Service recommendation method, device, equipment, medium and computer program product

Abstract

The embodiments of the present invention disclose a service recommendation method, device, equipment, medium and computer program product, wherein the service recommendation method includes: obtaining historical behavior data of a target user in a target space; identifying the service needs corresponding to different spatial regions of the target space based on the historical behavior data and the current spatial state and event information determined by a spatial object model; the spatial object model is constructed based on the regional attribute information and device information in the target space; based on the service functions that the spatial object model can provide, generating and pushing service functions adapted to the service needs. By combining the spatial object model and user data, the present application can more accurately identify user needs, and analyze the user's historical behavior and current spatial state, thereby providing users with personalized service recommendations that meet user needs, effectively improving the accuracy and relevance of service recommendations, and thus improving user experience.

Description

Technical Field

[0001] The present invention relates to the field of computer processing technology, and in particular to a service recommendation method, device, electronic device, storage medium and computer program product. Background Art

[0002] With the rapid development of Internet of Things technology and the continuous advancement of artificial intelligence, intelligent service recommendation systems have gradually become a key component of modern life and business operations. Intelligent service recommendation systems collect data through smart devices and sensors, and use advanced data analysis technology to provide users with personalized service recommendations, thereby improving efficiency, optimizing resource allocation and enhancing user experience. For example, the smart home system, based on the Internet of Things technology, uses various smart devices and sensors to achieve intelligent management and services for the home environment. It has become a new lifestyle that can achieve home automation, remote control, energy optimization, security monitoring and personalized living experience.

[0003] However, traditional intelligent product and service recommendations are often only based on users' personal preferences or behavioral data, and fail to fully consider the characteristics of the space and the relationships between users, resulting in low accuracy and personalization of service recommendations.

[0004] Application Contents

[0005] In view of this, the present application provides a service recommendation method, device, electronic device, storage medium and computer program product, which are used to solve the problem of low accuracy and personalization of service recommendations in the prior art. In order to achieve one or part or all of the above purposes or other purposes, the present application proposes a service recommendation method, device, electronic device, storage medium and computer program product. In the first aspect:

[0006] A service recommendation method, comprising:

[0007] Obtain historical behavior data of target users in the target space;

[0008] Based on the historical behavior data and the current space state and event information determined by the space object model, identifying the service needs of the target user corresponding to different space areas under the target space; the space object model is constructed based on the area attribute information and device information under the target space;

[0009] Based on the service functions that the spatial object model can provide, service functions adapted to the service requirements are generated and pushed.

[0010] A second aspect: A service recommendation device, comprising:

[0011] The acquisition module is used to obtain the historical behavior data of the target user in the target space;

[0012] an identification module, for identifying the service needs of target users corresponding to different spatial areas in a target space based on the historical behavior data and the current spatial state and event information determined by a spatial object model; the spatial object model is constructed based on the regional attribute information and device information in the target space;

[0013] The recommendation module is used to generate and push service functions adapted to the service requirements based on the service functions that the space object model can provide.

[0014] In one embodiment, the step of constructing the spatial object model includes:

[0015] Obtain the hierarchical relationship of each area in the target space, as well as the device information of the smart devices bound to each area;

[0016] Determine the functions that each smart device can provide based on the device information of each smart device;

[0017] Based on the hierarchical relationship between each area and the functions that each smart device can provide, a spatial object model corresponding to the target space is constructed.

[0018] In one embodiment, the identification module includes:

[0019] An extraction unit, configured to extract features from the user behavior data and the device status data based on a spatial object model to obtain corresponding user behavior features and device status features;

[0020] A determination unit, configured to determine a current space state and event information according to the user behavior characteristics and the device state characteristics;

[0021] The identification unit is used to identify the service requirements corresponding to different spatial areas of the target user in the target space according to the current spatial state and the event information.

[0022] In one embodiment, the identification unit comprises:

[0023] A mining subunit, used to perform data mining processing on the current space state, the device state data and the user behavior data respectively to obtain corresponding data processing results;

[0024] The identification subunit is used to identify the service requirements corresponding to different spatial areas of the target user in the target space based on the data processing result.

[0025] In one embodiment, the data processing results include data clustering results; the mining subunit is used to: perform cluster analysis on the current spatial state, the device state data and the user behavior data respectively to obtain corresponding data clustering results; based on the data clustering results, determine at least one of the regional characteristics of each spatial area under the target space, the user behavior pattern under each spatial area and the device association results between each device under each spatial area; based on the device association results, the regional characteristics and at least one of the user behavior patterns, predict the service needs of the target user corresponding to different spatial areas under the target space.

[0026] In one embodiment, the recommendation module includes:

[0027] A service determination unit, configured to determine the service functions that the space object model can provide according to the device attribute information and device capability information of each smart device in the target space;

[0028] The service push unit is used to obtain and push service functions that are adapted to the service requirements from the service functions that can be provided by the space object model.

[0029] In one embodiment, the recommendation module further includes:

[0030] A calculation unit, used to calculate the matching degree between the service demand and each service function, wherein the matching degree represents the adaptability between the service demand and each service function;

[0031] A first recommendation unit, configured to activate a target device in a target space area based on the service function if the matching degree reaches a first threshold range, so as to instruct the target device to provide a recommended service corresponding to the service function;

[0032] The second recommendation unit is used to push the service function to the target user when the recommendation degree reaches a second threshold range, so as to remind the target user whether to apply the recommended service corresponding to the service function.

[0033] In one embodiment, the identification module is also used to: obtain configured preference configuration data; identify the service needs of the target user corresponding to different spatial areas under the target space based on the historical behavior data and the preference configuration data, as well as the current spatial state and event information determined by the spatial object model.

[0034] In one embodiment, the service function is obtained by processing a machine learning model, and the training steps of the machine learning model include:

[0035] Acquire sample data and corresponding sample labels; the sample data includes sample historical behavior data, sample space status, and sample event information; the sample labels include labels corresponding to service requirements;

[0036] Extracting service demand characteristics corresponding to the sample data;

[0037] Matching the sample data with corresponding estimated service demands based on the service demand characteristics through an initial machine learning model;

[0038] Based on the difference loss value between the estimated service demand and the sample label, the model parameters of the initial machine learning model are adjusted, and the initial machine learning model is continued to be trained based on the adjusted model parameters until the training conditions are met, thereby obtaining a trained machine learning model.

[0039] A third aspect: An electronic device includes a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of the service recommendation method of each embodiment of the present application when executing the computer program.

[0040] A fourth aspect: A storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the steps of the service recommendation method of each embodiment of the present application.

[0041] A fifth aspect: A computer program product, comprising a computer program, characterized in that when the computer program is executed by a processor, the steps of the service recommendation method of each embodiment of the present application are implemented.

