WO2023168853A1 - 使用意图的预测方法和装置、存储介质及电子装置 - Google Patents

使用意图的预测方法和装置、存储介质及电子装置 Download PDF

Info

Publication number
WO2023168853A1
WO2023168853A1 PCT/CN2022/099875 CN2022099875W WO2023168853A1 WO 2023168853 A1 WO2023168853 A1 WO 2023168853A1 CN 2022099875 W CN2022099875 W CN 2022099875W WO 2023168853 A1 WO2023168853 A1 WO 2023168853A1
Authority
WO
WIPO (PCT)
Prior art keywords
target
usage
intention
spatial information
predicting
Prior art date
Application number
PCT/CN2022/099875
Other languages
English (en)
French (fr)
Inventor
赵仕军
Original Assignee
青岛海尔科技有限公司
海尔智家股份有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 青岛海尔科技有限公司, 海尔智家股份有限公司 filed Critical 青岛海尔科技有限公司
Publication of WO2023168853A1 publication Critical patent/WO2023168853A1/zh

Links

Images

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/20Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
    • G06F16/24Querying
    • G06F16/245Query processing
    • G06F16/2458Special types of queries, e.g. statistical queries, fuzzy queries or distributed queries
    • G06F16/2462Approximate or statistical queries
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/20Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
    • G06F16/23Updating
    • G06F16/2379Updates performed during online database operations; commit processing
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/20Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
    • G06F16/24Querying
    • G06F16/245Query processing
    • G06F16/2458Special types of queries, e.g. statistical queries, fuzzy queries or distributed queries
    • G06F16/2474Sequence data queries, e.g. querying versioned data
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/20Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
    • G06F16/24Querying
    • G06F16/248Presentation of query results
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/20Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
    • G06F16/29Geographical information databases
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00Pattern recognition
    • G06F18/20Analysing
    • G06F18/22Matching criteria, e.g. proximity measures
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06QINFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00Administration; Management
    • G06Q10/04Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
    • GPHYSICS
    • G16INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16YINFORMATION AND COMMUNICATION TECHNOLOGY SPECIALLY ADAPTED FOR THE INTERNET OF THINGS [IoT]
    • G16Y10/00Economic sectors
    • G16Y10/80Homes; Buildings
    • GPHYSICS
    • G16INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16YINFORMATION AND COMMUNICATION TECHNOLOGY SPECIALLY ADAPTED FOR THE INTERNET OF THINGS [IoT]
    • G16Y20/00Information sensed or collected by the things
    • G16Y20/10Information sensed or collected by the things relating to the environment, e.g. temperature; relating to location
    • GPHYSICS
    • G16INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16YINFORMATION AND COMMUNICATION TECHNOLOGY SPECIALLY ADAPTED FOR THE INTERNET OF THINGS [IoT]
    • G16Y20/00Information sensed or collected by the things
    • G16Y20/20Information sensed or collected by the things relating to the thing itself
    • GPHYSICS
    • G16INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16YINFORMATION AND COMMUNICATION TECHNOLOGY SPECIALLY ADAPTED FOR THE INTERNET OF THINGS [IoT]
    • G16Y20/00Information sensed or collected by the things
    • G16Y20/40Information sensed or collected by the things relating to personal data, e.g. biometric data, records or preferences
    • YGENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02PCLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
    • Y02P90/00Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
    • Y02P90/02Total factory control, e.g. smart factories, flexible manufacturing systems [FMS] or integrated manufacturing systems [IMS]

