WO2021027437A1 - 资源调度方法及系统、计算机可读存储介质 - Google Patents

资源调度方法及系统、计算机可读存储介质 Download PDF

Info

Publication number
WO2021027437A1
WO2021027437A1 PCT/CN2020/100339 CN2020100339W WO2021027437A1 WO 2021027437 A1 WO2021027437 A1 WO 2021027437A1 CN 2020100339 W CN2020100339 W CN 2020100339W WO 2021027437 A1 WO2021027437 A1 WO 2021027437A1
Authority
WO
WIPO (PCT)
Prior art keywords
function
knowledge graph
user
started
devices
Prior art date
Application number
PCT/CN2020/100339
Other languages
English (en)
French (fr)
Inventor
魏德平
贾巨涛
黄姿荣
胡志华
胡广绪
Original Assignee
珠海格力电器股份有限公司
珠海联云科技有限公司
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 珠海格力电器股份有限公司, 珠海联云科技有限公司 filed Critical 珠海格力电器股份有限公司
Priority to US17/623,677 priority Critical patent/US20220276897A1/en
Publication of WO2021027437A1 publication Critical patent/WO2021027437A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00—Arrangements for program control, e.g. control units
    • G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/46—Multiprogramming arrangements
    • G06F9/50—Allocation of resources, e.g. of the central processing unit [CPU]
    • G06F9/5005—Allocation of resources, e.g. of the central processing unit [CPU] to service a request
    • G06F9/5027—Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resource being a machine, e.g. CPUs, Servers, Terminals
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
    • G06F16/36—Creation of semantic tools, e.g. ontology or thesauri
    • G06F16/367—Ontology
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
    • G06F3/01—Input arrangements or combined input and output arrangements for interaction between user and computer
    • G06F3/011—Arrangements for interaction with the human body, e.g. for user immersion in virtual reality
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00—Arrangements for program control, e.g. control units
    • G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/46—Multiprogramming arrangements
    • G06F9/48—Program initiating; Program switching, e.g. by interrupt
    • G06F9/4806—Task transfer initiation or dispatching
    • G06F9/4843—Task transfer initiation or dispatching by program, e.g. task dispatcher, supervisor, operating system
    • G06F9/4881—Scheduling strategies for dispatcher, e.g. round robin, multi-level priority queues

Definitions

  • the present disclosure relates to the technical field of electric curtains, in particular to a resource scheduling method, and also to a resource scheduling system and a computer-readable storage medium.
  • the technical problem to be solved by the present disclosure is: the current resource scheduling lacks a personalized resource scheduling strategy.
  • the present disclosure provides a resource scheduling method and system, and computer-readable storage medium.
  • a resource scheduling method which includes:
  • the constructed knowledge graph is queried for candidate devices associated with the function.
  • the knowledge graph stores multiple functions and Equipment associated with each function
  • control instruction is a voice instruction
  • determining the intended device includes: performing semantic analysis on the voice instruction, and determining the intended device based on the semantic analysis result.
  • the device to be started is determined from all candidate devices, including:
  • the device to be started is determined according to the feedback information.
  • the knowledge graph also stores a connection weight between a function and a device, and the connection weight increases as the probability that the user selects the device when facing the function increases,
  • the candidate device corresponding to the largest connection weight is used as the device to be started.
  • the weight of the connection between the function and the device is reflected by the length of the path connecting the function and the device.
  • the greater the connection weight between the function and the device the shorter the path connecting the function and the device.
  • the method before acquiring the control instruction input by the user, the method further includes constructing the knowledge graph, which includes:
  • the function is connected to the device with the function through a path.
  • constructing the knowledge graph further includes:
  • connection weight between the function and each device having the function is determined, and the path length between the function and each device having the function is adjusted based on the connection weight.
  • a resource scheduling system which includes:
  • Instruction acquisition module configured to acquire control instructions input by the user
  • An intention determination module configured to determine an intention device based on the control instruction
  • the storage module is configured to store the constructed knowledge graph, and the knowledge graph stores multiple functions and devices associated with each function;
  • the query module is configured to, for each function of the intent device, query the candidate device associated with the function from the constructed knowledge graph when the intent device cannot be started;
  • the startup module is configured to determine and start the device to be started from all candidate devices.
