Method and device for constructing network atlas of smart home system, and electronic equipment
The model and attention mechanism are constructed through graph neural network maps, which solves the problem that smart home systems are difficult to adapt to environmental changes, and achieves dynamic optimization of smart home systems and matching user expectations.
Patent Information
- Application Number
- CN202411051645.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-01
- Publication Date
- 2024-11-22
AI Technical Summary
Smart home systems are difficult to adapt to environmental changes, resulting in the failure of control strategies.
The graph neural network map is used to build a model, combine device information and user interaction data, adjust the association weight through the attention mechanism, and build a dynamic smart home network map to adapt to environmental changes and optimize device interaction.
It realizes the adaptability of smart home systems to environmental changes, improves the matching degree of device and user interaction, and ensures that the interaction results meet user expectations.
Smart Images

Figure CN119005304A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of smart home technology, for example, to a method and device, and electronic device for constructing a network map of a smart home system. Background Art
[0002] At present, in the smart home system, many living appliances and devices are connected and interacted through the Internet, forming a complex network system. Although smart homes have greatly facilitated our lives, due to the wide variety of device types and the varying degrees of interaction between various devices, how to effectively manage and optimize this large and complex system is a technical challenge.
[0003] In order to realize intelligent control of smart home, the related technology discloses a smart home control method based on AI model, including: predicting predicted corpus data corresponding to scene corpus data through a language model, and determining the smart home control strategy corresponding to the predicted corpus data in the smart home control strategy database, and controlling at least one smart home in the current smart home scene through the smart home control strategy.
[0004] In the process of implementing the embodiments of the present disclosure, it is found that there are at least the following problems in the related art:
[0005] The smart home control strategy for controlling the smart home is obtained from the smart home control strategy database, which is difficult to adapt to changes in the smart home environment.
[0006] It should be noted that the information disclosed in the above background technology section is only used to enhance the understanding of the background of the present application, and therefore may include information that does not constitute the prior art known to ordinary technicians in the field. Summary of the invention
[0007] In order to provide a basic understanding of some aspects of the disclosed embodiments, a brief summary is given below. The summary is not an extensive review, nor is it intended to identify key / critical components or delineate the scope of protection of these embodiments, but rather serves as a prelude to the detailed description that follows.
[0008] The embodiments of the present disclosure provide a method, an apparatus, and an electronic device for constructing a network map of a smart home system, so that the smart home system can adapt to changes in the smart home environment.
[0009] In some embodiments, the method for constructing a network graph of a smart home system includes: obtaining first system data of the smart home system; wherein the first system data includes at least device information data, environmental information data, and user interaction data; inputting the first system data of the smart home into a graph construction model to obtain a graph neural network graph output by the graph construction model; wherein the association relationship of the graph neural network graph is used to characterize the association relationship between devices and devices in the smart home environment, and / or, between devices and users; according to the second system data, using an attention mechanism to adjust the weight of the association relationship of the graph neural network graph to obtain a network graph of the smart home system.
[0010] Optionally, the first system data of the smart home is input into the graph construction model to obtain a graph neural network graph output by the graph construction model, including: establishing nodes in the graph construction model according to the device information data and user interaction data in the first system data; encoding the nodes to construct the association relationship of the graph neural network graph; wherein the association relationship represents the relationship of mutual influence between the nodes.
[0011] Optionally, based on the device information data and user interaction data in the first system data, nodes are established in the graph construction model, including: converting the device information data and user interaction data into continuous vectors; inputting the continuous vectors into the graph convolution layer of the graph construction model to update the properties of the initial nodes in the graph construction model using a convolutional neural network to obtain the nodes of the graph neural network graph.
[0012] Optionally, the device information data includes discrete device data and continuous device data; converting the device information data into a continuous vector includes: inputting the discrete device data and the continuous device data into an embedding layer of a graph construction model to convert the input discrete device data and the continuous device data into data in vector form to obtain a continuous vector in a multi-dimensional continuous space.