[0042] Implementing the embodiments of the present application will have the following beneficial effects:

[0043] First, the historical behavior data of the target user in the target space is obtained; then, based on the historical behavior data and the current spatial state and event information determined by the spatial object model, the service needs corresponding to the target user in different spatial areas in the target space are identified; wherein, the spatial object model is constructed based on the regional attribute information and device information in the target space; finally, based on the service functions that the spatial object model can provide, service functions adapted to the service needs are generated and pushed. This application accurately identifies the services that the user may need in the current space by combining the user's historical behavior and current spatial state, and then generates the service functions that best match the service needs through the service functions of the spatial object model and pushes them to the user, thereby providing the user with personalized service recommendations that meet the user's needs, effectively improving the accuracy and relevance of service recommendations, and thus improving the user experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0045] in: Figure 1 A schematic diagram of an application environment of a service recommendation method provided in an embodiment;

[0046] Figure 2 A flowchart of a service recommendation method provided in one embodiment;

[0047] Figure 3 A schematic diagram of the process of step S2 provided in one embodiment;

[0048] Figure 4 A schematic diagram of the process of step S3 provided in one embodiment;

[0049] Figure 5 A schematic diagram of the structure of a service recommendation device provided in one embodiment;

[0050] Figure 6 The figure is a diagram of the internal structure of a computer device provided in one embodiment. DETAILED DESCRIPTION

[0051] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0052] In the following description, reference is made to “some embodiments”, which describe a subset of all possible embodiments, but it will be understood that “some embodiments” may be the same subset or different subsets of all possible embodiments and may be combined with each other without conflict.

[0053] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art to which the present invention belongs. The terms used herein are only for the purpose of describing the embodiments of the present invention and are not intended to limit the present invention.

[0054] The service recommendation method provided by this application can be applied to Figure 1 In the application environment shown in the figure, Figure 1A service recommendation system is provided. The service recommendation system includes a smart device 100, a gateway 200 communicatively connected to the smart device 100, a router 300, a server 400, and a terminal 500.

[0055] The smart device 100 accesses the gateway 200 in the service recommendation system, and communicates with the gateway 200 through its own configured communication module, and is then controlled by the gateway 200. In one embodiment, the smart device 100 accesses the gateway 200 through a local area network path or a wide area network path, and is thus deployed in the gateway 200. The local area network may include ZIGBEE or Bluetooth, and the wide area network may include 2G / 3G / 4G / 5G / WIFI, etc.

[0056] The gateway 200 establishes a network connection with the terminal 500 or the server 400 through the router 300. In one embodiment, the gateway 200 and the terminal 500 may establish a network connection through a local area network or a wide area network path. The network connection interacts with the terminal 500, so that the user can control the smart device 100 connected to the gateway 200 to perform corresponding actions through the terminal 500 or the smart device 100.

[0057] Among them, the smart device 100 can be a smart camera, a smart curtain motor, a smart air conditioner, a smart door lock, a smart light, a smart TV, a smart printer, a smart sweeper, a smart robot, and can also be an electronic device such as a human body sensor, a door and window sensor, a temperature and humidity sensor, a water immersion sensor, a natural gas alarm, a smoke alarm, a wall switch, a wall socket, a wireless switch, a wireless wall switch, a magic cube controller, etc. equipped with a communication module. The terminal 500 can be a smart phone, a laptop, a personal computer, a tablet computer, an intelligent control panel, or other electronic device that can implement network connection, which is not limited here. The server 400 can be implemented by an independent server or a server cluster composed of multiple servers.

[0058] In the specific implementation process, the gateway 200 in the service recommendation system can be accessed through the smart device 100, and then the historical behavior data of the target user in the target space can be obtained through the smart device 100 deployed in the target space; then, based on the historical behavior data and the current spatial state and event information determined by the spatial object model, the service needs corresponding to the target user in different spatial areas in the target space are identified; finally, based on the service functions that the spatial object model can provide, service functions that are adapted to the service needs are generated and pushed.

[0059] In addition, the present application can also be implemented through hardware circuits or hardware circuits combined with software instructions. Therefore, the implementation of the present application is not limited to any specific hardware circuits, software, or a combination of the two.

[0060] It should be noted that the order of description of the following embodiments is not intended to limit the priority order of the embodiments.

[0061] See also Figure 2 A service recommendation method is provided, and the method is applied to an electronic device as an example for explanation. The electronic device may be Figure 1 The smart device, terminal, gateway, server and terminal, etc., are not specifically limited in this application. The service recommendation method may specifically include the following steps:

[0062] S1. Obtain the historical behavior data of the target user in the target space.

[0063] Specifically, for step S1, first obtain the historical behavior data of the target user in the target space (such as home, office, etc.), wherein the target user may refer to a user group for which service recommendation is required, for example, specifically may be the target user corresponding to a registered account.

[0064] Among them, historical behavior data can refer to data that records and analyzes the behavior patterns of users in a specific environment, which can specifically include but is not limited to information such as user activity tracks, device usage, timestamps, etc., as well as the collection activity tracks, device usage, timestamps, etc. of the above data. The collection method of historical behavior data can be achieved through smart home devices, sensors, user input, etc.

[0065] S2. Based on historical behavior data and the current space status and event information determined by the space object model, identify the service needs of the target user in different spatial areas under the target space; the space object model is constructed based on the regional attribute information and device information under the target space.

[0066] Among them, the spatial object model can be used to describe and simulate the physical characteristics and equipment layout of the home space. The spatial object model combines spatial information and equipment information, and can provide the status and behavior of equipment in the spatial area. For example, in a smart home system, the spatial object model can be used to identify users' service needs in different spatial areas and provide personalized service recommendations. For example, the spatial object model can specifically include information about devices such as smart lights, temperature controllers, and smart cameras in the living room, as well as the layout and functions of these devices in the living room space. By analyzing the user's historical behavior data and current spatial status in the living room, it is possible to identify the user's service needs, such as adjusting the light brightness, starting the entertainment system, and other service needs.

[0067] It can be understood that the current space state may refer to the actual environmental conditions, device status and specific circumstances of user activities in the target space area at a specific moment, which is used to reflect the latest state of the target space, including various device statuses, environmental parameters and user behavior data. Among them, the specific moment may specifically be the moment when the service recommendation process is being performed. For example, taking the smart home service recommendation system as an example, the current space state can be used to identify and predict the user's service needs in order to provide personalized service recommendations. For example, the current space state of the target space may include: the smart lights in the living room are on, the brightness is set to 70%, the indoor temperature is 22°C, the TV is playing the user's favorite channel, and the user is watching TV in the living room.

[0068] Among them, event information refers to various information describing events that occur in the target space area in the smart system, such as device state changes, user behavior, environmental changes, etc. Taking the smart home service recommendation system as an example, event information can be used to identify user needs and space state changes in real time and provide timely service recommendations. For example, event information may include: the smart door lock is successfully unlocked at a specific time, indicating that the user has returned home; the smart light automatically turns on at night because the ambient light dims and the user is detected in the room; the smart smoke alarm sounds an alarm because smoke is detected in the kitchen; the temperature recorded by the smart thermostat suddenly drops, which may be because the window was accidentally opened.