Definitions

  • the present disclosure relates to the field of smart homes, and specifically to a method and device for predicting usage intention, a storage medium and an electronic device.
  • Embodiments of the present disclosure provide a method and device, a storage medium, and an electronic device for predicting intention of use, so as to at least solve the problems in related technologies such as the inability to predict intention of use in association with time and space, and the single prediction method.
  • a method for predicting usage intention including: obtaining first spatial information and second spatial information of a target device, wherein the first spatial information is used to indicate the geographical location of the target device, The second spatial information is used to indicate the characteristics of the space where the target device is located in the home area; determine the target usage environment of the target device based on the first spatial information and the second spatial information; determine the match with the target usage environment from the database of the Internet of Things cloud device operation events; predict the target object's intention to use the target device based on the device operation events.
  • a device for predicting usage intention including: an acquisition module configured to acquire first spatial information and second spatial information of a target device, wherein the first spatial information is used for Indicates the geographical location of the target device, and the second spatial information is used to indicate the characteristics of the space where the target device is located in the home area; the first determination module is configured to determine the location of the target device according to the first spatial information and the second spatial information. Determine the target usage environment of the target device; a second determination module configured to determine device operation events matching the target usage environment from the database of the Internet of Things cloud; a prediction module configured to predict targets based on the device operation events The object's intended use with the target device.
  • a storage medium is also provided, and a computer program is stored in the storage medium, wherein the computer program is configured to execute the steps in any of the above method embodiments when running.
  • an electronic device including a memory and a processor.
  • a computer program is stored in the memory, and the processor is configured to run the computer program to perform any of the above. Steps in method embodiments.
  • the first spatial information and the second spatial information of the target device are obtained, wherein the first spatial information is used to indicate the geographical location of the target device, and the second spatial information is used to indicate that the target device is in the home area. Characteristics of the space in which it is located; determine the target usage environment of the target device according to the first space information and the second space information; determine device operation events matching the target usage environment from the database of the Internet of Things cloud ; Predict the target object's intention to use the target device based on the device operation event.
  • the target usage environment of the target device is determined, and then the device operation event corresponding to the target usage environment is determined from the database, and the target device is used for the target object based on the device operation event.
  • the usage intention is determined, and the time corresponding to the device operation event is coupled and associated with the space corresponding to the target device, so that the predicted usage intention is more accurate. Therefore, it can solve the problem that the existing technology cannot correlate time and space to determine usage intention. Prediction, single prediction method and other problems, incorporate time and spatial correlation information into the prediction information, and improve the ability to predict the usage intention of the target object to use the target device, making the prediction process of usage intention more convenient, efficient and accurate.
  • Figure 1 is a schematic diagram of the hardware environment of a method for predicting usage intentions according to an embodiment of the present disclosure
  • Figure 2 is a flow chart of a method for predicting usage intention according to an embodiment of the present disclosure
  • Figure 3 is a schematic diagram of predicting usage intention through space and time according to an optional embodiment of the present disclosure
  • Figure 4 is a structural block diagram (1) of a device for predicting usage intention according to an embodiment of the present disclosure
  • FIG. 5 is a structural block diagram (2) of a device for predicting usage intention according to an embodiment of the present disclosure.
  • a prediction method of usage intention is provided.
  • This usage intention prediction method is widely used in whole-house intelligent digital control application scenarios such as smart home, smart home, smart home device ecology, and smart residence (Intelligence House) ecology.
  • the above prediction method of usage intention can be applied to the hardware environment composed of the terminal device 102 and the server 104 as shown in FIG. 1 .
  • the server 104 is connected to the terminal device 102 through the network and can be used to provide services (such as application services, etc.) for the terminal or the client installed on the terminal.
  • a database can be set up on the server or independently from the server.
  • cloud computing and/or edge computing services can be configured on the server or independently of the server to provide data computing services for the server 104.
  • the above-mentioned network may include but is not limited to at least one of the following: wired network, wireless network.
  • the above-mentioned wired network may include but is not limited to at least one of the following: wide area network, metropolitan area network, and the above-mentioned wireless network may include at least one of the following: WIFI (Wireless Fidelity, Wireless Fidelity), Bluetooth.
  • the terminal device 102 may be, but is not limited to, a PC, a mobile phone, a tablet, a smart air conditioner, a smart hood, a smart refrigerator, a smart oven, a smart stove, a smart washing machine, a smart water heater, a smart washing equipment, a smart dishwasher, or a smart projection device.
  • smart TV smart clothes drying rack, smart curtains, smart audio and video, smart sockets, smart audio, smart speakers, smart fresh air equipment, smart kitchen and bathroom equipment, smart bathroom equipment, smart sweeping robot, smart window cleaning robot, smart mopping robot, Smart air purification equipment, smart steamers, smart microwave ovens, smart kitchen appliances, smart purifiers, smart water dispensers, smart door locks, etc.
  • Figure 2 is a flow chart of a method for predicting intention to use according to an embodiment of the present disclosure. The process includes the following steps:
  • Step S202 Obtain first spatial information and second spatial information of the target device, where the first spatial information is used to indicate the geographical location of the target device, and the second spatial information is used to indicate that the target device is in the home area. characteristics of the space;
  • the above-mentioned first spatial information can be provided through the target object bound to the target device, or can be positioned through the GPS module or GPS function carried by the target device; the above-mentioned second spatial information can be combined with the room through the location of the device. Confirmation of spatial information, or confirmation of spatial information selected based on the target object. This disclosure does not limit this too much.
  • Step S204 determine the target usage environment of the target device according to the first spatial information and the second spatial information
  • Step S206 Determine device operation events that match the target usage environment from the database of the Internet of Things cloud;
  • the above-mentioned device operation events include the time, action and ancillary information of the device executing the operation event, the device, and the included target objects as the core elements of the event, and then the information is unified.
  • the information source of the above-mentioned device operation events Can be provided by IoT suites.
  • Step S208 Predict the target object's intention to use the target device based on the device operation event.
  • the first spatial information and the second spatial information of the target device are obtained, wherein the first spatial information is used to indicate the geographical location of the target device, and the second spatial information is used to indicate that the target device is in the home area. Characteristics of the space in which it is located; determine the target usage environment of the target device according to the first space information and the second space information; determine device operation events matching the target usage environment from the database of the Internet of Things cloud ; Predict the target object's intention to use the target device based on the device operation event.
  • the target usage environment of the target device is determined, and then the device operation event corresponding to the target usage environment is determined from the database, and the target device is used for the target object based on the device operation event.
  • the usage intention is determined, and the time corresponding to the device operation event is coupled and associated with the space corresponding to the target device, so that the predicted usage intention is more accurate. Therefore, it can solve the problem that the existing technology cannot correlate time and space to determine usage intention. Prediction, single prediction method and other problems, incorporate time and spatial correlation information into the prediction information, and improve the ability to predict the usage intention of the target object to use the target device, making the prediction process of usage intention more convenient, efficient and accurate.
  • determining the target usage environment of the target device based on the first spatial information and the second spatial information includes: inputting the first spatial information into a spatial prediction model to obtain the first environmental characteristics of the target device, Among them, the spatial prediction model is trained through machine learning using multiple sets of data. Each set of data in the multiple sets of data includes: the first spatial information of the target device, and the environmental characteristics corresponding to the first spatial information of the target device; according to The first environment characteristics and the second spatial information determine the target usage environment of the target device.
  • the spatial prediction part for the target device includes two parts of space, the geographical location space (equivalent to the first spatial information in the embodiment of the present disclosure), provinces and cities, etc.; and the space of the family room (equivalent to the first spatial information in the embodiment of the present disclosure).
  • Second space information in the example bedroom, living room, bathroom, kitchen, balcony, etc.;
  • a GNN graph convolutional network type model can be used for prediction of the spatial part.
  • information from other provinces and cities is also referenced to further improve the accuracy of predictions.
  • the accuracy of the results will be improved whether it is fine-grained prediction for counties or coarse-grained predictions at the provincial or municipal level.
  • the models used for the time series prediction part are not limited to RNN, CNN and other types of models (LSTM, etc.).
  • determining device operation events that match the target usage environment from the database of the Internet of Things cloud includes: determining the timing information of the target object using the target device; matching multiple events from the database of the Internet of Things cloud based on the timing information. sequential device operation events; determine the usage environment corresponding to each sequential device operation event among the multiple sequential device operation events, and obtain multiple usage environments; compare multiple usage environments with the target usage environment to determine the device that matches the target usage environment Operational events.
  • comparing multiple usage environments with a target usage environment to determine device operation events that match the target usage environment includes: determining a similarity between each usage environment in the multiple usage environments and the target usage environment; When the similarity meets the preset threshold, the current timing device operation event is determined to be a device operation event that matches the target usage environment.
  • the timing information of the target device using the target device is determined, that is, which time period the target object likes to be in Use the target device to match multiple sequential device operation events in the database of the Internet of Things cloud, and determine the usage environment corresponding to the sequential operation event.
  • the usage environment we can then operate from multiple sequential device operations.
  • the device operation events that match the target usage environment are determined from the events, and the timing and spatial information are integrated to predict the target object's intention to use the target device.
  • predicting the target object's intention to use the target device based on the device operation event includes: when there are multiple device operation events matching the target use environment, obtaining the operation preference of the target object, wherein , the operation preference is used to indicate the most frequent operation events performed by the target object using the target device; determine multiple matching degrees of multiple device operation events and operation preferences; arrange multiple device operation events according to the size of the multiple matching degrees, and select Use the most matching device operation events to predict the target object's intention to use the target device.
  • the above method further includes: obtaining a feedback result of the target object regarding the predicted intention to use; when the feedback result is that execution is allowed Next, it indicates that the currently predicted usage intention meets the usage requirements of the target object, and ends the prediction process of usage intention.
  • the above method further includes: associating the successfully predicted usage intention, the target object, the first spatial information, and the second spatial information. Binding; Upload the results of association binding to the database on the IoT cloud to generate the usage intention database of the target object.
  • a personalized prediction database belonging to the target object is generated by collecting successfully predicted usage intentions and corresponding target objects, spatial information, etc., through which the target object's needs can be directly determined based on this database. At the same time, the corresponding usage intention is quickly obtained, which improves the prediction efficiency of predicting usage intention.
  • an intention recognition method for smart home user habits based on spatiotemporal prediction is proposed.
  • time series prediction geographical location and home room space information are added, and graph neural network prediction is performed on it. , merging the prediction results of the two, effectively incorporating time and spatial correlation information into the prediction information to improve the ability of home users to use home appliances, thus improving the accuracy of prediction.
  • Figure 3 is a schematic diagram of predicting usage intention through space and time according to an optional embodiment of the present disclosure; specifically, by determining the events in which users use home appliances at each moment of the day, corresponding time series data is obtained; further acquisition space Linked data, by fusing time series data and spatial linked data, obtains spatiotemporal prediction results.
  • the above time series data and spatial correlation data can be shown in Table 1 below.
  • the data in Table 1 can be used to predict the water usage time, water consumption, and water temperature of the gas water heater the next day.
  • the location information is provided by the user or obtained from GPS; the room information is confirmed by the location information of the device, and the user chooses to fill in the home room; time, action and ancillary information, equipment, and users are unified as the core elements of the event by the Internet of Things. Kit provided.