  • a computer-readable storage medium in which a computer program is stored, and when the computer program is executed by a processor, the resource scheduling method as described above is implemented.
  • one or more embodiments of the above solutions may have the following advantages or beneficial effects:
  • the candidate device is reasonably determined according to the relationship between the function and the device in the constructed knowledge graph, and then the candidate device is determined and activated to be started equipment.
  • the present disclosure solves the problem that resources cannot be individually scheduled in related technologies, improves the degree of system individualization, meets the scheduling requirements of different resources, and improves user experience.
  • FIG. 1 shows a schematic flowchart of a resource scheduling method according to an embodiment of the present disclosure
  • Figure 2 shows a schematic diagram of a knowledge graph in an embodiment of the present disclosure
  • FIG. 3 shows a schematic diagram of a process of determining a device to be started from all candidate devices in an embodiment of the present disclosure
  • FIG. 4 shows another schematic flow chart of determining a device to be started from all candidate devices in an embodiment of the present disclosure
  • FIG. 5 shows a schematic diagram of a process of constructing a knowledge graph in an embodiment of the present disclosure
  • Fig. 6 shows a schematic structural diagram of a resource scheduling system according to an embodiment of the present disclosure.
  • embodiments of the present disclosure provide a resource scheduling method.
  • Fig. 1 shows a schematic flowchart of a resource scheduling method according to an embodiment of the present disclosure.
  • the resource scheduling method of the embodiment of the present disclosure mainly includes step S101 to step S106.
  • step S101 a control instruction input by the user is acquired.
  • the user can input control instructions through the remote control or directly through voice instructions.
  • step S102 the intended device is determined based on the control instruction.
  • control command input by the user is analyzed to confirm the device that the user wants to turn on (herein referred to as the intended device).
  • this step also includes performing semantic analysis on the voice instruction, and then determining the intended device based on the semantic analysis result.
  • semantic analysis methods commonly used in related technologies to perform semantic analysis on voice instructions to obtain the intended device, which is not described in this article.
  • step S103 it is determined whether the intended device can be started.
  • step S104 if the intended device can be activated, the intended device is normally activated.
  • step S105 in the case that the intended device cannot be started, for each function of the intended device, a candidate device associated with the function is queried from the constructed knowledge graph.
  • a candidate device associated with the function is queried from the constructed knowledge graph.
  • the resource scheduling system determines the candidate device of the intended device based on the pre-built knowledge graph.
  • FIG. 2 shows a schematic diagram of a knowledge graph in an embodiment of the present disclosure.
  • the knowledge graph saves three functions and the devices associated with each function.
  • the cooling function, dehumidification function and humidification function are stored in the knowledge map.
  • the devices associated with the cooling function include fans, air conditioners, refrigerators, and air conditioning fans.
  • the equipment associated with dehumidification includes air conditioners and dehumidifiers.
  • the equipment associated with humidification includes air conditioning fans and humidifiers. It can be seen that the system establishes a knowledge graph for resources of different types and functions by classifying various resources. For example, the establishment of a knowledge map for home appliances, and various electrical appliances are directly or indirectly connected by attributes or functions.
  • the air conditioner and the fan also have cooling properties, then the air conditioner and the fan are connected with the cooling property, and the air conditioner and the dehumidifier have the dehumidification function, then the air conditioner and the dehumidifier are directly connected with the dehumidification property, and the fan is connected with the dehumidifier.
  • Dehumidifiers form an indirect connection.
  • the system determines that the candidate devices include: fans, refrigerators and air-conditioning fans associated with the cooling function of the air conditioner, and dehumidifiers associated with the dehumidification function of the air conditioner.
  • step S106 determine and start the device to be started from all the candidate devices.
  • the device to be started can be determined by further querying the user, or the device to be started can be determined by the connection weight saved in the knowledge graph. The description will be given below in conjunction with FIG. 3 and FIG. 4.