[0069] Among them, service requirements may refer to specific functions, experiences or support requirements that users expect to be satisfied in the process of using products or services. In this embodiment, service requirements may refer to specific service requirements that users expect to be provided with automation, personalization or optimization of family life in a smart home environment. For example, service requirements may include: automatically adjusting the brightness and color of lights based on the user's activity time and light preferences in the living room; automatically adjusting the indoor temperature according to the outdoor temperature and the comfort level set by the user; automatically reducing the light and volume in the bedroom when it is detected that the user is ready to go to bed; and providing home security monitoring and alarm services when the user is out.

[0070] Specifically, for step S2, it is also necessary to use the regional attribute information (such as room size, layout, orientation, etc.) and device information (smart lights, temperature controllers, and smart cameras, etc.) under the target space to build a spatial object model. The construction of the spatial object model is mainly based on the regional attribute information and device information under the target space, including defining the hierarchical structure of the home space (such as floors, rooms, areas, etc.), and defining attributes for each spatial element (such as room, equipment), such as name, location, type, and function.

[0071] For example, the physical features and equipment layout of the home space are described through the spatial object model, including but not limited to the type of room, the location and function of the equipment, etc. After obtaining the user's historical behavior data and determining the current spatial state of the target space through the spatial object model (for example, determining the current spatial state through data collected in real time by sensors), the service needs of the target user in different spatial areas are identified. For example, if it is detected that the user is currently in the living room area and the light is dim, it is recognized that the user may need to turn on the lights.

[0072] S3. Based on the service functions that the spatial object model can provide, generate and push service functions that are adapted to the service needs.

[0073] Among them, service functions can refer to specific operations that can provide intelligent and personalized functions or optimize existing functions. Taking the smart home system as an example, service functions can provide users with a comfortable, convenient, safe, energy-saving, intelligent and personalized home living environment. For example, the service function can be an automated lighting function that automatically adjusts the light brightness according to the time or light sensor. Or an intelligent temperature control function that automatically adjusts the indoor temperature according to the indoor and outdoor temperature difference and user settings. Or it can be pre-set multiple scene modes based on user habits and activities, such as "movie viewing mode" that automatically dims the lights and turns on the TV, etc.

[0074] Specifically, for step S3, after identifying the service needs of the target user in different spatial areas, the service functions that are adapted to the user's service needs are generated and pushed according to the service functions that the spatial object model can provide. For example, if it is identified that the user has the need to watch TV in the living room area, the command to control the smart TV or recommend a specific TV program can be pushed to the user.

[0075] It can be seen that the service recommendation method provided in this embodiment can provide users with more personalized service recommendations and enhance users' smart home experience by analyzing users' behavioral data and spatial object models; at the same time, combined with real-time spatial status and event information, it can more accurately identify users' service needs in different spatial areas, thereby improving the accuracy and relevance of service recommendations.

[0076] In some embodiments, the steps of constructing the spatial object model may specifically include:

[0077] Obtain the hierarchical relationship of each area in the target space, as well as the device information of the smart devices bound to each area;

[0078] Determine the functions that each smart device can provide based on the device information of each smart device;

[0079] Based on the hierarchical relationship between each area and the functions that each smart device can provide, a spatial object model corresponding to the target space is constructed.

[0080] Specifically, for the construction of the spatial object model in this embodiment, the specific steps are as follows: First, it is necessary to obtain the hierarchical relationship between each area under the target space, such as floors, rooms, areas, etc., as well as the device information of the smart devices bound to these areas. The device information may specifically include the type, location, function and status of the device.

[0081] The process of determining the hierarchical relationship of each area is as follows: Predefined association structure In the database or model definition of the system, the general hierarchical structure of the home space is predefined. It includes floors, which are used to define different floors of the home; rooms define rooms under each floor, such as living room, bedroom, kitchen, etc.; areas refer to further division of areas within the room. User actual room and equipment construction import Provide a user interface or application, and the user enters or selects the corresponding room and equipment information according to their own home layout and equipment configuration. Identification and matching The system identifies and matches the room information entered by the user and the predefined spatial element association structure, and dynamically builds the spatial hierarchical relationship Based on the results of identification and matching, the hierarchical relationship of the user's home space is dynamically constructed, such as associating rooms with floors, and associating the rooms identified by the user with the corresponding floors. Associate devices with rooms, associate the devices selected by the user with the corresponding rooms, and clarify the installation location and control range of the devices. Users can also add additional areas or custom areas, and associate these areas with the corresponding rooms through the operations provided by the interface to further improve the hierarchical relationship of the entire home space.

[0082] Then, based on the device information of each smart device, determine the corresponding functions that each smart device can provide. Specifically, it is necessary to collect the basic information of various smart devices, including the type, model, functional characteristics, current status, location, etc. of the device. Define the capabilities of each smart device based on the basic information of the smart device, that is, what operations the device can perform or what services it can provide. Then map the capabilities of the device to specific service functions, and generate corresponding service functions based on user needs and the available capabilities of the device. For example, lighting equipment can adjust brightness and color, and temperature controllers can adjust indoor temperature. Finally, using the regional hierarchical relationship and device functions obtained above, construct the corresponding spatial object model of the target space. The spatial object model serves as a basic tool for subsequent service recommendation, and is used to describe the physical characteristics and device layout under different target spaces, as well as the relationship between physical characteristics and device layout.

[0083] This embodiment constructs a spatial object model to more accurately describe the physical characteristics and equipment layout of the target space, more effectively manage and optimize the use of the target space, provide an accurate and reliable data basis for subsequent service recommendations, and identify the functions of various smart devices to better adapt to user needs in different environments.

[0084] In some embodiments, Figure 3 As shown, step S2 "identifying the service needs of target users corresponding to different spatial areas in the target space based on historical behavior data and the current spatial state and event information determined by the spatial object model" may specifically include:

[0085] S21. Extract features from the user behavior data and the device status data based on the spatial object model to obtain corresponding user behavior features and device status features;

[0086] S22. Determine the current space state and event information based on the user behavior characteristics and device state characteristics;

[0087] S23. According to the current space state and event information, identify the service needs of the target user corresponding to different space areas under the target space.

[0088] Among them, user behavior characteristics can refer to the description of the user's behavior patterns and habits in a specific environment or situation. Taking the smart home environment as an example, user behavior characteristics can refer to the mode of user interaction with smart home devices, including usage frequency, time, preference settings, etc. For example, user behavior characteristics may include: the user prefers to dim the living room lights between 8 and 10 pm for relaxation and watching TV; the user tends to set the indoor temperature at 22°C in summer and 20°C in winter; the user is accustomed to turning off all lights and starting the security system when leaving home. Device status characteristics can refer to the set of operating conditions and performance indicators of the device at a specific point in time, including the device's on / off status, performance parameters, fault status, etc. Taking the smart home environment as an example, device status characteristics can refer to the detailed operating conditions and performance indicators of smart home devices at a specific point in time, such as the brightness and color settings of smart lights, the current temperature setting of smart thermostats, the lock status of smart door locks, etc.