  • the above spatiotemporal prediction usage intention can exist in the form of tool software, and the following conditions need to be ensured when the obtained data is used for prediction: reduce data noise, reduce the impact of missing information, and ensure the quality of the data; analyze trends and trends in time series.
  • Various dimensions such as cycles and bursts are included to ensure the applicability of the data; in the spatial dimension, it breaks the limitation of previous prediction models that can only predict at a single point, and can accurately predict and utilize the correlation impact in the spatial structure.
  • the models used in the time series prediction part are not limited to RNN, CNN and other types of models (LSTM, etc.); when performing spatial prediction through spatial correlation data, include Two parts of space, for example, (1) geographical location space, provinces and cities, etc.; (2) family room space, bedroom, living room, bathroom, kitchen, balcony, etc.
  • a GNN graph convolutional network type model can be used for prediction of the spatial part.
  • graph neural networks information from other provinces and cities is also referenced to further improve the accuracy of predictions. After the graph convolution network is introduced, the accuracy of the results will be improved whether it is fine-grained prediction for counties or coarse-grained predictions at the provincial or municipal level. Furthermore, it is necessary to further superimpose the temporal regularity information of adjacent stations for spatial information aggregation.
  • the habitual intention prediction method based on the fusion of geographical location information and time information, such as using the xgboost method, in which the geographical location and time information are only used as two-dimensional information of the table model, and each other There is no correlation between them.
  • the method based on spatio-temporal prediction in the above-mentioned optional embodiments of the present disclosure also adds the coupling and correlation of time and space to improve the prediction ability of user usage habits.
  • time series prediction geographical location and home room spatial information are added, and graph neural network prediction is performed on it, and the prediction results of the two are merged to improve the accuracy of prediction.
  • the method according to the above embodiments can be implemented by means of software plus the necessary general hardware platform. Of course, it can also be implemented by hardware, but in many cases the former is Better implementation.
  • the technical solution of the present disclosure can be embodied in the form of a software product in essence or that contributes to the existing technology.
  • the computer software product is stored in a storage medium (such as ROM/RAM, disk, CD), including several instructions to cause a terminal device (which can be a mobile phone, a computer, a server, or a network device, etc.) to perform the prediction of usage intentions described in various embodiments of the present disclosure.
  • module may be a combination of software and/or hardware that implements a predetermined function.
  • the apparatus described in the following embodiments is preferably implemented in software, implementation in hardware, or a combination of software and hardware, is also possible and contemplated.
  • Figure 4 is a structural block diagram (1) of a device for predicting usage intention according to an embodiment of the present disclosure. As shown in Figure 4, the device includes:
  • the acquisition module 42 is configured to acquire the first spatial information and the second spatial information of the target device, wherein the first spatial information is used to indicate the geographical location of the target device, and the second spatial information is used to indicate Characteristics of the space in which the target device is located in the home area;
  • the first determination module 44 is configured to determine the target usage environment of the target device according to the first spatial information and the second spatial information
  • the second determination module 46 is configured to determine device operation events that match the target usage environment from the database of the Internet of Things cloud;
  • the prediction module 48 is configured to predict the target object's intention to use the target device according to the device operation event.
  • the first spatial information and the second spatial information of the target device are obtained, wherein the first spatial information is used to indicate the geographical location of the target device, and the second spatial information is used to indicate that the target device is in the home area.
  • characteristics of the space in which it is located determine the target usage environment of the target device according to the first space information and the second space information; determine device operation events matching the target usage environment from the database of the Internet of Things cloud ; Predict the target object's intention to use the target device based on the device operation event.
  • the target usage environment of the target device is determined, and then the device operation event corresponding to the target usage environment is determined from the database, and the target device is used for the target object based on the device operation event.
  • the usage intention is determined, and the time corresponding to the device identification operation event and the space corresponding to the target device are coupled and associated, so that the predicted usage intention is more accurate. Therefore, it can solve the problem that the existing technology cannot correlate time and space to determine usage intention.
  • Prediction, single prediction method and other problems incorporate time and spatial correlation information into the prediction information, and improve the ability to predict the usage intention of the target object to use the target device, making the prediction process of usage intention more convenient, efficient and accurate.
  • the above-mentioned first determination module is also used to input the first spatial information into a spatial prediction model to obtain the first environmental characteristics of the target device, wherein the spatial prediction model uses multiple sets of data to pass Trained by machine learning, each set of data in the multiple sets of data includes: the first spatial information of the target device, and the environmental characteristics corresponding to the first spatial information of the target device; the target is determined based on the first environmental characteristics and the second spatial information.
  • the intended use environment of the device is also used to input the first spatial information into a spatial prediction model to obtain the first environmental characteristics of the target device, wherein the spatial prediction model uses multiple sets of data to pass Trained by machine learning, each set of data in the multiple sets of data includes: the first spatial information of the target device, and the environmental characteristics corresponding to the first spatial information of the target device; the target is determined based on the first environmental characteristics and the second spatial information.
  • the intended use environment of the device is also used to input the first spatial information into a spatial prediction model to obtain the first environmental characteristics of the target device, wherein
  • the spatial prediction part for the target device includes two parts of space, the geographical location space (equivalent to the first spatial information in the embodiment of the present disclosure), provinces and cities, etc.; and the space of the family room (equivalent to the first spatial information in the embodiment of the present disclosure).
  • Second space information in the example bedroom, living room, bathroom, kitchen, balcony, etc.;
  • a GNN graph convolutional network type model can be used for prediction of the spatial part.
  • information from other provinces and cities is also referenced to further improve the accuracy of predictions.
  • the accuracy of the results will be improved whether it is fine-grained prediction for counties or coarse-grained predictions at the provincial or municipal level.
  • the models used for the time series prediction part are not limited to RNN, CNN and other types of models (LSTM, etc.).
  • the above-mentioned second determination module is also used to determine the timing information of the target device used by the target object; match multiple timing device operation events from the database of the Internet of Things cloud based on the timing information; determine multiple timing devices
  • the usage environment corresponding to each sequential device operation event in the operation event is used to obtain multiple usage environments; the multiple usage environments are compared with the target usage environment to determine the device operation events that match the target usage environment.
  • the above-mentioned second determination module is also used to determine the similarity between each usage environment in the multiple usage environments and the target usage environment; when the similarity meets the preset threshold, determine the current timing device Operation events are device operation events that match the target usage environment.
  • the timing information of the target device using the target device is determined, that is, which time period the target object likes to be in Use the target device to match multiple sequential device operation events in the database of the Internet of Things cloud, and determine the usage environment corresponding to the sequential operation event.
  • the usage environment we can then operate from multiple sequential device operations.
  • the device operation events that match the target usage environment are determined from the events, and the timing and spatial information are integrated to predict the target object's intention to use the target device.
  • the above prediction module is also used to obtain the operation preference of the target object when there are multiple device operation events matching the target usage environment, wherein the operation preference is used to indicate the target object to use the target The most frequent operation events of the device; determine the multiple matching degrees of multiple device operation events and operation preferences; arrange multiple device operation events according to the size of the multiple matching degrees, and select the device operation event with the largest matching degree to predict the target The object's intended use with the target device.
  • FIG. 5 is a structural block diagram (2) of a device for predicting intention to use according to an embodiment of the present disclosure.
  • the above device may also include: a feedback module 50 and an association module 52 .
  • the above device further includes: a feedback module, configured to obtain the target object's feedback result for the predicted usage intention; when the feedback result is that execution is allowed, it indicates that the currently predicted usage intention satisfies the goal.
  • the usage requirements of the object end the prediction process of usage intention.
  • the above device further includes: an association module, configured to associate and bind successfully predicted usage intentions, target objects, first spatial information, and second spatial information; and upload the results of the association binding to The database in the IoT cloud generates the usage intention database of the target object.
  • an association module configured to associate and bind successfully predicted usage intentions, target objects, first spatial information, and second spatial information; and upload the results of the association binding to The database in the IoT cloud generates the usage intention database of the target object.
  • a personalized prediction database belonging to the target object is generated by collecting successfully predicted usage intentions and corresponding target objects, spatial information, etc., through which the target object's needs can be directly determined based on this database. At the same time, the corresponding usage intention is quickly obtained, which improves the prediction efficiency of predicting usage intention.
  • orientation or positional relationship indicated by the terms “center”, “upper”, “lower”, “front”, “back”, “left”, “right”, etc. is based on The orientation or positional relationship shown in the drawings is only to facilitate the description of the present disclosure and simplify the description, and does not indicate or imply that the device or component referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as Limitations on the Disclosure.
  • first and second are used for descriptive purposes only and are not to be understood as indicating or implying relative importance.
  • connection should be understood in a broad sense.
  • it can be a fixed connection or a detachable connection. , or integrally connected; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium; it can be an internal connection between two components.
  • a component is referred to as being “fixed” or “mounted to” another component, it can be directly on the other component or intervening components may also be present.
  • a component is said to be “connected” to another component, it may be directly connected to the other component or there may also be an intervening component present.
  • the specific meanings of the above terms in this disclosure can be understood on a case-by-case basis.
  • each of the above modules can be implemented through software or hardware.
  • it can be implemented in the following ways, but is not limited to this: the above modules are all located in the same processor; or the above modules can be implemented in any combination.
  • the forms are located in different processors.
  • Embodiments of the present disclosure also provide a storage medium in which a computer program is stored, wherein the computer program is configured to execute the steps in any of the above method embodiments when running.
  • the above-mentioned storage medium may be configured to store a computer program for performing the following steps:
  • S1 obtain the first spatial information and the second spatial information of the target device, where the first spatial information is used to indicate the geographical location of the target device, and the second spatial information is used to indicate the location of the target device in the home area. characteristics of the space;
  • S2 determine the target usage environment of the target device according to the first spatial information and the second spatial information
  • S4 Predict the target object's intention to use the target device according to the device operation event.
  • the above-mentioned storage medium may include but is not limited to: U disk, read-only memory (Read-Only Memory, referred to as ROM), random access memory (Random Access Memory, referred to as Various media that can store computer programs such as RAM), removable hard drives, magnetic disks or optical disks.
  • ROM read-only memory
  • RAM random access memory
  • removable hard drives magnetic disks or optical disks.
  • Embodiments of the present disclosure also provide an electronic device, including a memory and a processor.
  • a computer program is stored in the memory, and the processor is configured to run the computer program to perform the steps in any of the above method embodiments.
  • the above-mentioned electronic device may further include a transmission device and an input-output device, wherein the transmission device is connected to the above-mentioned processor, and the input-output device is connected to the above-mentioned processor.
  • the above-mentioned processor may be configured to perform the following steps through a computer program:
  • S1 obtain the first spatial information and the second spatial information of the target device, where the first spatial information is used to indicate the geographical location of the target device, and the second spatial information is used to indicate the location of the target device in the home area. characteristics of the space;
  • S2 determine the target usage environment of the target device according to the first spatial information and the second spatial information
  • S4 Predict the target object's intention to use the target device according to the device operation event.
  • modules or steps of the present disclosure can be implemented using general-purpose computing devices, and they can be concentrated on a single computing device, or distributed across a network composed of multiple computing devices.
  • they may be implemented in program code executable by a computing device, such that they may be stored in a storage device for execution by the computing device, and in some cases may be implemented in a format different from
  • the steps shown or described here are performed sequentially, or are implemented as separate integrated circuit modules, or multiple modules or steps among them are implemented as a single integrated circuit module.
  • the present disclosure is not limited to any specific combination of hardware and software.