  • Fig. 3 shows a schematic flow chart of determining a device to be started from all candidate devices in an embodiment of the present disclosure. As shown in FIG. 3, determining the device to be started from all candidate devices includes step S201 to step S203.
  • step S201 the prompt information of all candidate devices recorded is shown to the user.
  • step S202 the user's feedback information for the prompt information is received, and the feedback information indicates the device selected by the user from all the candidate devices.
  • step S203 the device to be started is determined according to the feedback information.
  • the client displays and records the prompt information of all candidate devices, so that the customer can select one of these candidate devices as the device to be activated.
  • Fig. 4 shows a schematic flow chart of determining a device to be started from all candidate devices in an embodiment of the present disclosure.
  • the knowledge graph also stores the connection weight between the function and the device.
  • the connection weight increases with the increase in the probability that the user selects the device when facing the function.
  • the connection weight between the function and the device The size of is reflected by the length of the path connecting the function and the device. The greater the weight of the connection between the function and the device, the shorter the path connecting the function and the device.
  • determining the device to be started from all candidate devices includes step S301 and step S302.
  • step S301 the connection weights associated with each candidate device are sorted.
  • step S302 the candidate device corresponding to the largest connection weight is used as the device to be activated.
  • the device selected by the user frequently is selected from the candidate devices as the device to be activated.
  • a knowledge graph needs to be constructed offline in advance.
  • Fig. 5 shows a schematic diagram of a process of constructing a knowledge graph in an embodiment of the present disclosure.
  • the method of constructing a knowledge graph mainly includes steps S401 to S404.
  • step S401 a list of devices to be scheduled is determined, and each device in the device list is recorded in the knowledge graph.
  • step S402 the functions of all devices in the device list are summarized into a function list, and each function in the function list is recorded in the knowledge graph.
  • step S403 for each function, the function is connected to the device having the function through a path.
  • step S404 for each function, obtain the record of the device selected by the user when facing the function, determine the connection weight between the function and each device with the function according to the record, and adjust the function based on the connection weight The path length between each device with this function.
  • the list includes six equipment: fans, air conditioners, refrigerators, air conditioners, dehumidifiers, and humidifiers. Summarize the functions of these six devices to get a function list, which includes the cooling function, dehumidification function, and humidification function.
  • the related functions and devices are connected by paths, and the length of the path represents the connection weight.
  • the shorter the path the greater the connection weight, and the greater the probability that the user will select the device when facing this function.
  • the longer the path the smaller the connection weight, and the lower the probability that the user will select the device when facing the function, thus constructing a knowledge graph including functions, devices, and paths.
  • the candidate device is reasonably determined according to the relationship between the function and the device in the constructed knowledge graph, and then the candidate device is determined And start the device to be started.
  • the embodiments of the present disclosure solve the problem that resources cannot be individually scheduled in related technologies, improve the degree of system personalization, meet the scheduling requirements of different resources, and improve user experience.
  • the system builds a knowledge graph based on the relationship between the physical devices and the user's preference for the entity, and performs resource scheduling, so that the scheduling result is more likely to be the real choice of the user, which helps to further enhance the user experience.
  • air conditioners have the function of cooling and dehumidification
  • dehumidifiers also have dehumidification functions. Based on the dehumidification function, the air conditioners and dehumidifiers are connected with the attribute of dehumidification. .
  • the cloud server records the usage history of the dehumidifier, and sometimes the user also turns on the dehumidification function of the air conditioner, and the cloud server also records user usage data.
  • a weight value is established between the attribute and the entity, and the weight value increases according to the increase in the frequency of use of the device by the user. Simply put, it is the length between this connection (path). The larger the weight value, the shorter the path line. Through this relationship, a corresponding user knowledge graph is established.
  • the user records on the cloud server the user's preferences for resources required for each operation, such as user 1 common resource 1.
  • the system analyzes the user instruction and puts the user instruction into the NLP semantic parser, for example : I want to turn on the air conditioner.
  • the semantic parser analyzes the user's semantics and needs to find the air conditioner.
