not require high performance, but resource usage can be optimized by reducing power consumption;

[0238] For example, based on the optimization of the Q-learning algorithm described above, the following personalized management strategy is generated:

[0239] (1) Equipment switching control strategy

[0240] Threshold setting: If device A (or device B) is used less than twice a day, the system will automatically shut down the device;

[0241] For example, if device B is used twice a day, and the system determines that device B's energy consumption and current status are not worth continuing to keep it running, then the system can choose to turn off the device.

[0242] Control strategy: When the device is used ≤ 2 times / day, the system will automatically shut down the device. If the device is running in energy-saving mode and the usage time is less than the set threshold (e.g., 4 hours), then shut it down.

[0243] As an example, device A is used 5 times / day, and device B is used 2 times / day. According to the threshold setting, the device will be shut down when the usage frequency is ≤2 times / day. Therefore, for device A, the frequency is high (5 times / day), so the device will not be shut down. For device B, the frequency is low (2 times / day), so the system will automatically shut down the device.

[0244] (2) Device mode switching strategy

[0245] Threshold setting: If the device is currently in "normal mode" or "high energy consumption mode" and the device's usage time (e.g., more than 6 hours) has not reached a certain threshold, the system assumes that the device does not need to operate at high energy consumption and switches to energy-saving mode.

[0246] When the device is used infrequently and operates for extended periods (e.g., device A is used for ≥6 hours and is operated infrequently), the system will switch to energy-saving mode. In other words, if the device is in high-energy-consumption mode and the user is not performing frequent operations, the system will switch the device to energy-saving mode.

[0247] For example, if device A is used 5 times / day and works for 10 hours, the system may consider that the device is consuming too much energy and automatically switch to energy-saving mode. In state s1 (device A is used for 10 hours), after the Q value is updated, the system considers that switching to energy-saving mode can reduce energy consumption, so it will choose action a2 (switch to energy-saving mode).

[0248] (3) Equipment configuration adjustment strategy

[0249] Threshold setting: If the current configuration of the device does not meet the user's preferred value (such as temperature too high or too low), the system will automatically adjust the device configuration. For example, when the device is at a high temperature and the ambient temperature is low, the system will adjust the device temperature to decrease.

[0250] If the device does not adjust its configuration according to user preferences (e.g., the temperature is too high or the brightness is too low), the system will automatically adjust the device configuration.

[0251] For example, if the temperature of a device exceeds 22°C, the temperature will be automatically adjusted to the set value. If the current temperature of device A is 25°C and the user's preferred temperature is 22°C, the system will automatically adjust the configuration to achieve the user-set temperature value.

[0252] In an optional implementation, the device on / off control includes: automatically turning off the device if the device is used less than twice a day; and keeping the device on if the device is used more than twice a day.

[0253] In an optional implementation, the device mode switching includes: if the device is in a high-energy-consumption mode for a long time and is used infrequently, then switching to an energy-saving mode; if the device is used frequently and has high energy efficiency requirements, then maintaining the current mode (e.g., high-efficiency mode).

[0254] In an optional implementation, configuration adjustments include automatically adjusting the device configuration (such as temperature or brightness) if it does not meet user preferences.

[0255] Preferably, the processing in step 1 of the method of the present invention ensures the comprehensiveness and accuracy of data collection, laying the foundation for subsequent data processing. Combined with the time synchronization in step 2, it solves the problem of inconsistent timestamps of different devices and improves the accuracy of time series analysis. Through the implementation of step 3, the correlation between different devices is explored, overcoming the shortcomings of traditional methods in that they cannot effectively model the interaction relationship between devices. The operation in step 4 enables the management system to continuously optimize management strategies based on the actual behavior and feedback of users, and provide more intelligent and personalized services.

[0256] Preferably, the method of the present invention solves the problems of difficulty in integrating multi-source heterogeneous data, inaccurate time synchronization, insufficient user behavior modeling, and lack of personalized management strategies in the prior art. By introducing time synchronization, temporal correlation network algorithms and reinforcement learning strategies, it achieves efficient management and analysis of user usage information on multiple IoT devices, can provide users with personalized behavior management solutions, improves the real-time performance and efficiency of data processing, and enhances user experience.