[0089] Specifically, the identification of service requirements in step S2 includes the following steps: first, extracting features from user behavior data and device status data based on the spatial object model, such as analyzing the user's behavior pattern in the current space and the status of the smart device in the current space, thereby determining the user behavior features and device status features;

[0090] Then, the extracted user behavior characteristics and device status characteristics are used to determine the current space status and event information, such as identifying the activity status of the target user in the current room and the working status of all smart devices in the current room; finally, based on the current space status and event information, the service needs of the target user in different spatial areas under the target space are identified. Through the sensors and smart devices deployed in the target space, the space status and event information are collected in real time, and the user's behavior patterns in specific spatial areas, such as activity time, frequency and duration, are analyzed. Cluster analysis is performed on user behavior data and device status data to identify similar behavior patterns or device configurations. Based on user behavior characteristics and device status characteristics, the service needs of users in different spatial areas are predicted. For example, when the user's activity in the kitchen area is monitored to increase, it is identified that the user may need to adjust the kitchen lights or start the coffee machine.

[0091] By analyzing user behavior and device rotation characteristics, this embodiment can more accurately identify users' service needs in different spaces, thereby improving the accuracy of subsequent service recommendations, allowing users to obtain service recommendations that are more in line with their current activity status and preferences, thereby improving user experience.

[0092] In some embodiments, step S23 "identifying service requirements corresponding to different spatial areas of the target user under the target space according to the current spatial state and event information" may specifically include:

[0093] S231. Perform data mining on the current space status, device status data and user behavior data to obtain corresponding data processing results;

[0094] S232. Based on the data processing results, identify the service needs of the target user in different spatial areas under the target space.

[0095] Specifically, for step S23, before identifying the service needs of the target user, it is also necessary to mine and process the spatial state, device state data and user behavior data of the target space where the target user is located, including but not limited to data mining through cluster analysis, association rule mining and classification algorithms. The obtained data processing results can be used to analyze the behavior pattern or trend of the target user in the target space. It can be understood that the data processing results can be cluster analysis results, association rule mining results and prediction model results, etc. Through clustering algorithms, such as K-means or hierarchical clustering, user behavior data and device state data are grouped into different patterns or categories. Or use association rule mining technology, such as Apriori algorithm, to find the correlation between different device states and user behavior. Or use machine learning algorithms, such as support vector machines (SVM) or random forests, to establish a prediction model to predict the user's future behavior or device state changes, etc., which is not limited in this application.

[0096] For example, clustering algorithms are used to perform cluster analysis on the space status, device status data, and user behavior data respectively to obtain corresponding data clustering results. The activity behavior patterns and device usage status of the target user in the target space are determined based on the data clustering results, and the target user's preferences in different spatial areas are determined, thereby identifying or predicting the target user's corresponding service needs in different spatial areas in the target space. For example, when it is identified that the user's activities in the kitchen area increase, it is predicted that the user may need to adjust the kitchen lights or start the coffee machine, etc.

[0097] This embodiment more accurately identifies the user's behavior patterns and preferences by performing data mining on the current space status, device status data, and user behavior data, thereby providing the user with more relevant and personalized service recommendations, so that the user subsequently receives service recommendations that are more in line with their habits and needs, significantly improving the user experience.

[0098] In some embodiments, the data processing results in this embodiment may specifically include data clustering results; step S231 "perform data mining processing on the current space state, device state data and user behavior data respectively to obtain corresponding data processing results" may specifically include:

[0099] Perform cluster analysis on the current space status, device status data and user behavior data respectively to obtain the corresponding data clustering results;

[0100] Based on the clustering results of each data, determine at least one of the regional characteristics of each spatial area in the target space, the user behavior pattern in each spatial area, and the device association result between each device in each spatial area;

[0101] Based on at least one of the device association result, the regional characteristics and the user behavior pattern, the service needs of the target user corresponding to different spatial areas under the target space are predicted.

[0102] Among them, the regional characteristics of the spatial area may refer to the physical layout, functional division and usage mode of a specific spatial area. In this embodiment, the regional characteristics may refer to the characteristics of the automation equipment configuration, user behavior pattern and environmental settings of a specific spatial area in the target space. Taking the target space as a house as an example, the target space may include areas such as the living room, dining room, and bedroom. For example, the regional characteristics of the living room may include: the types and layout of smart lights, smart thermostats, and smart entertainment systems, the types of activities of users in the living room (such as watching TV, family gatherings) and activity times (such as 8 to 10 p.m.), and the user's preferred light brightness, color, and temperature settings.

[0103] Among them, user behavior patterns may refer to the user's behavior habits and behavior patterns in a specific application or service, such as usage frequency, activity time, function preferences, etc. For example, in a smart home environment, user behavior patterns may include regular behaviors of users interacting with smart home devices, including device usage time, usage frequency, setting preferences, etc.

[0104] It can be understood that the device association result between devices refers to the result of analyzing the data interaction and functional dependency between different devices in the Internet of Things or smart device system to determine whether there is a certain association or interaction between them. For example, the device association result can be the result of the mutual relationship and dependency between smart devices in terms of function, usage scenario or user behavior determined through data mining and analysis. For example, when a user watches TV in the living room, he tends to dim the smart lights and turn on the home theater system. This device association result can be used to automatically create a "movie viewing mode". When the user chooses to watch a movie, the light brightness is automatically adjusted and the theater system is started to enhance the user's viewing experience.

[0105] Specifically, the data processing results in step S231 may specifically include but are not limited to data clustering results, and step S231 may specifically include the following steps: First, cluster analysis is performed on the current spatial state, device state data, and user behavior data respectively through a data clustering analysis algorithm. Specifically, various data from smart devices and the current system can be obtained, including spatial state data (such as temperature, humidity, and light intensity), device state data (such as the switch state and setting parameters of smart devices), and user behavior data (such as user activity trajectories and device usage frequency). According to the characteristics of the data and the analysis objectives, a suitable clustering algorithm is selected. According to the selected algorithm, the clustering center is initialized or the clustering parameters are set. Then, the current spatial state, device state data, and user behavior data are clustered and analyzed by a clustering algorithm, and similar data points are classified into one category, thereby obtaining the data clustering results corresponding to the current spatial state, device state data, and user behavior data, respectively.

[0106] Then, based on different data clustering results, the regional characteristics of each spatial area under the target space, the user behavior patterns, and the association results between devices are determined, so as to identify the activities and preferences of users in different spatial areas. Specifically, by analyzing the spatial regional characteristics represented by each clustering result, such as a living room cluster may show a higher frequency of light use and a lower indoor temperature setting at night. Then, based on the spatial regional characteristics, the association between the devices is identified, such as smart lights and smart curtains may be activated at the same time when the user enters the room, indicating that the two have a synergistic effect. By analyzing the clustering results of user behavior data and the association between devices, the user's behavior patterns in different spatial areas are identified, such as users in the bedroom area may prefer to dim the lights before going to bed.