Abstract

提供了一种使用意图的预测方法和装置、存储介质及电子装置,涉及智慧家庭技术领域,使用意图的预测方法包括:获取目标设备的第一空间信息以及第二空间信息,其中,第一空间信息用于指示目标设备所在地的地理位置,第二空间信息用于指示目标设备在家庭区域中所处空间的特征(S202);根据第一空间信息和第二空间信息确定目标设备的目标使用环境(S204);从物联网云端的数据库中确定与目标使用环境匹配的设备操作事件(S206);根据设备操作事件预测目标对象对于与目标设备的使用意图(S208)。

Description

使用意图的预测方法和装置、存储介质及电子装置
本公开要求于2022年03月10日提交中国专利局、申请号为202210240965.2、发明名称“使用意图的预测方法和装置、存储介质及电子装置”的中国专利申请的优先权,其全部内容通过引用结合在本公开中。
技术领域
本公开涉及智慧家庭领域,具体而言,涉及一种使用意图的预测方法和装置、存储介质及电子装置。
背景技术
随着人工智能和互联技术的不断发展,智能应用越来越多的下沉到各种场景和设备,比如可以自动驾驶的智能汽车,可以智能服务的智能家居。其中,针对用户的使用习惯进行用户使用设备意图的预测又是各种场景中重要的交互方式。
但是,相关技术中,针对用户使用家电习惯的方法基本处于时间序列的预测,而众所周知的情况是家庭用户针对使用家电的习惯的数据稀疏度较高,单纯靠时间序列的预测很难有较明显的意图能力的提升。
针对相关技术中,无法关联时间和空间进行使用意图的预测,预测方式单一等问题,尚未提出有效的技术方案。
发明内容
本公开实施例提供了一种使用意图的预测方法和装置、存储介质、电子装置,以至少解决相关技术中,无法关联时间和空间进行使用意图的预测,预测方式单一等问题。
根据本公开的一个实施例,提供了一种使用意图的预测方法,包括:获取目标设备的第一空间信息以及第二空间信息,其中,第一空间信息用于指示目标设备所在地的地理位置,第二空间信息用于指示目标设备在家庭区域中所处空间的 特征;根据第一空间信息和第二空间信息确定目标设备的目标使用环境;从物联网云端的数据库中确定与目标使用环境匹配的设备操作事件;根据设备操作事件预测目标对象对于与目标设备的使用意图。
根据本公开的另一个实施例,提供了一种使用意图的预测装置,包括:获取模块,设置为获取目标设备的第一空间信息以及第二空间信息,其中,所述第一空间信息用于指示目标设备所在地的地理位置,所述第二空间信息用于指示目标设备在家庭区域中所处空间的特征;第一确定模块,设置为根据所述第一空间信息和所述第二空间信息确定所述目标设备的目标使用环境;第二确定模块,设置为从物联网云端的数据库中确定与所述目标使用环境匹配的设备操作事件;预测模块,用于根据所述设备操作事件预测目标对象对于与所述目标设备的使用意图。
根据本公开的又一个实施例,还提供了一种存储介质,所述存储介质中存储有计算机程序,其中,所述计算机程序被设置为运行时执行上述任一项方法实施例中的步骤。
根据本公开的又一个实施例,还提供了一种电子装置,包括存储器和处理器,所述存储器中存储有计算机程序,所述处理器被设置为运行所述计算机程序以执行上述任一项方法实施例中的步骤。
通过本公开,获取目标设备的第一空间信息以及第二空间信息,其中,所述第一空间信息用于指示目标设备所在地的地理位置,所述第二空间信息用于指示目标设备在家庭区域中所处空间的特征;根据所述第一空间信息和所述第二空间信息确定所述目标设备的目标使用环境;从物联网云端的数据库中确定与所述目标使用环境匹配的设备操作事件;根据所述设备操作事件预测目标对象对于与所述目标设备的使用意图。也就是说,通过确定目标设备的多种空间信息,对目标设备的目标使用环境进行确定,继而从数据库中确定与目标使用环境对应的设备操作事件,基于该设备操作事件对目标对象使用目标设备的使用意图进行确定,将设别操作事件对应的时间和目标设备对应的空间进行耦合、关联,使得预测出的使用意图更加准确,因此,可以解决现有技术中无法关联时间和空间进行使用意图的预测,预测方式单一等问题,将时间和空间相关性信息纳入预测信息中,而提升预测目标对象使用目标设备的使用意图的能力,使得使用意图的预测流程 更加便捷、高效、准确。
附图说明
此处的附图被并入说明书中并构成本说明书的一部分,示出了符合本公开的实施例,并与说明书一起用于解释本公开的原理。
为了更清楚地说明本公开实施例或现有技术中的技术方案,下面将对实施例或现有技术描述中所需要使用的附图作简单地介绍,显而易见地,对于本领域普通技术人员而言,在不付出创造性劳动性的前提下,还可以根据这些附图获得其他的附图。
图1是根据本公开实施例的一种使用意图的预测方法的硬件环境示意图;
图2是根据本公开实施例的使用意图的预测方法的流程图;
图3是根据本公开可选实施例的通过时空预测使用意图的示意图;
图4是根据本公开实施例的使用意图的预测装置的结构框图(一);
图5是根据本公开实施例的使用意图的预测装置的结构框图(二)。
具体实施方式
为了使本技术领域的人员更好地理解本公开方案,下面将结合本公开实施例中的附图,对本公开实施例中的技术方案进行清楚、完整地描述,显然,所描述的实施例仅仅是本公开一部分的实施例,而不是全部的实施例。基于本公开中的实施例,本领域普通技术人员在没有做出创造性劳动前提下所获得的所有其他实施例,都应当属于本公开保护的范围。
需要说明的是,本公开的说明书和权利要求书及上述附图中的术语“第一”、“第二”等是用于区别类似的对象,而不必用于描述特定的顺序或先后次序。应该理解这样使用的数据在适当情况下可以互换,以便这里描述的本公开的实施例能够以除了在这里图示或描述的那些以外的顺序实施。此外,术语“包括”和“具有”以及他们的任何变形,意图在于覆盖不排他的包含,例如,包含了一系列步骤或单元的过程、方法、系统、产品或设备不必限于清楚地列出的那些步骤 或单元,而是可包括没有清楚地列出的或对于这些过程、方法、产品或设备固有的其它步骤或单元。
根据本公开实施例的一个方面,提供了一种使用意图的预测方法。该使用意图的预测方法广泛应用于智慧家庭(Smart Home)、智能家居、智能家用设备生态、智慧住宅(Intelligence House)生态等全屋智能数字化控制应用场景。可选地,在本实施例中,上述使用意图的预测方法可以应用于如图1所示的由终端设备102和服务器104所构成的硬件环境中。如图1所示,服务器104通过网络与终端设备102进行连接,可用于为终端或终端上安装的客户端提供服务(如应用服务等),可在服务器上或独立于服务器设置数据库,用于为服务器104提供数据存储服务,可在服务器上或独立于服务器配置云计算和/或边缘计算服务,用于为服务器104提供数据运算服务。
上述网络可以包括但不限于以下至少之一:有线网络,无线网络。上述有线网络可以包括但不限于以下至少之一:广域网,城域网,,上述无线网络可以包括但不限于以下至少之一:WIFI(Wireless Fidelity,无线保真),蓝牙。终端设备102可以并不限定于为PC、手机、平板电脑、智能空调、智能烟机、智能冰箱、智能烤箱、智能炉灶、智能洗衣机、智能热水器、智能洗涤设备、智能洗碗机、智能投影设备、智能电视、智能晾衣架、智能窗帘、智能影音、智能插座、智能音响、智能音箱、智能新风设备、智能厨卫设备、智能卫浴设备、智能扫地机器人、智能擦窗机器人、智能拖地机器人、智能空气净化设备、智能蒸箱、智能微波炉、智能厨宝、智能净化器、智能饮水机、智能门锁等。
在本实施例中提供了一种使用意图的预测方法,图2是根据本公开实施例的使用意图的预测方法的流程图,该流程包括如下步骤:
步骤S202,获取目标设备的第一空间信息以及第二空间信息,其中,所述第一空间信息用于指示目标设备所在地的地理位置,所述第二空间信息用于指示目标设备在家庭区域中所处空间的特征;
可选的,上述第一空间信息可以是通过与目标设备绑定的目标对象提供,也可以是通过目标设备携带的GPS模块或者GPS功能进行定位;上述第二空间信息 可以通过设备所在位置结合房间空间信息确认,或者根据目标对象选填的空间信息确认,对此,本公开不做过多限定。
步骤S204,根据所述第一空间信息和所述第二空间信息确定所述目标设备的目标使用环境;
步骤S206,从物联网云端的数据库中确定与所述目标使用环境匹配的设备操作事件;
需要说明的是,上述设备操作事件包括设备执行操作事件的时间、动作以及附属信息、设备、包含的目标对象作为事件核心要素,继而进行信息的统一,可选的,上述设备操作事件的信息来源可以由物联网的套件提供。
步骤S208,根据所述设备操作事件预测目标对象对于与所述目标设备的使用意图。
通过上述步骤,获取目标设备的第一空间信息以及第二空间信息,其中,所述第一空间信息用于指示目标设备所在地的地理位置,所述第二空间信息用于指示目标设备在家庭区域中所处空间的特征;根据所述第一空间信息和所述第二空间信息确定所述目标设备的目标使用环境;从物联网云端的数据库中确定与所述目标使用环境匹配的设备操作事件;根据所述设备操作事件预测目标对象对于与所述目标设备的使用意图。也就是说,通过确定目标设备的多种空间信息,对目标设备的目标使用环境进行确定,继而从数据库中确定与目标使用环境对应的设备操作事件,基于该设备操作事件对目标对象使用目标设备的使用意图进行确定,将设别操作事件对应的时间和目标设备对应的空间进行耦合、关联,使得预测出的使用意图更加准确,因此,可以解决现有技术中无法关联时间和空间进行使用意图的预测,预测方式单一等问题,将时间和空间相关性信息纳入预测信息中,而提升预测目标对象使用目标设备的使用意图的能力,使得使用意图的预测流程更加便捷、高效、准确。