[0013] Optionally, based on the second system data, the attention mechanism is used to adjust the weight of the association relationship of the graph neural network map, including: when new devices are added and / or old devices are offline, the second system data of the smart home environment is obtained; wherein the second system data represents the attributes of the nodes of the graph neural network map and the interaction relationship between the nodes; according to the attributes of the nodes and the interaction relationship between the nodes, the attention score of the node is calculated; according to the attention score, the association relationship is assigned a corresponding weight; wherein the weight represents the degree of association between the nodes.
[0014] Optionally, obtaining first system data of the smart home environment includes: collecting entity data of the smart home environment; constructing a time series to divert or segment the entity data to obtain the first system data of the smart home environment.
[0015] Optionally, the method for constructing a network graph for a smart home system further includes: when a user interacts with the smart home system, controlling the smart home system to execute an interaction strategy using a graph neural network graph to obtain interaction feedback; wherein the interaction feedback is used to measure whether the interaction result meets user expectations; based on the interaction feedback, updating the attributes of the nodes of the graph neural network graph and the interaction relationships between the nodes so that the interaction results of the smart home system match the user expectations.
[0016] Optionally, based on the interaction feedback, the attributes of the nodes of the graph neural network map and the interaction relationships between the nodes are updated, including: calculating a reward value based on the interaction feedback; when the reward value is less than a threshold, updating the attributes of the nodes of the graph neural network map and the interaction relationships between the nodes; executing the interaction strategy and calculating a new reward value through the updated graph neural network map, and iteratively updating the attributes of the nodes of the graph neural network map and the interaction relationships between the nodes until the reward value is greater than or equal to the threshold.
[0017] In some embodiments, the device for constructing a network map of a smart home system includes: a processor and a memory storing program instructions, and the processor is configured to execute the method for constructing a network map of a smart home system as described above when running the program instructions.
[0018] In some embodiments, the electronic device includes: an electronic device body; and a device for constructing a network map of a smart home system as described above, installed on the electronic device body.
[0019] The method, device, and electronic device for constructing a network graph of a smart home system provided by the embodiments of the present disclosure can achieve the following technical effects:
[0020] The disclosed embodiment utilizes the graph construction model to construct a graph neural network graph of a smart home system. The first system data of the smart home system is used as input data during the construction process, so that the constructed graph neural network graph can reflect the connection between users and devices and between devices in the smart home system. In addition, the disclosed embodiment introduces an attention mechanism to adjust the weight of the association relationship of the graph neural network graph through the second system data, thereby adjusting the structure of the graph neural network graph, thereby making the obtained network graph of the smart home system more suitable for the current smart home environment.
[0021] The above general description and the following description are exemplary and explanatory only and are not intended to limit the present application. BRIEF DESCRIPTION OF THE DRAWINGS
[0022] One or more embodiments are exemplarily described by corresponding drawings, which do not limit the embodiments. Elements with the same reference numerals in the drawings are shown as similar elements, and the drawings do not constitute a scale limitation, and wherein:
[0023] Figure 1 It is a hardware environment diagram of a method for constructing a network map of a smart home system provided by an embodiment of the present disclosure;
[0024] Figure 2 is a schematic diagram of a method for constructing a network graph of a smart home system provided by an embodiment of the present disclosure;
[0025] Figure 3 It is a schematic diagram of a data input from an embedding layer to a graph convolution layer provided by an embodiment of the present disclosure;
[0026] Figure 4 is a schematic diagram of another method for constructing a network graph of a smart home system provided by an embodiment of the present disclosure;
[0027] Figure 5 It is an application schematic diagram of a method for constructing a network graph of a smart home system provided by an embodiment of the present disclosure;
[0028] Figure 6 is a schematic diagram of a device for constructing a network map of a smart home system provided by an embodiment of the present disclosure;
[0029] Figure 7 It is a schematic diagram of a network map of a smart home system provided by an embodiment of the present disclosure. DETAILED DESCRIPTION
[0030] In order to be able to understand the features and technical contents of the embodiments of the present disclosure in more detail, the implementation of the embodiments of the present disclosure is described in detail below in conjunction with the accompanying drawings. The attached drawings are for reference only and are not used to limit the embodiments of the present disclosure. In the following technical description, for the convenience of explanation, a full understanding of the disclosed embodiments is provided through multiple details. However, one or more embodiments can still be implemented without these details. In other cases, to simplify the drawings, well-known structures and devices can be simplified for display.