[0107] Finally, based on one or more of the device association results, regional characteristics, and user behavior patterns, the service needs of the target user in different spatial areas under the target space are predicted. Specifically, various data from smart devices and the current system can be collected, including space status data (such as temperature, humidity, light intensity), device status data (such as the switch status of smart devices, setting parameters), and user behavior data (such as user activity trajectory, device usage frequency). Analyze the correlation between different devices, such as the linkage relationship between smart lights and smart door locks, and the association between smart thermostats and window opening and closing states. According to the spatial object model, analyze the characteristics of different spatial areas, such as the device layout, user activity patterns, and environmental parameters in the living room, bedroom, kitchen, etc. Analyze the user's behavior patterns in different spatial areas, such as activity time, frequency, and duration. Or use sequence pattern mining and machine learning techniques, such as hidden Markov model (HMM) and support vector machine (SVM), to identify the regularity and preference of user behavior. Finally, the device association results, regional characteristics, and user behavior patterns are combined to predict the user's service needs in different spatial areas. For example, when it is recognized that the user's behavior pattern in the bedroom area indicates that he is ready to go to bed, it is predicted that the user may currently have personalized service needs such as dimming the lights or lowering the temperature.

[0108] The data clustering analysis algorithm of this embodiment performs in-depth analysis of user behavior and device usage patterns, and can more accurately predict user service needs and provide more personalized service recommendations, so that users can receive service recommendations that are more in line with their habits and needs.

[0109] In some embodiments, Figure 4 As shown, step S3 "generating and pushing service functions adapted to service requirements based on the service functions that the spatial object model can provide" may specifically include:

[0110] S31. Determine the service functions that the space object model can provide based on the device attribute information and device capability information of each smart device in the target space;

[0111] S32. From the service functions that the space object model can provide, obtain the service functions that are adapted to the service requirements and push them.

[0112] Device attribute information refers to detailed information describing the functions, performance, status, and user settings of smart devices, which is used by the system to better control and optimize the operation of the device. For example, in a smart home environment, the device attribute information of a smart thermostat includes the current set temperature, temperature adjustment range, energy-saving mode settings, and user temperature preferences.

[0113] Device capability information may refer to information describing the specific operations that a smart device can perform and the scope of services it can provide. For example, in a smart home environment, the device capability information of a smart lighting system includes the brightness level, color change, timer switch, and voice control functions that it can adjust.

[0114] Specifically, for step S3, first analyze the device attribute information and device capability information of each smart device in the target space. Specifically, obtain the basic information of each smart device, including the device type (such as smart lights, smart thermostats, smart door locks, etc.), manufacturer, model, installation location, and user-configured settings (such as brightness preferences, temperature settings, etc.). Obtain the functions and performance parameters of each smart device, such as the dimming range of smart lights, the temperature adjustment range of smart thermostats, and the unlocking methods of smart door locks. Classify the devices according to their functions and attribute information, such as classifying all lighting devices into one category and temperature control devices into another category. Extract key features from the device attribute information, such as the response speed of the device, energy consumption level, user-defined settings, etc. For example, the input and output parameters, operating range, and role of smart devices in the smart home ecosystem of smart devices can be included to determine what service functions each smart device can currently provide.

[0115] Then, from the service functions that the spatial object model can provide, by comparing and matching the device functions and device attribute information with user needs, matching rules can be formulated based on the characteristics of device functions and user needs. For example, if the user prefers to use soft lights at night, the matching rules will recommend that the smart lights automatically adjust to low brightness mode at night. And use algorithms (such as collaborative filtering, content-based recommendation algorithms, deep learning, etc.) to achieve the match between device functions and user needs. Based on the results of the matching algorithm, functions that match the user's service needs are screened out.

[0116] For example, when the user is ready to go to bed, it is determined that the user's needs are to dim the lights or lower the temperature. At this time, the status of the smart lights and air conditioner in the master bedroom is analyzed to filter out the functions that match the user's service needs (dimming the lights or lowering the temperature). Finally, the service functions are pushed to the user. The push can be automatically executed (such as automatically adjusting the light brightness and automatically lowering the temperature) or recommended to the user (such as suggesting that the user turn on a certain device), so that the user can receive service recommendations that are more in line with their habits and needs.

[0117] This embodiment reduces the user's operating burden through automated device control, improves the automation and intelligence level of the smart home system, enables users to receive service recommendations that are highly matched to their needs, and improves the user's overall satisfaction with the smart home system.

[0118] In some embodiments, Figure 4As shown, after step S3 of "generating and pushing a service function adapted to the service demand", the service recommendation method may further specifically include:

[0119] S33. Calculate the matching degree between the service demand and each service function, where the matching degree represents the adaptability between the service demand and each service function;

[0120] S34. If the matching degree reaches the first threshold range, the target device in the target space area is started based on the service function to instruct the target device to provide a recommended service corresponding to the service function;

[0121] S35. If the recommendation degree reaches the second threshold range, the service function is pushed to the target user to remind the target user whether to apply the recommended service corresponding to the service function.

[0122] Specifically, after the service function adapted to the service demand is generated and pushed, the matching degree between the service demand and the service functions adapted to it is calculated first. The matching degree is a quantitative indicator for measuring the degree of adaptation between the service demand and the service function. The first threshold range and the second threshold range can be set in advance. The first threshold range is used to identify highly matching service functions, and the second threshold range is used to identify possible matching service functions.

[0123] Specifically, the user's service needs can be identified through user behavior data, device status data, and space status data. For example, based on the user's activity pattern and time in the living room, identify the environment that the user needs to relax between 8 and 10 pm. Then extract key features from the service needs, such as time period, environmental parameters (brightness, temperature), user preferences, etc., to determine the service functions that each device can provide, and extract key features from the service functions. By matching the key features corresponding to the service functions with the key features corresponding to the service needs. Specifically, methods such as cosine similarity, Euclidean distance or other similarity metrics can be used to calculate the similarity between the feature vector corresponding to the service demand and the feature vector corresponding to the service function, and use the similarity as the matching degree between the service demand and the service functions it adapts to.

[0124] If the matching degree between the service demand and a service function is within the first threshold range, the target device in the target space area will be automatically started and instructed to provide the corresponding recommended service without user confirmation or operation; if the matching degree between the service demand and a service function is within the second threshold range, the service function will be pushed to the target user, and the target user will be provided with a recommendation service corresponding to the service function, so that the user can flexibly decide whether to accept the recommendation according to his current situation. In addition, the first threshold range and the second threshold range can be automatically adjusted according to user feedback and behavior patterns, or the user can set the threshold and matching rules according to his own preferences.

[0125] This embodiment determines whether the device needs to be automatically started by the matching degree between the service demand and the service functions it is adapted to, thereby reducing the user's operating burden, providing more automated and intelligent services, and effectively enhancing the user experience; and can intelligently schedule the device according to the user's actual needs, avoiding unnecessary energy waste and improving the efficiency of device use.