  • the air conditioner cannot be turned on or the network is cut off, the user cannot control the air conditioner well. That is, the user cannot operate the corresponding resource.
  • the system uses the established knowledge map, the attributes or functions of the resources, for example, the system uses the attributes of air-conditioning, such as cooling, cooling, and dehumidification, to find appliances with cooling functions in the user’s home equipment as a substitute.
  • air-conditioning such as cooling, cooling, and dehumidification
  • the system finds out the corresponding information through the knowledge graph and feeds back the corresponding information to the user, such as the air conditioner cannot be controlled, whether to turn on the fan, after the user chooses, the user instruction is understood through the NLP semantic parser again, and if yes, the fan is turned on. That is, when there is a problem with resource 1, the system searches for other devices with corresponding functions through the knowledge graph, such as resource 5, recommends it to the user, and performs resource scheduling after obtaining user feedback.
  • Fig. 6 shows a schematic structural diagram of a resource scheduling system according to an embodiment of the present disclosure.
  • the resource scheduling system of this embodiment mainly includes an instruction acquisition module, an intention determination module, a storage module, a query module, and a startup module.
  • the instruction acquisition module is configured to acquire the control instruction input by the user.
  • the intent determination module is configured to determine the intent device based on the control instruction.
  • the storage module is configured to store the constructed knowledge graph.
  • the knowledge graph stores multiple functions and devices associated with each function.
  • the query module is configured to query the candidate device associated with the function from the constructed knowledge graph for each function of the intent device when the intent device cannot be started.
  • the startup module is configured to determine and start the device to be started from all candidate devices.
  • This embodiment also provides a computer-readable storage medium in which a computer program is stored.
  • the computer program is executed by a processor, the resource scheduling method as in the first or second embodiment is implemented.