[0031] The terms "first", "second", etc. in the technical solutions described in the embodiments of the present disclosure are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the terms used in this way can be interchanged where appropriate, so as to describe the embodiments of the present disclosure described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions.
[0032] Unless otherwise stated, the term "plurality" means two or more.
[0033] In the embodiment of the present disclosure, the character " / " indicates that the preceding and following objects are in an "or" relationship. For example, A / B indicates: A or B.
[0034] The term "and / or" is a description of the association relationship between objects, indicating that three relationships can exist. For example, A and / or B means: A or B, or, A and B.
[0035] The term "correspondence" may refer to an association relationship or a binding relationship. The correspondence between A and B means that there is an association relationship or a binding relationship between A and B.
[0036] In the smart home system, smart devices include air conditioners, televisions, washing machines, lighting equipment, etc. They obtain environmental information through sensors and can operate according to user needs or preset programs. User needs and usage environments will also change over time and conditions. The technical problem that urgently needs to be solved is how to make the system adapt to the dynamic environment and make the best decision.
[0037] According to one aspect of an embodiment of the present application, a method for constructing a network map of a smart home system is provided. The method for constructing a network map of a smart home system is widely used in smart home (Smart Home), smart home, smart home device ecology, smart residential (Intelligence House) ecology and other whole-house intelligent digital control application scenarios. Optionally, in this embodiment, the above-mentioned method for constructing a network map of a smart home system can be applied to Figure 1 In the hardware environment composed of the terminal device 102 and the server 104 shown in FIG. Figure 1 As shown, the server 104 is connected to the terminal device 102 via a network, and can be used to provide services (such as application services, etc.) for the terminal or a client installed on the terminal. A database can be set on the server or independently of the server to provide data storage services for the server 104. Cloud computing and / or associated edge computing services can be configured on the server or independently of the server to provide data computing services for the server 104.
[0038] The network may include but is not limited to at least one of the following: wired network, wireless network. The wired network may include but is not limited to at least one of the following: wide area network, metropolitan area network, local area network, and the wireless network may include but is not limited to at least one of the following: WIFI (Wireless Fidelity), Bluetooth. The terminal device 102 may be but is not limited to a PC, a mobile phone, a tablet computer, a smart air conditioner, a smart range hood, a smart refrigerator, a smart oven, a smart stove, a smart washing machine, a smart water heater, a smart washing device, a smart dishwasher, a smart projection device, a smart TV, a smart clothes drying rack, a smart curtain, a smart audio and video, a smart socket, a smart speaker, a smart fresh air device, a smart kitchen and bathroom device, a smart bathroom device, a smart sweeping robot, a smart window cleaning robot, a smart mopping robot, a smart air purification device, a smart steamer, a smart microwave oven, a smart kitchen treasure, a smart purifier, a smart water dispenser, a smart door lock, etc.
[0039] It should be understood that Figure 1 The number of terminal devices and servers in the figure is only for illustration. According to actual needs, there may be any number of terminal devices and servers. For example, one server may correspond to multiple terminal devices.
[0040] Combination Figure 2 As shown, the embodiment of the present disclosure provides a method for constructing a network graph of a smart home system, including:
[0041] S201, the server obtains first system data of the smart home system; wherein the first system data at least includes device information data, environment information data and user interaction data.
[0042] S202, the server inputs the first system data of the smart home into the graph construction model, and obtains the graph neural network graph output by the graph construction model; wherein the association relationship of the graph neural network graph is used to characterize the association relationship between devices and devices in the smart home environment, and / or, between devices and users.
[0043] S203, the server uses the attention mechanism to adjust the weight of the association relationship of the graph neural network map according to the second system data to obtain the network map of the smart home system.