[0257] The embodiments of the present invention further illustrate that traditional methods, which employ device information management based on simple statistics and rules to process and manage user data of IoT devices, have the following drawbacks:

[0258] Lack of time synchronization and correlation analysis: Due to the lack of time synchronization, it is impossible to accurately analyze the timing relationship between different devices, resulting in inaccurate understanding of device interactions and overall user behavior;

[0259] Low data processing efficiency: With the increase in the number of IoT devices and the amount of data, traditional statistical methods cannot efficiently process massive amounts of data and are difficult to meet real-time requirements;

[0260] Lack of personalization and adaptability: Device management strategies cannot be dynamically adjusted according to the user's actual usage and preferences, resulting in a lack of personalized services and a poor user experience;

[0261] Inefficient energy use: Inability to effectively optimize equipment operation, resulting in energy waste and a lack of intelligent management of energy consumption;

[0262] This experiment aims to verify the effectiveness of the proposed user information management method based on IoT big data in managing user information using IoT devices in a smart home environment. The experimental comparison indicators include:

[0263] Accuracy of User Behavior Pattern Recognition: Evaluate the accuracy of the method of the present invention in identifying and predicting user behavior on IoT devices, and verify its advantages in capturing complex user behavior patterns and inter-device interaction relationships.

[0264] Energy efficiency improvement effect: Verify the optimization effect of personalized behavior management strategy on energy consumption, and evaluate the role of the method of the present invention in reducing energy waste and improving equipment operating efficiency;

[0265] Data processing and analysis efficiency: Compare the efficiency differences between the method of this invention and traditional methods in processing and analyzing IoT big data, and verify its advantages in real-time performance and high efficiency;

[0266] The design experiment was conducted in a typical smart home environment, specifically including:

[0267] Living space: A three-bedroom, two-living-room apartment with an area of ​​approximately 120 square meters;

[0268] IoT devices: A total of 10 units, distributed in different rooms, including:

[0269] Device A: Smart Lighting (Device ID: Device001);

[0270] Device B: Intelligent thermostat (Device ID: Device002);

[0271] Device C: Smart TV (Device ID: Device003);

[0272] Device D: Smart Speaker (Device ID: Device004);

[0273] Device E: Smart Curtain (Device ID: Device005);

[0274] Device F: Smart Mattress (Device ID: Device006);

[0275] Device G: Smart Refrigerator (Device ID: Device007);

[0276] Device H: Smart Oven (Device ID: Device008);

[0277] Device I: Smart water heater (Device ID: Device009);

[0278] Device J: Smart door lock (Device ID: Device010);

[0279] Number of users: Three resident users, namely User1 (adult), User2 (adult) and User3 (child);

[0280] User characteristics:

[0281] User1: Works early and returns late on weekdays, prefers energy-saving mode, and likes to use a smart speaker to relax at night;

[0282] User2: Working from home, frequently using smart thermostats and lighting equipment, with high requirements for environmental comfort;

[0283] User3: After school, he enjoys watching smart TV and using a smart mattress at home, and goes to bed early.

[0284] Experiment duration: 28 consecutive days, covering weekdays and weekends;

[0285] Network environment: High-speed and stable Wi-Fi network to ensure reliable data transmission for IoT devices;

[0286] Time synchronization: The Precision Time Protocol (PTP) is used to synchronize the time of all devices, with a time accuracy of milliseconds.

[0287] Data storage and processing platform: Using a cloud server configured with an 8-core CPU, 16GB of memory and high-speed SSD storage, a comparative test of the algorithm of this invention and traditional methods is run;

[0288] Experimental operation of traditional methods: User operation data is collected independently from each IoT device, including device ID, user ID, operation time (local timestamp), device status and operation mode. The data of each device is independently statistically analyzed to calculate the number of daily uses and average usage time. Based on the historical usage data of the device, the moving average method is used to predict the future usage of the device. According to the prediction results and timed switches, corresponding management strategies are generated and applied to the automatic control of the device. The device operates according to preset rules.