在一个示例性实施例中,根据第一空间信息和第二空间信息确定目标设备的目标使用环境,包括:将第一空间信息输入到空间预测模型中,以得到目标设备的第一环境特征,其中,空间预测模型为使用多组数据通过机器学习训练出的, 多组数据中的每组数据均包括:目标设备的第一空间信息,以及目标设备的第一空间信息对应的环境特征;根据第一环境特征和第二空间信息确定目标设备的目标使用环境。
例如,对于目标设备的空间预测的部分包含两部分的空间,地理位置的空间(相当于本公开实施例中的第一空间信息),省市等部分;家庭房间的空间(相当于本公开实施例中的第二空间信息),卧室、客厅、洗漱间、厨房、阳台等;
可选的,针对空间部分的预测可以采用GNN图卷积网络类型模型。利用图神经网络,也将其他省市的信息参考进来,进一步提升预测的精度。引入图卷积网络后,无论是针对县区的细粒度预测,还是省市级别的粗粒度预测,结果的准确性都会提升。
可选的,对于时序预测的部分使用的模型不限于RNN、CNN等类型的模型(LSTM等)。
在一个示例性实施例中,从物联网云端的数据库中确定与目标使用环境匹配的设备操作事件,包括:确定目标对象使用目标设备的时序信息;基于时序信息从物联网云端的数据库匹配出多个时序设备操作事件;确定多个时序设备操作事件中每一个时序设备操作事件对应的使用环境,得到多个使用环境;比较多个使用环境与目标使用环境,以确定与目标使用环境匹配的设备操作事件。
在一个示例性实施例中,比较多个使用环境与目标使用环境,以确定与目标使用环境匹配的设备操作事件,包括:确定多个使用环境中每一个使用环境与目标使用环境的相似度;在相似度符合预设阈值的情况下,确定当前时序设备操作事件为与目标使用环境匹配的设备操作事件。
可以理解的是,在确定与目标使用环境对应的设备操作事件时,为了更好的贴合目标对象的使用习惯,通过确定目标对象使用目标设备的时序信息,即目标对象喜欢在哪一个时间段进行目标设备的使用,从而在物联网云端的数据库匹配出多个时序设备操作事件,并对时序操作事件对应的使用环境进行确定,通过比较使用环境与目标使用环境,进而从多个时序设备操作事件中确定出与目标使用环境匹配的设备操作事件,综合时序和空间信息来对目标对象使用目标设备的使用意图进行预测。
在一个示例性实施例中,根据设备操作事件预测目标对象对于与目标设备的使用意图,包括:在与目标使用环境匹配的设备操作事件存在多个的情况下,获取目标对象的操作偏好,其中,操作偏好用于指示目标对象使用目标设备进行次数最多的操作事件;确定多个设备操作事件与操作偏好的多个匹配度;依据多个匹配度的大小对多个设备操作事件进行排列,选择匹配度最大的设备操作事件来预测目标对象对于与目标设备的使用意图。
简单来说,为了提升对使用意图预测的准确性,排除其他多余选项对预测结果的干扰,还可以通过确定目标对象的操作偏好,进而通过确定多个设备操作事件与目标对象对应操作偏好的匹配度,继而通过匹配度对多个设备操作事件进行排列,优先使用匹配度最大的设备操作事件来进行目标对象的使用意图的预测。
在一个示例性实施例中,根据设备操作事件预测目标对象对于与目标设备的使用意图之后,上述方法还包括:获取目标对象对于预测出的使用意图的反馈结果;在反馈结果为允许执行的情况下,表明当前预测出的使用意图满足目标对象的使用要求,结束使用意图的预测流程。
也就是说,在预测出目标对象的使用意图之后,还可以通过获取目标对象对预测出的使用意图的反馈结果,确定本次使用意图预测的准确性,即通过关联目标对象的反馈结果对预测出的使用意图进行校验,保证预测出的使用意图与目标对象的使用习惯相互贴合。
在一个示例性实施例中,根据设备操作事件预测目标对象对于与目标设备的使用意图之后,上述方法还包括:将预测成功的使用意图、目标对象、第一空间信息、第二空间信息进行关联绑定;将关联绑定的结果上传至物联网云端的数据库,生成目标对象的使用意图数据库。
即为了使得对于目标对象使用目标设备的预测更加快速,通过收集成功预测使用意图以及对应的目标对象、空间信息等,生成属于目标对象的个性化预测数据库,通过该数据库可以直接在目标对象需求相同时,快速得到对应的使用意图,提升了预测使用意图的预测效率。
为了更好的理解上述使用意图的预测方法的过程,以下结合几个可选实施例对上述使用意图的预测方法流程进行说明。
作为一种可选的实施例,提出了一种基于时空预测针对智慧家庭用户习惯进行的意图识别方法,在时序预测的基础上增加地理位置和家庭房间空间信息,并对其进行图神经网络预测,融合两者的预测结果,有效的将时间和空间相关性信息纳入预测信息中而提升家庭用户使用家电的使用意图的能力,从而提升预测的准确率。提升智慧家庭中用户使用智慧家电设备使用意图的预测的能力。
可选的,图3是根据本公开可选实施例的通过时空预测使用意图的示意图;具体的,通过确定用户每日每个时刻使用家电的事件,得到对应的时间序列数据;进一步的获取空间关联数据,通过融合时间序列数据和空间关联数据,继而得到时空预测结果。
可选的,上述时间序列数据和空间关联数据可以如下表1所示,例如,可以通过表1中的数据预测第二天燃气热水器用水时间、用水量、用水温度预测。
表1、燃热零冷水使用预测
序号 日期 时间 位置 房间 设备 动作 水量
1 2020-01-01 08:20 青岛 厨房 燃气热水器 用水 100
2 2020-01-20 09:21 青岛 厨房 燃气热水器 用水 150
3 2020-01-25 14:20 青岛 厨房 燃气热水器 用水 200
4 2020-01-28 20:25 青岛 厨房 燃气热水器 用水 100
5 2020-02-01 16:36 青岛 厨房 燃气热水器 用水 150
6 2020-03-01 21:21 青岛 厨房 燃气热水器 用水 200
7 2020-04-10 22:00 青岛 厨房 燃气热水器 用水 300
8 2020-04-10 06:34 青岛 客厅 空调 开机  
可选的,位置信息通过用户提供或者gps获取;房间信息有设备所在位置信息确认,及用户选填在家庭房间;时间、动作及附属信息、设备、用户作为事件核心要素来统一由物联网的套件提供。
可选的,上述时空预测使用意图可以通过工具软件的方式存在,并且获取的数据进行预测时需要保证以下条件:降低数据噪音,减小信息缺失的影响,保证数据的质量;时序上对趋势、周期、突发等各类维度进行包容,保证数据的适用性;在空间维度上打破以往预测模型只能单点预测的局限性,能够在空间结构中准确预测并利用关联影响。
作为一种可选的实施方式,在通过时间序列数据进行时序预测时,时序预测部分使用的模型不限于RNN、CNN等类型的模型(LSTM等);在通过空间关联数据进行空间预测时,包含两部分的空间,例如,(1)地理位置的空间,省市等部分;(2)家庭房间的空间,卧室、客厅、洗漱间、厨房、阳台等。针对空间部分的预测可以采用GNN图卷积网络类型模型。利用图神经网络,也将其他省市的信息参考进来,进一步提升预测的精度。引入图卷积网络后,无论是针对县区的细粒度预测,还是省市级别的粗粒度预测,结果的准确性都会提升。进一步的,则需要进一步叠加相邻站点的时序规律信息进行空间上的信息聚合。
例如,A用户家庭与其相邻家庭之间往往存在这样的关系——当相邻家庭家电使用增加时。在这种情况下,当在时序上预测出该家庭次日用水,同时又看到空间层上相邻家庭次日用水预计会增加时,就可以预估出该家庭次日的设备用水或其他的行为可能将准确率增加,这样就将站点空间上的关联关系也融入到了模型中。对于关联性越大的节点,在预测时就越要优先考虑他们的关联关系。
综上,通过上述预测方式,避免相关基于将地理位置信息和时间信息融合进行的习惯意图预测的方法,比如使用xgboost方法,其中,地理位置和时间信息仅作为表格模型的两维信息,相互之间并未有相互之间的关联。而本公开上述可选实施例中基于时空预测的方法将时间和空间的耦合性、相关性也加入进来,以用来提升用户使用习惯的预测能力。在时序预测的基础上增加地理位置和家庭房间空间信息,并对其进行图神经网络预测,融合两者的预测结果,从而提升预测的准确率。
通过以上的实施方式的描述,本领域的技术人员可以清楚地了解到根据上述实施例的方法可借助软件加必需的通用硬件平台的方式来实现,当然也可以通过 硬件,但很多情况下前者是更佳的实施方式。基于这样的理解,本公开的技术方案本质上或者说对现有技术做出贡献的部分可以以软件产品的形式体现出来,该计算机软件产品存储在一个存储介质(如ROM/RAM、磁碟、光盘)中,包括若干指令用以使得一台终端设备(可以是手机,计算机,服务器,或者网络设备等)执行本公开各个实施例所述使用意图的预测。
在本实施例中还提供了一种使用意图的预测装置,该装置用于实现上述实施例及优选实施方式,已经进行过说明的不再赘述。如以下所使用的,术语“模块”可以实现预定功能的软件和/或硬件的组合。尽管以下实施例所描述的装置较佳地以软件来实现,但是硬件,或者软件和硬件的组合的实现也是可能并被构想的。
图4是根据本公开实施例的使用意图的预测装置的结构框图(一),如图4所示,该装置包括:
(1)获取模块42,设置为获取目标设备的第一空间信息以及第二空间信息,其中,所述第一空间信息用于指示目标设备所在地的地理位置,所述第二空间信息用于指示目标设备在家庭区域中所处空间的特征;