[0044] The method for constructing a network graph of a smart home system provided by the embodiment of the present disclosure is adopted, and a graph construction model is used to construct a graph neural network graph of the smart home system. In the construction process, the first system data of the smart home system is used as input data, so that the constructed graph neural network graph can reflect the connection between users and devices and between devices in the smart home system. In addition, the attention mechanism is introduced in the embodiment of the present disclosure, and the weight of the association relationship of the graph neural network graph is adjusted through the second system data, so that the structure of the graph neural network graph can be adjusted, thereby making the obtained network graph of the smart home system more suitable for the current smart home environment.
[0045] Environmental information data plays a decisive role in optimizing equipment operation. During the data collection stage, basic information such as room temperature and humidity needs to be obtained. In addition, the number of people in the room and the location of people change over time and will directly affect the working status of the equipment (for example, the air conditioner is automatically adjusted to standby mode when no one is in the room). Therefore, environmental information data also includes the number of people in the room and the location of people.
[0046] In actual applications, a smart home environment may include user A, user B, a Bluetooth speaker, a refrigerator, and a mobile phone. Figure 7 In the network graph shown, the graph construction model establishes nodes corresponding to user A, user B, Bluetooth speakers, refrigerators, and mobile phones in the smart home environment. These nodes will affect each other during the interaction process. According to the impact of the interaction, the graph construction model creates an association relationship between nodes. For example, user A can turn on the Bluetooth speaker through voice interaction control commands, and the mobile phone can control the refrigeration temperature of the refrigerator through wireless network control commands.
[0047] Optionally, the first system data of the smart home is input into the graph construction model to obtain a graph neural network graph output by the graph construction model, including: establishing nodes in the graph construction model according to the device information data and user interaction data in the first system data; encoding the nodes to construct the association relationship of the graph neural network graph; wherein the association relationship represents the relationship of mutual influence between the nodes.
[0048] In the disclosed embodiment, to construct the graph neural network graph, it is first necessary to create a corresponding unique node for each device in the smart home environment and the user who has interacted with the device. Each node contains device information data and user interaction data.
[0049] Device information data is the most critical part of the smart home environment. Each device (such as smart lighting, air conditioning, TV, etc.) has its own unique operating parameters (such as on / off status, current and voltage, working mode, etc.), which can be obtained through the device's built-in sensors. In addition, there may be interactions between devices (such as multiple speakers may form a stereo effect), and the interaction parameters are also important parameters that need to be examined and recorded.
[0050] User interaction data includes the commands issued by users to the device through mobile applications or voice assistants and the execution results obtained. The data generated during the execution of commands is a very important source of feedback information. By analyzing and learning the user's commands, the system can predict user needs more accurately.
[0051] In the process of encoding nodes, nodes represent the location of devices or users in the vector space, and associations represent the existence of certain relationships between nodes. For example, two adjacent smart lighting lamps that adjust their working status in an interrelated manner can be regarded as two nodes, and because of their mutual influence, an "association relationship" is created in the entire graph.
[0052] Optionally, based on the device information data and user interaction data in the first system data, nodes are established in the graph construction model, including: converting the device information data and user interaction data into continuous vectors; inputting the continuous vectors into the graph convolution layer of the graph construction model to update the properties of the initial nodes in the graph construction model using a convolutional neural network to obtain the nodes of the graph neural network graph.
[0053] In the disclosed embodiment, the device information data and user interaction data need to be converted into a form that is easy for the neural network to learn and understand. The embedding method can be used to convert discrete variables into continuous variables. For example, the device type represented by the discrete identifier is converted into a vector in a multi-dimensional continuous space through embedding. Similarly, different user interaction data can also be converted into corresponding continuous vectors.
[0054] After all the data is collected and embedded, a convolutional neural network is used in the graph convolution layer to calculate and update each node. Specifically, during the convolution process, each node updates its node attributes based on its own information and that of its neighboring nodes, so that information can be efficiently propagated in the graph network and the status of each node can be updated.