  • the candidate device is reasonably determined according to the relationship between the function and the device in the constructed knowledge graph, and the device to be started is determined and activated from the candidate device.
  • the embodiments of the present disclosure solve the problem that resources cannot be individually scheduled in related technologies, improve the degree of system personalization, meet the scheduling requirements of different resources, and improve user experience.
  • the system builds a knowledge graph based on the relationship between the physical devices and the user's preference for the entity, and performs resource scheduling, so that the scheduling result is more likely to be the real choice of the user, which helps to further enhance the user experience.

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Software Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Human Computer Interaction (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Animal Behavior & Ethology (AREA)
  • Computational Linguistics (AREA)
  • Data Mining & Analysis (AREA)
  • Databases & Information Systems (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)

Abstract

一种资源调度方法及系统、计算机可读存储介质，方法包括：获取用户输入的控制指令（S101）；基于所述控制指令，确定意图设备（S102）；在所述意图设备无法启动的情况下，针对所述意图设备的每个功能，从构建的知识图谱中查询与该功能相关联的备选设备，所述知识图谱中保存有多个功能以及与各个功能相关联的设备（S103，S105）；从所有备选设备中确定并启动待启动设备（S106）。所述资源调度方法，在用户指定的意图设备无法启动的情况下，根据构建好的知识图谱中功能与设备的关联关系来合理确定备选设备，进而从备选设备中确定与启动待启动设备，解决了相关技术中资源无法个性化调度的问题，提升系统个性化程度，满足不同资源的调度需求，提升用户体验感。

Description

资源调度方法及系统、计算机可读存储介质
本公开要求于2019年08月12日提交中国专利局、申请号为201910739694.3、发明名称为“资源调度方法及系统、计算机可读存储介质”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
本公开涉及电动窗帘技术领域，尤其涉及一种资源调度方法，还涉及一种资源调度系统以及计算机可读存储介质。
背景技术
随着互联网技术和产业的发展以及用户生活水平的提高，人们对系统提出的要求有了更高的水准。在大数据的环境背景下，资源在系统中的调度有了更多的个性化选择。
然而，目前的资源调度中，缺少个性化的资源调度策略，因此亟需一种资源调度的相应机制去满足这种需求，以提高用户的体验。
发明内容
本公开要解决的技术问题是：目前的资源调度中，缺少个性化的资源调度策略。为解决上述技术问题，本公开提供了一种资源调度方法及系统、计算机可读存储介质。
根据本公开的一个方面，提供了一种资源调度方法，其包括：
获取用户输入的控制指令；
基于所述控制指令，确定意图设备；
在所述意图设备无法启动的情况下，针对所述意图设备的每个功能，从构建的知识图谱中查询与该功能相关联的备选设备，所述知识图谱中保存有多个功能以及与各个功能相关联的设备；
从所有备选设备中确定并启动待启动设备。
在一些实施方式中，所述控制指令为语音指令，
基于所述控制指令，确定意图设备，包括：对所述语音指令进行语义解析，并基于语义解析结果确定所述意图设备。
在一些实施方式中，从所有备选设备中确定待启动设备，包括：
向用户展示记录有所有备选设备的提示信息；
接收用户针对所述提示信息的反馈信息，所述反馈信息指示用户从所有备选设备中选择的设备；
根据所述反馈信息确定所述待启动设备。
在一些实施方式中，所述知识图谱中还保存有功能与设备之间的连接权重，所述连接权重随着用户在面对该功能时选择该设备的概率的增加而增加，
从所有备选设备中确定待启动设备，包括：
将与各个备选设备相关联的连接权重进行排序；
将最大的连接权重对应的备选设备作为所述待启动设备。
在一些实施方式中，所述功能与设备之间的连接权重的大小通过连接该功能和设备的路径长度体现。
在一些实施方式中，所述功能与设备之间的连接权重越大，连接该功能与该设备的路径越短。
在一些实施方式中，在获取用户输入的控制指令之前，所述方法还包括构建所述知识图谱，其包括：