(2)第一确定模块44,设置为根据所述第一空间信息和所述第二空间信息确定所述目标设备的目标使用环境;
(3)第二确定模块46,设置为从物联网云端的数据库中确定与所述目标使用环境匹配的设备操作事件;
(4)预测模块48,设置为根据所述设备操作事件预测目标对象对于与所述目标设备的使用意图。
通过上述装置,获取目标设备的第一空间信息以及第二空间信息,其中,所述第一空间信息用于指示目标设备所在地的地理位置,所述第二空间信息用于指示目标设备在家庭区域中所处空间的特征;根据所述第一空间信息和所述第二空间信息确定所述目标设备的目标使用环境;从物联网云端的数据库中确定与所述目标使用环境匹配的设备操作事件;根据所述设备操作事件预测目标对象对于与所述目标设备的使用意图。也就是说,通过确定目标设备的多种空间信息,对目标设备的目标使用环境进行确定,继而从数据库中确定与目标使用环境对应的设备操作事件,基于该设备操作事件对目标对象使用目标设备的使用意图进行确定, 将设别操作事件对应的时间和目标设备对应的空间进行耦合、关联,使得预测出的使用意图更加准确,因此,可以解决现有技术中无法关联时间和空间进行使用意图的预测,预测方式单一等问题,将时间和空间相关性信息纳入预测信息中,而提升预测目标对象使用目标设备的使用意图的能力,使得使用意图的预测流程更加便捷、高效、准确。
在一个示例性实施例中,上述第一确定模块,还用于将第一空间信息输入到空间预测模型中,以得到目标设备的第一环境特征,其中,空间预测模型为使用多组数据通过机器学习训练出的,多组数据中的每组数据均包括:目标设备的第一空间信息,以及目标设备的第一空间信息对应的环境特征;根据第一环境特征和第二空间信息确定目标设备的目标使用环境。
例如,对于目标设备的空间预测的部分包含两部分的空间,地理位置的空间(相当于本公开实施例中的第一空间信息),省市等部分;家庭房间的空间(相当于本公开实施例中的第二空间信息),卧室、客厅、洗漱间、厨房、阳台等;
可选的,针对空间部分的预测可以采用GNN图卷积网络类型模型。利用图神经网络,也将其他省市的信息参考进来,进一步提升预测的精度。引入图卷积网络后,无论是针对县区的细粒度预测,还是省市级别的粗粒度预测,结果的准确性都会提升。
可选的,对于时序预测的部分使用的模型不限于RNN、CNN等类型的模型(LSTM等)。
在一个示例性实施例中,上述第二确定模块,还用于确定目标对象使用目标设备的时序信息;基于时序信息从物联网云端的数据库匹配出多个时序设备操作事件;确定多个时序设备操作事件中每一个时序设备操作事件对应的使用环境,得到多个使用环境;比较多个使用环境与目标使用环境,以确定与目标使用环境匹配的设备操作事件。
在一个示例性实施例中,上述第二确定模块,还用于确定多个使用环境中每一个使用环境与目标使用环境的相似度;在相似度符合预设阈值的情况下,确定当前时序设备操作事件为与目标使用环境匹配的设备操作事件。
可以理解的是,在确定与目标使用环境对应的设备操作事件时,为了更好的 贴合目标对象的使用习惯,通过确定目标对象使用目标设备的时序信息,即目标对象喜欢在哪一个时间段进行目标设备的使用,从而在物联网云端的数据库匹配出多个时序设备操作事件,并对时序操作事件对应的使用环境进行确定,通过比较使用环境与目标使用环境,进而从多个时序设备操作事件中确定出与目标使用环境匹配的设备操作事件,综合时序和空间信息来对目标对象使用目标设备的使用意图进行预测。
在一个示例性实施例中,上述预测模块,还用于在与目标使用环境匹配的设备操作事件存在多个的情况下,获取目标对象的操作偏好,其中,操作偏好用于指示目标对象使用目标设备进行次数最多的操作事件;确定多个设备操作事件与操作偏好的多个匹配度;依据多个匹配度的大小对多个设备操作事件进行排列,选择匹配度最大的设备操作事件来预测目标对象对于与目标设备的使用意图。
简单来说,为了提升对使用意图预测的准确性,排除其他多余选项对预测结果的干扰,还可以通过确定目标对象的操作偏好,进而通过确定多个设备操作事件与目标对象对应操作偏好的匹配度,继而通过匹配度对多个设备操作事件进行排列,优先使用匹配度最大的设备操作事件来进行目标对象的使用意图的预测。
可选的,图5是根据本公开实施例的使用意图的预测装置的结构框图(二),上述装置还可以包括:反馈模块50,关联模块52。
在一个示例性实施例中,上述装置还包括:反馈模块,用于获取目标对象对于预测出的使用意图的反馈结果;在反馈结果为允许执行的情况下,表明当前预测出的使用意图满足目标对象的使用要求,结束使用意图的预测流程。
也就是说,在预测出目标对象的使用意图之后,还可以通过获取目标对象对预测出的使用意图的反馈结果,确定本次使用意图预测的准确性,即通过关联目标对象的反馈结果对预测出的使用意图进行校验,保证预测出的使用意图与目标对象的使用习惯相互贴合。
在一个示例性实施例中,上述装置还包括:关联模块,用于将预测成功的使用意图、目标对象、第一空间信息、第二空间信息进行关联绑定;将关联绑定的结果上传至物联网云端的数据库,生成目标对象的使用意图数据库。
即为了使得对于目标对象使用目标设备的预测更加快速,通过收集成功预测 使用意图以及对应的目标对象、空间信息等,生成属于目标对象的个性化预测数据库,通过该数据库可以直接在目标对象需求相同时,快速得到对应的使用意图,提升了预测使用意图的预测效率。
在本公开的描述中,需要理解的是,术语中“中心”、“上”、“下”、“前”、“后”、“左”、“右”等指示的方位或位置关系为基于附图所示的方位或位置关系,仅是为了便于描述本公开和简化描述,而不是指示或暗示所指的装置或组件必须具有特定的方位、以特定的方位构造和操作,因此不能理解为对本公开的限制。此外,术语“第一”、“第二”仅用于描述目的,而不能理解为指示或暗示相对重要性。
在本公开的描述中,需要说明的是,除非另有明确的规定和限定,术语“安装”、“连接”、“相连”应做广义理解,例如,可以是固定连接,也可以是拆卸连接,或一体地连接;可以是机械连接,也可以是电连接;可以是直接相连,也可以是通过中间媒介间接相连,可以是两个组件内部的连通。当组件被称为“固定于”或“设置于”另一个元件,它可以直接在另一个组件上或者也可以存在居中的组件。当一个组件被认为是“连接”另一个元件,它可以是直接连接到另一个元件或者可能同时存在居中元件。对于本领域的普通技术人员而言,可以具体情况理解上述术语在本公开的具体含义。
需要说明的是,上述各个模块是可以通过软件或硬件来实现的,对于后者,可以通过以下方式实现,但不限于此:上述模块均位于同一处理器中;或者,上述各个模块以任意组合的形式分别位于不同的处理器中。
本公开的实施例还提供了一种存储介质,该存储介质中存储有计算机程序,其中,该计算机程序被设置为运行时执行上述任一项方法实施例中的步骤。
在一个示例性实施例中,在本实施例中,上述存储介质可以被设置为存储用于执行以下步骤的计算机程序:
S1,获取目标设备的第一空间信息以及第二空间信息,其中,所述第一空间信息用于指示目标设备所在地的地理位置,所述第二空间信息用于指示目标设备在家庭区域中所处空间的特征;
S2,根据所述第一空间信息和所述第二空间信息确定所述目标设备的目标使用环境;
S3,从物联网云端的数据库中确定与所述目标使用环境匹配的设备操作事件;
S4,根据所述设备操作事件预测目标对象对于与所述目标设备的使用意图。
在一个示例性实施例中,在本实施例中,上述存储介质可以包括但不限于:U盘、只读存储器(Read-Only Memory,简称为ROM)、随机存取存储器(Random Access Memory,简称为RAM)、移动硬盘、磁碟或者光盘等各种可以存储计算机程序的介质。
本公开的实施例还提供了一种电子装置,包括存储器和处理器,该存储器中存储有计算机程序,该处理器被设置为运行计算机程序以执行上述任一项方法实施例中的步骤。
在一个示例性实施例中,上述电子装置还可以包括传输设备以及输入输出设备,其中,该传输设备和上述处理器连接,该输入输出设备和上述处理器连接。
在一个示例性实施例中,在本实施例中,上述处理器可以被设置为通过计算机程序执行以下步骤:
S1,获取目标设备的第一空间信息以及第二空间信息,其中,所述第一空间信息用于指示目标设备所在地的地理位置,所述第二空间信息用于指示目标设备在家庭区域中所处空间的特征;
S2,根据所述第一空间信息和所述第二空间信息确定所述目标设备的目标使用环境;
S3,从物联网云端的数据库中确定与所述目标使用环境匹配的设备操作事件;
S4,根据所述设备操作事件预测目标对象对于与所述目标设备的使用意图。
在一个示例性实施例中,本实施例中的具体示例可以参考上述实施例及可选实施方式中所描述的示例,本实施例在此不再赘述。
显然,本领域的技术人员应该明白,上述的本公开的各模块或各步骤可以用通用的计算装置来实现,它们可以集中在单个的计算装置上,或者分布在多个计算装置所组成的网络上,在一个示例性实施例中,它们可以用计算装置可执行的程序代码来实现,从而,可以将它们存储在存储装置中由计算装置来执行,并且在某些情况下,可以以不同于此处的顺序执行所示出或描述的步骤,或者将它们 分别制作成各个集成电路模块,或者将它们中的多个模块或步骤制作成单个集成电路模块来实现。这样,本公开不限制于任何特定的硬件和软件结合。
以上所述仅是本公开的优选实施方式,应当指出,对于本技术领域的普通技术人员来说,在不脱离本公开原理的前提下,还可以做出若干改进和润饰,这些改进和润饰也应视为本公开的保护范围。