[0055] Optionally, the device information data includes discrete device data and continuous device data; converting the device information data into a continuous vector includes: inputting the discrete device data and the continuous device data into an embedding layer of a graph construction model to convert the input discrete device data and the continuous device data into data in vector form to obtain a continuous vector in a multi-dimensional continuous space.
[0056] In the disclosed embodiment, discrete device data includes device type and status, and continuous device data includes specific operating parameters (such as current, voltage, etc.). Discrete device data and continuous device data are converted into multidimensional vectors through word embedding technology. Word embedding is a language model that can map discrete words or objects (device type, status, etc.) to continuous vector space. Word embedding can greatly reduce data dimensionality and break data sparsity.
[0057] After word embedding, a set of "standard" vectorized device data is obtained. Each device type and its operating status are represented by a unique vector and saved as nodes in the constructed graph model. The distance between vectors represents the degree of content similarity.
[0058] Combination Figure 3 As shown in the figure, the embedding layer encodes the discrete device data and the continuous device data into a vectorized form and inputs them into the next graph convolution layer for further processing.
[0059] Optionally, based on the second system data, the attention mechanism is used to adjust the weight of the association relationship of the graph neural network map, including: when new devices are added and / or old devices are offline, the second system data of the smart home environment is obtained; wherein the second system data represents the attributes of the nodes of the graph neural network map and the interaction relationship between the nodes; according to the attributes of the nodes and the interaction relationship between the nodes, the attention score of the node is calculated; according to the attention score, the association relationship is assigned a corresponding weight; wherein the weight represents the degree of association between the nodes.
[0060] In the disclosed embodiment, after receiving specific events such as new devices joining or old devices going offline, the graph construction model will be dynamically updated according to the feedback information. The attention mechanism can solve the problems of large number of devices and complex interaction relationships in the smart home environment. The attention mechanism can assign different weights to the nodes of each smart home device so that the graph construction model can pay more attention to the device nodes that have an important impact on the user.
[0061] In the graph neural network, each node (smart home device or user) is assigned an attention score. The attention score is calculated based on the properties of the node (such as device type, status, user behavior) and the interaction relationship between nodes (such as connection strength, interaction frequency). Combined with the attention score, the weight of each association in the graph can be adjusted to reflect the strength of the association between different devices and the degree of their impact on users.
[0062] When new devices are added and old devices are offline, various interactions in the smart home environment will change. The system can obtain the second system data of the smart home environment when new devices are added and / or old devices are offline. At this time, the graph construction model can be dynamically adjusted according to the second system data to adapt to the current environment structure. For example, it is necessary to recalculate the attention score corresponding to each node in the new environment, and adjust the weight of each associated relationship connected to it according to the score.
[0063] Optionally, obtaining first system data of the smart home environment includes: collecting entity data of the smart home environment; constructing a time series to divert or segment the entity data to obtain the first system data of the smart home environment.
[0064] In the disclosed embodiment, the collected entity data includes device information data, environmental information data and user interaction data. Entity data is usually complex and not suitable for direct application in the model, so the entity data is preprocessed. First, data cleaning can eliminate invalid, erroneous or incomplete data. Secondly, numerical processing operations can perform numerical operations suitable for neural network learning according to model requirements, including but not limited to standardization / normalization / bucketing, etc.
[0065] Furthermore, changes in user needs and the passage of time will cause changes in the smart home environment. In order for the system to improve future interactions, it needs to learn from past behaviors (similar to the human experience learning process). Therefore, it is necessary to construct a time series for data diversion or window segmentation so that the graph construction model has memory.
[0066] Optionally, the method for constructing a network graph for a smart home system further includes: when a user interacts with the smart home system, controlling the smart home system to execute an interaction strategy using a graph neural network graph to obtain interaction feedback; wherein the interaction feedback is used to measure whether the interaction result meets user expectations; based on the interaction feedback, updating the attributes of the nodes of the graph neural network graph and the interaction relationships between the nodes so that the interaction results of the smart home system match the user expectations.
[0067] Combination Figure 4 As shown, the embodiment of the present disclosure provides another method for constructing a network graph of a smart home system, including:
[0068] S401, the server obtains first system data of the smart home system; wherein the first system data at least includes device information data, environment information data and user interaction data.