[0289] Experimental operation of the method of this invention: Obtain device information used by the user from all IoT devices, including device ID, user ID, operation time (unified UTC timestamp), device status and operation mode. Use the PTP protocol to synchronize the time of all devices to ensure the time consistency of the event sequence. Calculate the temporal similarity between different devices, consider the degree of time proximity and the correlation of user operations, construct an association matrix, quantify the similarity between devices, form a graph structure representation, use devices as nodes and temporal similarity as edge weights, extract user behavior patterns from the graph structure, generate an immediate reward value r to evaluate the effect of the current policy, use the Q-learning algorithm in reinforcement learning to continuously update the Q-value table based on the immediate reward and the current state of the device, generate personalized device management policies, and realize intelligent control of devices.

[0290] The experimental comparison metrics include behavior pattern recognition accuracy, energy efficiency improvement rate, and data processing and analysis efficiency, among which:

[0291] In the experimental operation of the behavior pattern recognition accuracy index, traditional methods predict the user's next device operation based on the historical data of a single device. The method of this invention utilizes temporal correlation networks and Q-learning algorithms to comprehensively consider the correlation between devices and the user's behavior patterns to predict the user's next device operation. The calculation formula is as follows:

[0292] Behavioral pattern recognition accuracy =

[0293] (Number of correctly predicted user's next device operation / Total number of predictions) × 100%

[0294] In the experimental operation of the energy efficiency improvement rate index, the daily energy consumption of all equipment was recorded under two methods, the energy efficiency improvement rate was calculated, and the energy-saving effect of the method of the present invention was evaluated. The calculation formula is as follows:

[0295] Daily energy consumption = Total daily energy usage of all equipment (unit: kWh)

[0296] Energy efficiency improvement rate =

[0297] (Energy consumption of conventional methods - Energy consumption of the present invention) / Energy consumption of conventional methods × 100%

[0298] In the experimental operation of data processing and analysis efficiency indicators, the data processing time of the two methods under the same data volume was recorded to evaluate the efficiency of the method of the present invention in big data processing. The calculation formula is as follows:

[0299] Data processing time = The time required daily to process and analyze data from all devices (in minutes).

[0300] Data volume = Number of data entries collected daily

[0301] The experimental results were obtained as follows Figure 2 , Figure 3 , Figure 4 As shown.

[0302] Reference Figure 2 The dashed line (circular marker) represents the traditional method, with accuracy fluctuating between 58% and 62%. Traditional methods rely solely on historical data from a single device and cannot capture user behavior patterns across devices, resulting in low accuracy and slow improvement. The solid line (square marker) represents the method of this invention, with accuracy increasing from 80% to 96% and stabilizing after 15 days. The method of this invention utilizes temporal association networks and reinforcement learning algorithms to quickly learn user behavior habits, significantly improving accuracy and reaching a stable high level after 15 days.

[0303] Reference Figure 3 The dashed line (circular mark) represents the traditional method, with energy consumption between 150-126 kWh, showing a slow downward trend. Due to the lack of personalized management strategies, the energy consumption of the traditional method decreases slowly. The solid line (square mark) represents the method of this invention, with energy consumption decreasing from 140 kWh to 85 kWh, a greater reduction. The method of this invention effectively reduces unnecessary energy consumption through intelligent equipment management strategies, resulting in significant energy-saving effects.

[0304] Reference Figure 4 The dashed lines (circular markers) represent traditional methods, where processing time increases linearly with the amount of data, resulting in decreased efficiency. Traditional methods are inefficient when processing large amounts of data due to their reliance on simple statistical analysis. The solid lines (square markers) represent the method of this invention, where processing time increases slowly and exhibits better scalability. This invention utilizes optimized algorithms and data structures (such as association matrices and graph structures) to improve the efficiency of big data processing.