[0126] In some embodiments, the service recommendation method may further include:

[0127] Get the configured preference configuration data;

[0128] Based on historical behavior data and preference configuration data, as well as the current space status and event information determined by the space object model, the service needs of target users in different spatial areas under the target space are identified.

[0129] Among them, preference configuration data can refer to a series of parameters and choices set by users to personalize their smart home experience, reflecting the user's expected behavior of smart devices. For example, users can set their preferred indoor temperature. For example, some users may like to set the air conditioner to 22°C in summer and 20°C in winter. Users can set the light brightness in different rooms or different time periods, such as preferring brighter lights when reading and softer lights when relaxing.

[0130] Specifically, the service recommendation method provided in this embodiment also provides a method for further improving the personalization of service recommendations by incorporating the user's preference configuration data. The specific process is as follows: first, obtain the user's configured preference configuration data, including the user's personalized settings for smart home devices, such as temperature preferences, lighting brightness, and music types; then, combine the user's historical behavior data with the configured preference data, where the historical behavior data provides the user's past behavior patterns in specific situations, and the preference data provides the user's explicit preference settings; determine the current spatial state and event information through the spatial object model, and finally identify the target user's service needs in different spatial areas by combining the historical behavior data, preference configuration data, and the current spatial state and event information. Specifically, data mining technology can be used to analyze the preprocessed data to mine information about user needs and behavior patterns, such as identifying the characteristics of different spatial areas and user behavior patterns through a cluster analysis algorithm. The correlation and influence between devices are discovered through an association rule mining algorithm. Recommendations are made based on the user's needs and preferences in different spatial areas through a recommendation algorithm. According to the current spatial state, device state, and user behavior data, the user's service needs in different spatial areas are identified. From the service functions that the spatial object model can provide, select the functions that match the user's service needs.

[0131] This embodiment can provide more customized service recommendations and enhance user experience by considering the user's personalized preferences. It can more accurately predict and meet the user's service needs by combining the user's real-time status, historical behavior and explicit preferences.

[0132] In some embodiments, the service function is obtained by processing a machine learning model, and the training steps of the machine learning model include:

[0133] Obtain sample data and corresponding sample labels; the sample data includes sample historical behavior data, sample space status, and sample event information; the sample labels include labels used to indicate service requirements;

[0134] Extracting service demand characteristics corresponding to sample data;

[0135] Match the sample data to the corresponding estimated service demand based on the service demand characteristics through the initial machine learning model;

[0136] Based on the difference loss value between the estimated service demand and the sample label, the model parameters of the initial machine learning model are adjusted, and the initial machine learning model is continuously trained based on the adjusted model parameters until the training conditions are met, thereby obtaining a trained machine learning model.

[0137] Among them, sample data refers to data used to train machine learning models, and sample data includes sample historical behavior data, sample space status and sample event information; among them, sample historical behavior data may include historical data recording the user's interaction with smart devices, such as the frequency of turning on and off lights, adjusting temperature settings, time periods for watching TV, and other data; sample space status includes the space status corresponding to the corresponding time in space, which may specifically be the space status corresponding to the sample historical behavior data, such as the temperature, humidity, light intensity, etc. of the space; sample event information is event information recorded in the sample data, which may specifically be event information corresponding to the sample historical behavior data, and is used to record the occurrence of specific events, such as the user returning home, leaving home, starting sleep mode, etc.

[0138] It can be understood that in machine learning, the sample label is the value associated with each sample data, which is what the model tries to predict or learn. Sample labels include service demand labels, which are used for the expected service results corresponding to the sample data, such as "need to turn on the lights", "set the temperature to 22℃", "play relaxing music", etc.; user feedback labels are used to characterize the user's feedback on the recommended service, such as "accept", "reject", "adjust brightness", etc.

[0139] Specifically, this embodiment obtains the service function through the processing of the machine learning model. For the machine learning model, it is necessary to train the machine learning model before use. The specific steps are as follows: collect sample data, including the user's historical behavior data, spatial status and event information, and assign corresponding sample labels to these sample data. The sample labels represent service requirements. Extract key service demand features related to service requirements from the sample data, including time patterns, device usage frequency, environmental conditions and other features. Use the extracted features to train an initial machine learning model to predict service requirements. Specifically, key features related to service requirements can be extracted from the sample data. These features include time features, environmental features, user behavior features and device status features; select a suitable machine learning algorithm as the initial model. Use the extracted features and corresponding sample labels to train the initial machine learning model, including data segmentation, model training and hyperparameter adjustment, and finally evaluate the performance of the model to ensure that it can accurately predict service requirements. Make the model learn how to match the estimated service requirements based on the input features. By comparing the difference (loss value) between the estimated service requirements and the actual labels, adjust the parameters of the machine learning model, and minimize the prediction error through this back propagation process. Continue to train the model until the preset training conditions are met, such as reaching a certain accuracy rate or number of training rounds. Finally, you will get an optimized, trained machine learning model that can accurately predict service demand.

[0140] This embodiment uses a machine learning model to more accurately predict users' service needs, thereby providing services that better meet users' expectations; through continuous learning, the prediction model is continuously optimized to adapt to changes in user behavior and improve the accuracy of subsequent service recommendations.

[0141] The present application also provides an application scenario, which applies the above-mentioned service recommendation method. Specifically, taking the application of the method to a smart home scenario as an example, the application of the service recommendation method in the application scenario is as follows:

[0142] Assume that a user has bound a home space in the smart home system. There are multiple spatial areas in the home space, including living room, dining room, bedroom, kitchen, office, etc. Family members have different living habits and preferences. The goal of this application is to provide automated and personalized home services.

[0143] First, build a spatial object model. In the spatial data model building phase, the system builds the spatial data model through the following steps:

[0144] Define the hierarchical structure of the home space, such as floors, rooms, areas, etc., to form a hierarchical relationship of the home space.

[0145] Define properties for each space element (such as room, equipment), such as name, location, type and function. Each space element has its own unique identifier to facilitate system identification and access. Define the hierarchy of home space and build it according to the user's actual rooms and equipment. This can be achieved in the system through the following detailed processing flow:

[0146] (1) System predefined association structure: The system predefines spatial elements at various levels (such as floors, rooms, and areas) and the associations between them (such as floors contain multiple rooms, and rooms belong to specific areas). These predefined association structures can serve as the basis for spatial data models.

[0147] (2) User actual room and device construction import: Provide a user interface or application to allow users to enter or select their actual room and device information. Users can manually enter or select the corresponding rooms and devices from the device list provided by the system based on their home layout and device configuration.

[0148] (3) Identification and matching: The system identifies and matches the room information entered by the user with the predefined spatial element association structure. The system can match the room name or other description information entered by the user with the predefined association structure to find the corresponding floor, area, and defined room.

[0149] For rooms that cannot be matched, the system can provide hints and suggestions to help users make correct matches or make manual adjustments.