[0069] S402, the server inputs the first system data of the smart home into the graph construction model, and obtains the graph neural network graph output by the graph construction model; wherein the association relationship of the graph neural network graph is used to characterize the association relationship between devices and devices in the smart home environment, and / or, between devices and users.
[0070] S403, the server uses the attention mechanism to adjust the weight of the association relationship of the graph neural network map according to the second system data to obtain the network map of the smart home system.
[0071] S404, when the user interacts with the smart home system, the server controls the smart home system to use the graph neural network graph to execute the interaction strategy and obtain interaction feedback; wherein the interaction feedback is used to measure whether the interaction result meets the user's expectations.
[0072] S405, the server updates the attributes of the nodes of the graph neural network graph and the interaction relationships between the nodes based on the interaction feedback, so that the interaction results of the smart home system match the user's expectations.
[0073] In the disclosed embodiment, the user interacts with the smart home system multiple times, and the system obtains feedback information from the interaction, and continuously adjusts the system's interaction strategy based on the feedback information, so that the interaction results of the smart home system match the user's expectations.
[0074] Optionally, based on the interaction feedback, the attributes of the nodes of the graph neural network map and the interaction relationships between the nodes are updated, including: calculating a reward value based on the interaction feedback; when the reward value is less than a threshold, updating the attributes of the nodes of the graph neural network map and the interaction relationships between the nodes; executing the interaction strategy and calculating a new reward value through the updated graph neural network map, and iteratively updating the attributes of the nodes of the graph neural network map and the interaction relationships between the nodes until the reward value is greater than or equal to the threshold.
[0075] In the disclosed embodiment, the entire smart home environment and all devices can be regarded as an explicit environment. Each user command or device state change (such as temperature adjustment, lighting adjustment, etc.) is regarded as an interaction with the environment. After each interaction, the system will obtain a corresponding interaction feedback, which is used to measure whether the interaction meets the user's expectations. The reward value is calculated based on the interaction feedback, and the effect of the currently adopted interaction strategy can be judged and fine-tuned. After enough steps of interaction and enough interaction feedback, the system will learn an optimal interaction strategy to maximize the reward value.
[0076] like Figure 5As shown, in actual applications, the data acquisition module of the system can be responsible for collecting the first system data in the smart home environment, and these first system data will be used as the input of the graph construction model. After the graph construction model receives the input data, it is encoded through the embedding layer. The embedding layer encodes the data into a vectorized form and inputs it to the graph convolution layer, which calculates and updates each node. In the process of the graph convolution layer calculating and updating each node, the attention mechanism is introduced and used to assign different weights to each device node, so that the graph neural network model can pay more attention to the device nodes that have an important impact on the user. Furthermore, the reinforcement learning mechanism is introduced and used to optimize the interaction strategy of the smart home system.
[0077] Combination Figure 6 As shown, the embodiment of the present disclosure provides a device 60 for constructing a network map of a smart home system, including a processor (processor) 600 and a memory (memory) 601. Optionally, the device 60 may also include a communication interface (Communication Interface) 602 and a bus 603. Among them, the processor 600, the communication interface 602, and the memory 601 can communicate with each other through the bus 603. The communication interface 702 can be used for information transmission. The processor 600 can call the logic instructions in the memory 601 to execute the method for constructing a network map of a smart home system in the above embodiment.
[0078] In addition, the logic instructions in the memory 601 described above can be implemented in the form of software functional units and can be stored in a computer-readable storage medium when sold or used as an independent product.
[0079] The memory 601 is a computer-readable storage medium that can be used to store software programs and computer executable programs, such as program instructions / modules corresponding to the method in the embodiment of the present disclosure. The processor 600 executes the functional application and data processing by running the program instructions / modules stored in the memory 601, that is, the method for constructing a network map of a smart home system in the above embodiment is implemented.
[0080] The memory 601 may include a program storage area and a data storage area, wherein the program storage area may store an operating system and an application required for at least one function; the data storage area may store data created according to the use of the terminal device, etc. In addition, the memory 601 may include a high-speed random access memory and may also include a non-volatile memory.