[0305] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A user information management method based on Internet of Things big data, characterized in that, include: The device information used by the user is obtained from multiple IoT devices, including user identity information and device usage behavior information. Each device is identified by a unique device identifier. The device information is timestamped and synchronized to form an event sequence containing device identifiers and timestamps; The local time of each device is synchronized to a unified standard time using the PTP protocol. Each device data record, after being updated with a timestamp, forms an event sequence containing a device identifier, a timestamp, and device behavior data. Each event record contains the device identifier, the timestamp after synchronization, and the device's behavior information. The behavioral data of the device includes at least the device's on / off status, usage frequency, and operation mode; The mathematical expression for the event sequence is as follows: E t ={(ID device ,timestamp t ,Data t )} Among them, E t Let ID represent the sequence of events at time t. device A unique identifier for the device, timestamp t The synchronized timestamp represents the standard time when the device action occurred. (Data) t This refers to the device's behavioral data, which includes device behavior information. The temporal correlation network algorithm is used to analyze the data interaction relationship between different IoT devices in the event sequence, calculate the temporal similarity between events, construct the correlation matrix, and form a graph structure representation. Based on timestamps and device behavior information between devices, the temporal similarity between two devices is calculated, and an event association matrix is ​​constructed. Each element of the event association matrix represents the similarity between two events. The higher the similarity between events, the stronger the correlation between the behaviors of the two devices. The formula for calculating the temporal similarity is: ΔT=|Timestamp i -Timestamp j | Where ΔT represents the time difference between device i and device j, and Adj i,j α is the element of the association matrix between device i and device j, α is a hyperparameter that adjusts the effect of time difference, and Behavior Similarity is the behavioral similarity, which is 1 if the behaviors are the same and 0 if the behaviors are different. The device data is fused and analyzed through graph convolution operations to construct a graph structure representation. Device identifiers are used as nodes in the graph, and temporal associations are used as edges to connect the nodes. The weight of each edge represents the temporal similarity between devices. The higher the similarity between a pair of devices, the greater the weight of the edge, indicating that the behavior is more similar. The user's behavioral feedback is analyzed from the graph structure representation to generate instant rewards. Based on the current state of device usage and the generated instant rewards, a Q-value optimization strategy is used to generate a personalized management method for the user. The Q-learning algorithm is used to analyze user behavior feedback from a graph structure representation. By calculating the state space, action space, and reward function, an immediate reward is generated. In Q-learning algorithms, the Q-value represents the cumulative reward obtained after performing an action in a given state. The Q-value is updated using the following formula: Where Q(s,a) represents the Q-value of performing action a in state s, α is the learning rate, which controls the weight of updating old information with new information, R(s,a) is the immediate reward, representing the reward obtained after taking action a, and γ is the discount factor, representing the degree of discount on future rewards. It is the maximum Q value of all possible actions a′ under the new state s′, used to reflect the maximum expected value of future rewards; Based on the Q-learning process, the system continuously updates the Q-value and selects the optimal action using an ε-greedy strategy: The system selects a random action with probability ∈ and the action with the largest Q value with probability 1-∈, that is: By continuously optimizing the Q-value through the Q-learning algorithm, personalized behavior management strategies are generated.

2. The user information management method based on IoT big data according to claim 1, characterized in that, The system binds user-used IoT devices to corresponding user IDs, collects device information through sensor technology and network communication, and records device status and user actions; among which: The user identity information includes username and user ID; The device usage behavior information includes the device's on / off status, usage frequency, and operation method; The device identifier is a unique identifier for each device, assigned by the device's hardware manufacturer or generated through a unique algorithm.

3. The user information management method based on IoT big data according to claim 2, characterized in that, The device on / off status records the device's on / off state at different points in time; the usage frequency records the number of times the device is used or the device's active duration within a certain period of time; and the operation method records the interaction method between the user and the device.

4. The user information management method based on IoT big data according to claim 1 or 2, characterized in that, The collected device information is used to form an initial dataset containing device identifiers, user identity information, and device behavior information, which is then stored in a structured manner. The mathematical expression of the initial dataset is as follows: D i =(ID device ,{Data behavior ,Data user }) Among them, D i For the data of the i-th device, ID device For device identifiers, Data behavior Data for device usage behavior. user For user identity data.

5. The user information management method based on IoT big data according to claim 1, characterized in that, The personalized management method includes controlling the on / off state of devices bound to the user, switching device modes, and adjusting configurations.

Images