Claims (16)

  1. 一种使用意图的预测方法,包括:
    获取目标设备的第一空间信息以及第二空间信息,其中,所述第一空间信息用于指示目标设备所在地的地理位置,所述第二空间信息用于指示目标设备在家庭区域中所处空间的特征;
    根据所述第一空间信息和所述第二空间信息确定所述目标设备的目标使用环境;
    从物联网云端的数据库中确定与所述目标使用环境匹配的设备操作事件;
    根据所述设备操作事件预测目标对象对于与所述目标设备的使用意图。
  2. 根据权利要求1所述使用意图的预测方法,其中,根据所述第一空间信息和所述第二空间信息确定所述目标设备的目标使用环境,包括:
    将所述第一空间信息输入到空间预测模型中,以得到所述目标设备的第一环境特征,其中,所述空间预测模型为使用多组数据通过机器学习训练出的,所述多组数据中的每组数据均包括:目标设备的第一空间信息,以及目标设备的第一空间信息对应的环境特征;
    根据所述第一环境特征和所述第二空间信息确定所述目标设备的目标使用环境。
  3. 根据权利要求1所述使用意图的预测方法,其中,从物联网云端的数据库中确定与所述目标使用环境匹配的设备操作事件,包括:
    确定目标对象使用所述目标设备的时序信息;
    基于所述时序信息从所述物联网云端的数据库匹配出多个时序设备操作事件;
    确定所述多个时序设备操作事件中每一个时序设备操作事件对应的使用环境,得到多个使用环境;
    比较所述多个使用环境与所述目标使用环境,以确定与所述目标使用环境匹配的设备操作事件。
  4. 根据权利要求3所述使用意图的预测方法,其中,比较所述多个使用环境与所述目标使用环境,以确定与所述目标使用环境匹配的设备操作事件,包括:
    确定所述多个使用环境中每一个使用环境与所述目标使用环境的相似度;
    在所述相似度符合预设阈值的情况下,确定当前时序设备操作事件为与所述目标使用环境匹配的设备操作事件。
  5. 根据权利要求1所述使用意图的预测方法,其中,根据所述设备操作事件预测目标对象对于与所述目标设备的使用意图,包括:
    在与所述目标使用环境匹配的设备操作事件存在多个的情况下,获取目标对象的操作偏好,其中,所述操作偏好用于指示目标对象使用目标设备进行次数最多的操作事件;
    确定多个所述设备操作事件与所述操作偏好的多个匹配度;
    依据所述多个匹配度的大小对多个所述设备操作事件进行排列,选择匹配度最大的所述设备操作事件来预测目标对象对于与所述目标设备的使用意图。
  6. 根据权利要求1所述使用意图的预测方法,其中,根据所述设备操作事件预测目标对象对于与所述目标设备的使用意图之后,所述方法还包括:
    获取所述目标对象对于预测出的使用意图的反馈结果;
    在所述反馈结果为允许执行的情况下,表明当前预测出的使用意图满足所述目标对象的使用要求,结束所述使用意图的预测流程。
  7. 根据权利要求1所述使用意图的预测方法,其中,根据所述设备操作事件预测目标对象对于与所述目标设备的使用意图之后,所述方法还包括:
    将预测成功的使用意图、所述目标对象、所述第一空间信息、所述第二空间信息进行关联绑定;
    将所述关联绑定的结果上传至物联网云端的数据库,生成所述目标对象的使用意图数据库。
  8. 一种使用意图的预测装置,包括:
    获取模块,设置为获取目标设备的第一空间信息以及第二空间信息,其中,所述第一空间信息用于指示目标设备所在地的地理位置,所述第二空间信息用于指示目标设备在家庭区域中所处空间的特征;
    第一确定模块,设置为根据所述第一空间信息和所述第二空间信息确定所述目标设备的目标使用环境;
    第二确定模块,设置为从物联网云端的数据库中确定与所述目标使用环境匹配的设备操作事件;
    预测模块,设置为根据所述设备操作事件预测目标对象对于与所述目标设备的使用意图。
  9. 根据权利要求8所述使用意图的预测装置,其中,所述第一确定模块,还用于将第一空间信息输入到空间预测模型中,以得到目标设备的第一环境特征,其中,空间预测模型为使用多组数据通过机器学习训练出的,多组数据中的每组数据均包括:目标设备的第一空间信息,以及目标设备的第一空间信息对应的环境特征;根据第一环境特征和第二空间信息确定目标设备的目标使用环境。
  10. 根据权利要求8所述使用意图的预测装置,其中,所述第二确定模块,还用于确定目标对象使用目标设备的时序信息;基于时序信息从物联网云端的数据库匹配出多个时序设备操作事件;确定多个时序设备操作事件中每一个时序设备操作事件对应的使用环境,得到多个使用环境;比较多个使用环境与目标使用环境,以确定与目标使用环境匹配的设备操作事件。
  11. 根据权利要求8所述使用意图的预测装置,其中,所述第二确定模块,还用于确定多个使用环境中每一个使用环境与目标使用环境的相似度;在相似度符合预设阈值的情况下,确定当前时序设备操作事件为与目标使用环境匹配的设备操作事件。
  12. 根据权利要求8所述使用意图的预测装置,其中,所述预测模块,还用于在与目标使用环境匹配的设备操作事件存在多个的情况下,获取目标对象的操作偏好,其中,操作偏好用于指示目标对象使用目标设备进行次数最多的操作事件;确定多个设备操作事件与操作偏好的多个匹配度;依据多个匹配度的大小对多个设备操作事件进行排列,选择匹配度最大的设备操作事件来预测目标对象对于与目标设备的使用意图。
  13. 根据权利要求8所述使用意图的预测装置,其中,所述使用意图的预测装置还包括:反馈模块,用于获取目标对象对于预测出的使用意图的反馈结果;在反馈结果为允许执行的情况下,表明当前预测出的使用意图满足目标对象的使用要求,结束使用意图的预测流程。
  14. 根据权利要求8所述使用意图的预测装置,其中,所述使用意图的预测装置还包括:关联模块,用于将预测成功的使用意图、目标对象、第一空间信息、第二空间信息进行关联绑定;将关联绑定的结果上传至物联网云端的数据库,生成目标对象的使用意图数据库。
  15. 一种计算机可读的存储介质,其中,所述计算机可读的存储介质包括存储的程序,其中,所述程序运行时执行权利要求1至7中任一项所述的方法。
  16. 一种电子装置,包括存储器和处理器,其中,所述存储器中存储有计算机程序,所述处理器被设置为通过所述计算机程序执行权利要求1至7中任一项所述的方法。
PCT/CN2022/099875 2022-03-10 2022-06-20 使用意图的预测方法和装置、存储介质及电子装置 WO2023168853A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202210240965.2 2022-03-10
CN202210240965.2A CN114691752A (zh) 2022-03-10 2022-03-10 使用意图的预测方法和装置、存储介质及电子装置