[0081] The disclosed embodiment provides an electronic device, including: an electronic device body, and the above-mentioned device 60 for constructing a network map of a smart home system. The device 60 for constructing a network map of a smart home system is installed on the electronic device body. The installation relationship described here is not limited to placement inside the electronic device body, but also includes installation connections with other components of the electronic device, including but not limited to physical connections, electrical connections, or signal transmission connections. It can be understood by those skilled in the art that the device 60 for constructing a network map of a smart home system can be adapted to a feasible electronic device body, thereby realizing other feasible embodiments.
[0082] The technical solution of the embodiment of the present disclosure can be embodied in the form of a software product, which is stored in a storage medium and includes one or more instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the method described in the embodiment of the present disclosure. The aforementioned storage medium may be a non-transient storage medium, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a disk or an optical disk, and other media that can store program codes.
[0083] The above description and the accompanying drawings fully illustrate the embodiments of the present disclosure so that those skilled in the art can practice them. Other embodiments may include structural, logical, electrical, process and other changes. The embodiments represent only possible changes. Unless explicitly required, separate components and functions are optional, and the order of operation may vary. The parts and features of some embodiments may be included in or replace the parts and features of other embodiments. Moreover, the words used in this application are only used to describe the embodiments and are not used to limit the claims. As used in the description of the embodiments and the claims, unless the context clearly indicates, the singular forms of "a", "an" and "the" are intended to include plural forms as well. Similarly, the term "and / or" as used in this application refers to any and all possible combinations of listings containing one or more associated ones. In addition, when used in the present application, the term "comprise" and its variants "comprises" and / or comprising refer to the presence of stated features, wholes, steps, operations, elements, and / or components, but do not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components and / or groups thereof. In the absence of further restrictions, the elements defined by the sentence "comprising a ..." do not exclude the presence of other identical elements in the process, method or device comprising the elements. In this article, each embodiment may focus on the differences from other embodiments, and the same and similar parts between the various embodiments may refer to each other. For the methods, products, etc. disclosed in the embodiments, if they correspond to the method part disclosed in the embodiments, then the relevant parts can refer to the description of the method part.
[0084] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software may depend on the specific application and design constraints of the technical solution. The technicians may use different methods for each specific application to implement the described functions, but such implementations should not be considered to exceed the scope of the embodiments of the present disclosure. The technicians may clearly understand that, for the convenience and simplicity of description, the specific working processes of the systems, devices and units described above may refer to the corresponding processes in the aforementioned method embodiments, and will not be repeated here.
[0085] In the embodiments disclosed herein, the disclosed methods and products (including but not limited to devices, equipment, etc.) can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units can be only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the coupling or direct coupling or communication connection between each other shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms. The units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the units may be selected according to actual needs to implement this embodiment. In addition, each functional unit in the embodiment of the present disclosure may be integrated in a processing unit, or each unit may exist physically alone, or two or more units may be integrated in one unit.
[0086] The flowchart and block diagram in the accompanying drawings show the possible architecture, function and operation of the system, method and computer program product according to the embodiment of the present disclosure. In this regard, each box in the flowchart or block diagram can represent a module, a program segment or a part of the code, and the module, the program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. In some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two consecutive boxes can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, which can depend on the functions involved. In the description corresponding to the flowchart and the block diagram in the accompanying drawings, the operations or steps corresponding to different boxes can also occur in a different order from the order disclosed in the description, and sometimes there is no specific order between different operations or steps. For example, two consecutive operations or steps can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, which can depend on the functions involved. Each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, may be implemented by a dedicated hardware-based system that performs the specified functions or actions, or may be implemented by a combination of dedicated hardware and computer instructions.
Claims
1. A method for constructing a network graph of a smart home system, characterized in that: include: Acquire first system data of the smart home system; wherein the first system data at least includes device information data, environment information data and user interaction data; Inputting the first system data of the smart home into the graph construction model to obtain a graph neural network graph output by the graph construction model; wherein the association relationship of the graph neural network graph is used to characterize the association relationship between devices and devices in the smart home environment, and / or, between devices and users; According to the second system data, the attention mechanism is used to adjust the weights of the association relationships of the graph neural network map to obtain the network map of the smart home system.