[0150] (4) Dynamically construct spatial hierarchical relationships: Based on the results of identification and matching, the system dynamically constructs the hierarchical relationship of the user's home space. For example, the identified room is associated with the corresponding predefined room, and the room is associated with the floor to form a correct hierarchical structure. Users can also add additional areas or custom areas, and through the operations provided by the interface, associate these areas with the corresponding rooms to further improve the hierarchical relationship of the entire home space. Establish associations between spatial elements, such as the association between devices and rooms, and the association between rooms and floors. These associations can be expressed through parent-child relationships, connection relationships, or other methods.

[0151] Then, the real-time status and event information of the equipment are obtained and reported. The sensor network and smart devices are used to collect the relevant status and event information of the space in real time and report it to the system. Sensors can include temperature, humidity, light, equipment switch status, human body sensing, etc. Through the collection and reporting of sensor data, the system can obtain real-time space status and event information, providing real-time data support for subsequent analysis and recommendations.

[0152] The space status and event information collected in real time can be reported to the smart home hub device or system. This hub device or system plays the role of data collection, processing and analysis. If there is no hub but only a basic gateway, it can be reported to the gateway and synchronized to the cloud platform through routing, and processed in the system cloud.

[0153] Next, the collected real-time status and event information needs to be preprocessed to improve data quality and accuracy. Preprocessing includes data cleaning, data conversion, feature extraction, and spatial data conversion. Data mining technology is used to analyze the preprocessed data to mine information about user needs and behavior patterns.

[0154] Finally, based on the spatial data model and the results of data mining analysis, real-time smart home service recommendations are generated. Based on the mined associations and analysis results, personalized smart home service recommendations are generated. The recommendation process comprehensively considers multiple factors such as spatial characteristics, real-time status and event information, user needs, and device capabilities to provide accurate and real-time service recommendations. The recommendation results are presented to users through smart home terminal devices, mobile applications, or voice assistants to help users select and control the corresponding smart home services.

[0155] In this smart home smart service recommendation system, real-time recommendation refers to recommending suitable smart home services to users in real time based on factors such as user needs in different spatial areas, spatial characteristics, real-time status and event information, and device capabilities. Specifically, real-time recommendation can include the following:

[0156] Control of smart home devices: Based on the space status and user needs, the system can recommend ways and parameter settings for users to control specific devices, such as adjusting lighting brightness, controlling temperature, opening windows, etc.

[0157] Home security and monitoring: The system can recommend security-related services based on the association between spatial elements and real-time status information, such as reminding users to lock doors and windows, monitor gas leaks, detect fire alarms or emergencies, etc.

[0158] Energy management: Based on real-time energy usage and user needs, the system can recommend energy-saving measures and control equipment to optimize energy efficiency.

[0159] Personalized services: Based on the user's needs and preferences in different spatial areas, the system can recommend personalized services that match the user's interests, hobbies and behavior patterns, such as music playback, TV program recommendations, scene mode switching, etc.

[0160] The content of real-time recommendations can be defined by system developers based on domain knowledge and user needs. In system design, real-time recommendation content can be personalized based on user preferences, personal settings, historical behavior data, and other information. The system can continuously optimize and adjust recommendation strategies based on the needs and preferences of different users through data mining analysis and machine learning algorithms to ensure that real-time recommendation services that better meet user needs are provided.

[0161] In order to better implement the service recommendation method of the embodiment of the present application, the embodiment of the present application also provides a service recommendation device based on the above service recommendation method. The meanings of the terms are the same as those in the above service recommendation method, and the specific implementation details can refer to the description in the method embodiment.

[0162] See also Figure 5 , Figure 5 20 is a schematic diagram of the structure of a service recommendation device provided in an embodiment of the present application. The service recommendation device includes an acquisition module 201, an identification module 202 and a recommendation module 203. Specifically, it can be as follows:

[0163] An acquisition module 201 is used to acquire historical behavior data of a target user in a target space;

[0164] Identification module 202, used to identify the service needs of target users corresponding to different spatial areas in the target space based on historical behavior data and current spatial status and event information determined by a spatial object model; the spatial object model is constructed based on regional attribute information and device information in the target space;

[0165] The recommendation module 203 is used to generate and push service functions adapted to the service requirements based on the service functions that the space object model can provide.

[0166] In some embodiments, the steps of constructing the spatial object model include:

[0167] Obtain the hierarchical relationship of each area in the target space, as well as the device information of the smart devices bound to each area;

[0168] Determine the functions that each smart device can provide based on the device information of each smart device;

[0169] Based on the hierarchical relationship between each area and the functions that each smart device can provide, a spatial object model corresponding to the target space is constructed.

[0170] In some embodiments, the identification module 202 may specifically include:

[0171] An extraction unit, used to extract features from the user behavior data and the device status data based on the spatial object model, to obtain corresponding user behavior features and device status features;

[0172] A determination unit, used to determine the current space state and event information according to the user behavior characteristics and device state characteristics;

[0173] The identification unit is used to identify the service needs of the target user corresponding to different spatial areas under the target space according to the current space state and event information.

[0174] In some embodiments, the identification unit may specifically include:

[0175] The mining subunit is used to perform data mining on the current space status, device status data and user behavior data to obtain corresponding data processing results;

[0176] The identification subunit is used to identify the service needs of the target user corresponding to different spatial areas under the target space based on the data processing results.

[0177] In some embodiments, the data processing results include data clustering results; the mining sub-unit is used to: perform cluster analysis on the current spatial status, device status data and user behavior data respectively to obtain corresponding data clustering results; based on each data clustering result, determine at least one of the regional characteristics of each spatial area under the target space, the user behavior pattern under each spatial area and the device association results between each device under each spatial area; based on at least one of the device association results, regional characteristics and user behavior patterns, predict the service needs of the target user corresponding to different spatial areas under the target space.

[0178] In some embodiments, the recommendation module 203 may specifically include:

[0179] A service determination unit, used to determine the service functions that the space object model can provide according to the device attribute information and device capability information of each smart device in the target space;

[0180] The service push unit is used to obtain and push service functions that are adapted to the service requirements from the service functions that the space object model can provide.

[0181] In some embodiments, the recommendation module 203 may further include:

[0182] A calculation unit, used to calculate the matching degree between the service demand and each service function, where the matching degree represents the adaptability between the service demand and each service function;

[0183] A first recommendation unit, configured to start a target device in a target space area based on a service function if the matching degree reaches a first threshold range, so as to instruct the target device to provide a recommended service corresponding to the service function;

[0184] The second recommendation unit is used to push the service function to the target user when the recommendation degree reaches a second threshold range, so as to remind the target user whether to apply the recommended service corresponding to the service function.

[0185] In some embodiments, the identification module 202 is also used to: obtain configured preference configuration data; identify the service needs of the target user corresponding to different spatial areas under the target space based on historical behavior data and preference configuration data, as well as the current spatial status and event information determined by the spatial object model.