Publications (1)

Publication Number Publication Date
WO2023168853A1 true WO2023168853A1 (zh) 2023-09-14

Family

ID=82140067

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2022/099875 WO2023168853A1 (zh) 2022-03-10 2022-06-20 使用意图的预测方法和装置、存储介质及电子装置

Country Status (2)

Country Link
CN (1) CN114691752A (zh)
WO (1) WO2023168853A1 (zh)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115299832A (zh) * 2022-08-05 2022-11-08 佛山市百斯特电器科技有限公司 一种开门控制方法、装置、洗涤设备及存储介质

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104614998A (zh) * 2014-12-19 2015-05-13 小米科技有限责任公司 控制家居设备的方法和装置
US20200035235A1 (en) * 2018-07-27 2020-01-30 Lg Electronics Inc. Home appliance and speech recognition server system using artificial intelligence and method for controlling thereof
CN111562748A (zh) * 2020-04-12 2020-08-21 李杰勇 一种智能家居环境控制系统及控制方法
CN112054947A (zh) * 2020-08-31 2020-12-08 海信(山东)空调有限公司 控制室内环境电器的方法、室内环境电器以及遥控终端
CN113934152A (zh) * 2021-11-10 2022-01-14 珠海格力电器股份有限公司 一种设备控制方法和装置、电子设备和存储介质

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104614998A (zh) * 2014-12-19 2015-05-13 小米科技有限责任公司 控制家居设备的方法和装置
US20200035235A1 (en) * 2018-07-27 2020-01-30 Lg Electronics Inc. Home appliance and speech recognition server system using artificial intelligence and method for controlling thereof
CN111562748A (zh) * 2020-04-12 2020-08-21 李杰勇 一种智能家居环境控制系统及控制方法
CN112054947A (zh) * 2020-08-31 2020-12-08 海信(山东)空调有限公司 控制室内环境电器的方法、室内环境电器以及遥控终端
CN113934152A (zh) * 2021-11-10 2022-01-14 珠海格力电器股份有限公司 一种设备控制方法和装置、电子设备和存储介质

Also Published As

Publication number Publication date
CN114691752A (zh) 2022-07-01

Similar Documents

Publication Publication Date Title
US10584892B2 (en) Air-conditioning control method, air-conditioning control apparatus, and storage medium
CN109445848B (zh) 设备联动方法及装置
Aydinalp-Koksal et al. Comparison of neural network, conditional demand analysis, and engineering approaches for modeling end-use energy consumption in the residential sector
CN104166396B (zh) 一种智能家居控制方法及系统
WO2023168862A1 (zh) 控制指令的预测方法和装置、存储介质及电子装置
CA2968009A1 (en) Generation of browser suggestions based on internet of things device data
WO2023168853A1 (zh) 使用意图的预测方法和装置、存储介质及电子装置
Karlin et al. Characterization and potential of home energy management (HEM) Technology
CN115167164A (zh) 设备场景的确定方法和装置、存储介质及电子装置
JP5691918B2 (ja) 在不在継続時間推定方法、在不在継続時間推定プログラム及び在不在継続時間推定装置
Zhou et al. Privacy-preserving and residential context-aware online learning for IoT-enabled energy saving with big data support in smart home environment
WO2024021546A1 (zh) 行为偏好表的生成方法和装置、存储介质及电子装置
WO2024001189A1 (zh) 食物存储信息的确定方法及装置、存储介质及电子装置
WO2024001196A1 (zh) 家电设备的控制方法、装置、存储介质及电子装置
US20140236324A1 (en) Apparatus and method for controlling terminal based on living pattern
WO2023207170A1 (zh) 洗涤程序的推荐方法及装置、存储介质及电子装置
WO2024001113A1 (zh) 图谱的确定方法和装置、存储介质及电子装置
WO2023165051A1 (zh) 身份确定方法、存储介质及电子装置
CN116540556A (zh) 基于用户习惯的设备控制方法及装置
CN115481315A (zh) 推荐信息的确定方法和装置、存储介质及电子装置
CN114459158B (zh) 一种家用供热控制方法、系统、电子设备及介质
CN114355782A (zh) 家居设备控制方法、装置、路由设备和存储介质
CN114115027A (zh) 目标环境参数的调节方法、系统、装置、设备及存储介质
CN113777948A (zh) 家电设备的控制方法及装置、非易失性存储介质
WO2024036804A1 (zh) 意图指令的确定方法及装置、存储介质及电子装置

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 22930482

Country of ref document: EP

Kind code of ref document: A1