2. The method for constructing a network graph of a smart home system according to claim 1, characterized in that: Input the first system data of the smart home into the graph construction model to obtain the graph neural network graph output by the graph construction model, including: Establishing nodes in the graph construction model according to the device information data and the user interaction data in the first system data; Encode the nodes and construct the association relationship of the graph neural network graph; the association relationship represents the mutual influence relationship between the nodes.
3. The method for constructing a network graph of a smart home system according to claim 2, characterized in that: According to the device information data and the user interaction data in the first system data, a node is established in the graph construction model, including: Convert device information data and user interaction data into continuous vectors; The continuous vector is input into the graph convolution layer of the graph construction model to update the properties of the initial nodes in the graph construction model using the convolutional neural network to obtain the nodes of the graph neural network graph.
4. The method for constructing a network graph of a smart home system according to claim 3, characterized in that: The device information data includes discrete device data and continuous device data; converting the device information data into a continuous vector includes: The discrete device data and the continuous device data are input into the embedding layer of the graph construction model to convert the input discrete device data and the continuous device data into data in vector form, thereby obtaining a continuous vector in a multi-dimensional continuous space.
5. The method for constructing a network graph of a smart home system according to claim 1, characterized in that: According to the data from the second system, the attention mechanism is used to adjust the weights of the association relationships of the graph neural network graph, including: When a new device is added and / or an old device is offline, second system data of the smart home environment is obtained; wherein the second system data represents the attributes of the nodes of the graph neural network graph and the interaction relationship between the nodes; Calculate the attention score of the node based on the attributes of the node and the interaction relationship between the nodes; The corresponding weight is assigned to the association relationship according to the attention score, wherein the weight represents the degree of association between the nodes.
6. The method for constructing a network graph of a smart home system according to claim 1, characterized in that: Get first-class system data for smart home environment, including: Collect physical data of smart home environment; Construct a time series to divert or segment the entity data and obtain the first system data of the smart home environment.
7. The method for constructing a network graph of a smart home system according to any one of claims 1 to 6, characterized in that: Also includes: When the user interacts with the smart home system, the smart home system is controlled to execute the interaction strategy using the graph neural network graph to obtain interaction feedback; the interaction feedback is used to measure whether the interaction result meets the user's expectations; Based on the interaction feedback, the attributes of the nodes in the graph neural network graph and the interaction relationships between the nodes are updated to make the interaction results of the smart home system match the user's expectations.
8. The method for constructing a network graph of a smart home system according to claim 7, characterized in that: According to the interactive feedback, update the properties of the nodes in the graph neural network graph and the interactive relationships between the nodes, including: Calculate reward value based on interaction feedback; When the reward value is less than the threshold, update the attributes of the nodes in the graph neural network graph and the interaction relationship between the nodes; The interaction strategy is executed and the new reward value is calculated through the updated graph neural network graph, and the attributes of the nodes of the graph neural network graph and the interaction relationship between the nodes are iteratively updated until the reward value is greater than or equal to the threshold.
9. A device for constructing a network map of a smart home system, comprising a processor and a memory storing program instructions, characterized in that: The processor is configured to execute the method for constructing a network map of a smart home system as described in any one of claims 1 to 8 when running the program instructions.
10. An electronic device, characterized in that: include: Electronic device body; The device for constructing a network map of a smart home system as described in claim 9 is installed on the electronic device body.
Citation Information
Patent Citations
User behavior prediction model training method and device for intelligent home control
CN107665230A
Knowledge graph reasoning relation prediction method based on graph neural network
CN113010691A
Method and device for man-machine interaction of intelligent equipment and intelligent equipment
CN117540016A
Method and device for man-machine interaction of intelligent equipment and intelligent equipment
CN117540017A
Service recommendation method and system based on knowledge graph and graph neural network
CN117688238A