[0186] In some embodiments, the service function is obtained by processing a machine learning model, and the training steps of the machine learning model include:

[0187] Obtain sample data and corresponding sample labels; the sample data includes sample historical behavior data, sample space status, and sample event information; the sample labels include labels used to indicate service requirements;

[0188] Extracting service demand characteristics corresponding to sample data;

[0189] Match the sample data to the corresponding estimated service demand based on the service demand characteristics through the initial machine learning model;

[0190] Based on the difference loss value between the estimated service demand and the sample label, the model parameters of the initial machine learning model are adjusted, and the initial machine learning model is continuously trained based on the adjusted model parameters until the training conditions are met, thereby obtaining a trained machine learning model.

[0191] In summary, the service recommendation device provided in this embodiment obtains the historical behavior data of the target user in the target space through the acquisition module 201; identifies the service needs corresponding to different spatial areas of the target user in the target space through the identification module 202 based on the historical behavior data and the current spatial state and event information determined by the spatial object model; the spatial object model is constructed based on the regional attribute information and device information in the target space; and generates and pushes service functions adapted to the service needs based on the service functions that the spatial object model can provide through the recommendation module 203. This embodiment accurately identifies the services that the user may need in the current space by combining the user's historical behavior and the current spatial state, and then generates the service functions that are most adapted to the service needs through the service functions of the spatial object model and pushes them to the user, thereby providing the user with personalized service recommendations that meet the user's needs, effectively improving the accuracy and relevance of the service recommendations, and thus improving the user experience.

[0192] In one embodiment, a computer device is provided. The computer device may be a terminal, and its internal structure diagram may be as follows: Figure 6As shown. The computer device includes a processor, a memory, a network interface, a display screen and an input device connected via a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The network interface of the computer device is used to communicate with an external terminal via a network connection. When the computer program is executed by the processor, a target detection method is implemented. The display screen of the computer device can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device can be a touch layer covered on the display screen, or it can be a computer device.

[0193] Those skilled in the art will understand that Figure 6 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0194] In one embodiment, a computer device is further provided, including a memory and a processor, wherein a computer program is stored in the memory, and the processor implements the steps in the above method embodiments when executing the computer program.

[0195] In one embodiment, a computer-readable storage medium is provided, storing a computer program, which implements the steps in the above method embodiments when executed by a processor.

[0196] In one embodiment, a computer program product is provided, including a computer program, which implements the steps in the above method embodiments when executed by a processor.

[0197] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0198] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0199] The above-described embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be construed as limiting the scope of the patent application. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the patent application shall be subject to the attached claims.

Claims

1. A service recommendation method, characterized in that: include: Obtain historical behavior data of target users in the target space; Based on the historical behavior data and the current space state and event information determined by the space object model, identifying the service needs of the target user corresponding to different space areas under the target space; The spatial object model is constructed based on the regional attribute information and device information in the target space; Based on the service functions that the spatial object model can provide, service functions adapted to the service requirements are generated and pushed.

2. The service recommendation method according to claim 1, characterized in that: The steps of constructing the spatial object model include: Obtain the hierarchical relationship of each area in the target space, as well as the device information of the smart devices bound to each area; Determine the functions that each smart device can provide based on the device information of each smart device; Based on the hierarchical relationship between each area and the functions that each smart device can provide, a spatial object model corresponding to the target space is constructed.

3. The service recommendation method according to claim 1, characterized in that: The identifying service requirements of target users corresponding to different spatial areas in the target space based on the historical behavior data and the current spatial state and event information determined by the spatial object model includes: Based on the spatial object model, feature extraction is performed on the user behavior data and the device status data to obtain corresponding user behavior features and device status features; Determine the current space state and event information according to the user behavior characteristics and the device state characteristics; According to the current space state and the event information, service requirements corresponding to different space areas of the target user in the target space are identified.

4. The service recommendation method according to claim 3, characterized in that: The identifying, according to the current space state and the event information, service requirements corresponding to different space areas of the target user in the target space includes: Performing data mining processing on the current space state, the device state data and the user behavior data respectively to obtain corresponding data processing results; Based on the data processing result, service requirements corresponding to different spatial areas of the target user in the target space are identified.

5. The service recommendation method according to claim 4, characterized in that: The data processing result includes a data clustering result; the data mining process is performed on the current space state, the device state data and the user behavior data respectively to obtain corresponding data processing results, including: Performing cluster analysis on the current space state, the device state data, and the user behavior data respectively to obtain corresponding data clustering results; Based on the clustering results of each data, determine at least one of the regional characteristics of each spatial area in the target space, the user behavior pattern in each spatial area, and the device association result between each device in each spatial area; Based on at least one of the device association result, the regional characteristics and the user behavior pattern, service demands of target users corresponding to different spatial areas under the target space are predicted.

6. The service recommendation method according to claim 1, characterized in that: The generating and pushing of service functions adapted to the service requirements based on the service functions that can be provided by the spatial object model includes: Determine the service functions that the space object model can provide based on the device attribute information and device capability information of each smart device in the target space; From the service functions that the space object model can provide, service functions that are adapted to the service requirements are obtained and pushed.

7. The service recommendation method according to claim 1, characterized in that: After generating and pushing the service function adapted to the service requirement, the method further includes: Calculating the matching degree between the service demand and each service function, wherein the matching degree represents the adaptability between the service demand and each service function; If the matching degree reaches a first threshold range, starting a target device in a target space area based on the service function to instruct the target device to provide a recommended service corresponding to the service function; If the recommendation degree reaches a second threshold range, the service function is pushed to the target user to remind the target user whether to apply the recommended service corresponding to the service function.

8. The service recommendation method according to claim 1, characterized in that: The method further comprises: Get the configured preference configuration data; According to the historical behavior data and the preference configuration data, as well as the current space state and event information determined by the space object model, the service needs of the target user corresponding to different space areas under the target space are identified.

9. The service recommendation method according to claim 1, characterized in that: The service function is obtained by processing a machine learning model, and the training steps of the machine learning model include: Acquire sample data and corresponding sample labels; the sample data includes sample historical behavior data, sample space status, and sample event information; the sample labels include labels corresponding to service requirements; Extracting service demand characteristics corresponding to the sample data; Matching the sample data with corresponding estimated service demands based on the service demand characteristics through an initial machine learning model; Based on the difference loss value between the estimated service demand and the sample label, the model parameters of the initial machine learning model are adjusted, and the initial machine learning model is continued to be trained based on the adjusted model parameters until the training conditions are met, thereby obtaining a trained machine learning model.

10. A service recommendation device, characterized in that: include: The acquisition module is used to obtain the historical behavior data of the target user in the target space; An identification module, used to identify the service needs of the target user corresponding to different spatial areas under the target space based on the historical behavior data and the current space state and event information determined by the space object model; The spatial object model is constructed based on the regional attribute information and device information in the target space; The recommendation module is used to generate and push service functions adapted to the service requirements based on the service functions that the space object model can provide.

11. An electronic device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 9 are implemented.

12. A storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 9 are implemented.

13. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 9 are implemented.

Images