确定待调度的设备列表，并将所述设备列表中的每个设备记录在所述知识图谱中；
将所述设备列表中所有设备的功能汇总成功能列表，并将所述功能列表中的每个功能记录在所述知识图谱中；
针对每个功能，将该功能与具有该功能的设备之间通过路径连接。
在一些实施方式中，构建所述知识图谱，还包括：
针对每个功能，执行以下步骤：
获取用户在面对该功能时选择设备的记录；
根据所述记录，确定该功能与具有该功能的各个设备之间的连接权重并基于该连接权重调整该功能与具有该功能的各个设备之间的路径长度。
根据本公开的另一方面，提供了一种资源调度系统，其包括：
指令获取模块，配置为获取用户输入的控制指令；
意图确定模块，配置为基于所述控制指令，确定意图设备；
存储模块，配置为存储有构建的知识图谱，所述知识图谱中保存有多个功能以及与各个功能相关联的设备；
查询模块，配置为在所述意图设备无法启动的情况下，针对所述意图设备的每个功能，从构建的知识图谱中查询与该功能相关联的备选设备；
启动模块，配置为从所有备选设备中确定并启动待启动设备。
根据本公开的再一方面，提供了一种计算机可读存储介质，其中存储有计算机程序，所述计算机程序被处理器执行时实现如上述资源调度方法。
与相关技术相比，上述方案中的一个或多个实施例可以具有如下优点或有益效果：
应用本公开资源调度方法，在用户指定的意图设备无法启动的情况下，根据构建好的知识图谱中功能与设备的关联关系来合理确定备选设备，进而从备选设备中确定与启动待启动设备。本公开解决了相关技术中资源无法个性化调度的问题，提升系统个性化程度，满足不同资源的调度需求，提升用户体验感。
附图说明
通过结合附图阅读下文示例性实施例的详细描述可更好地理解本公开的范围。其中所包括的附图是：
图1示出了本公开实施例资源调度方法的流程示意图；
图2示出了本公开实施例中知识图谱的示意图；
图3示出了本公开实施例中从所有备选设备中确定待启动设备的一种流程示意图；
图4示出了本公开实施例中从所有备选设备中确定待启动设备的另一种流程示意图；
图5示出了本公开实施例中构建知识图谱的流程示意图；
图6示出了本公开实施例资源调度系统的结构示意图。
具体实施方式
为使本公开的目的、技术方案和优点更加清楚，以下将结合附图及实施例来 详细说明本公开的实施方法，借此对本公开如何应用技术手段来解决技术问题，并达成技术效果的实现过程能充分理解并据以实施。
目前的资源调度中，缺少个性化的资源调度策略。为解决上述技术问题，本公开实施例提供了一种资源调度方法。
实施例一
图1示出了本公开实施例资源调度方法的流程示意图。如图1所示，本公开实施例资源调度方法主要包括步骤S101至步骤S106。
在步骤S101中，获取用户输入的控制指令。
具体地，用户可以通过遥控器或直接通过语音指令输入控制指令。
在步骤S102中，基于控制指令，确定意图设备。
具体地，对用户输入的控制指令进行分析以确认用户欲开启的设备(这里称为意图设备)。
当控制指令为语音指令时，本步骤还包括对语音指令进行语义解析，然后基于语义解析结果确定意图设备。本领域人员可以采用相关技术常用的语义解析方法对语音指令进行语义解析，以得到意图设备，本文对此不展开说明。
在步骤S103中，判断意图设备是否能够启动。
在步骤S104中，在意图设备能够启动的情况下，正常启动意图设备。
在步骤S105中，在意图设备无法启动的情况下，针对意图设备的每个功能，从构建的知识图谱中查询与该功能相关联的备选设备。这里，知识图谱中保存有多个功能以及与各个功能相关联的设备。
具体地，当意图设备因为设备自身问题无法启动或者外部配置问题(例如断电断网)而无法启动的情况下，资源调度系统结合预先构建的知识图谱确定意图设备的备选设备。
图2示出了本公开实施例中知识图谱的示意图。如图2所示，知识图谱保存有三个功能以及与各个功能相关联的设备。其中，知识图谱中保存有降温功能、除湿功能和加湿功能。与降温功能相关联的设备有风扇、空调、冰箱和空调扇。与降湿相关联的设备有空调和除湿机。与加湿相关联的设备有空调扇和加湿器。可见，系统通过对各种资源进行分类，对不同类型和功能的资源建立知识图谱。例如对家电进行知识图谱的建立，各种电器之间以属性或功能进行直接或间接的连接。举个例子：空调与风扇同样具有降温的属性，那么空调与风扇就以降温属 性进行相连，而空调与除湿机都具有除湿功能，那么空调与除湿机就以除湿属性直接相连，而风扇则与除湿机形成了间接相连的关系。
在用户指示的意图设备空调不能启动的情况下，系统确定备选设备包括：与空调的降温功能相关联的风扇、冰箱和空调扇，以及与空调的除湿功能相关联的除湿机。
在步骤S106中，从所有备选设备中确定并启动待启动设备。
具体地，可以通过进一步询问用户来确定待启动设备，也可以通过知识图谱中保存的连接权重来确定待启动设备。下面结合图3和图4进行说明。
图3示出了本公开实施例中从所有备选设备中确定待启动设备的一种流程示意图。如图3所示，从所有备选设备中确定待启动设备，包括步骤S201至步骤S203。
在步骤S201中，向用户展示记录有所有备选设备的提示信息。
在步骤S202中，接收用户针对提示信息的反馈信息，反馈信息指示用户从所有备选设备中选择的设备。
在步骤S203中，根据反馈信息确定待启动设备。
在本实施例中，这里，通过客户端显示记录有所有备选设备的提示信息，以让客户从这些备选设备中选择一个作为待启动设备。
图4示出了本公开实施例中从所有备选设备中确定待启动设备的一种流程示意图。在本实施例中，知识图谱中还保存有功能与设备之间的连接权重，连接权重随着用户在面对该功能时选择该设备的概率的增加而增加，功能与设备之间的连接权重的大小通过连接该功能和设备的路径长度体现，功能与设备之间的连接权重越大，连接该功能与该设备的路径越短。
如图4所示，从所有备选设备中确定待启动设备，包括步骤S301和步骤S302。
在步骤S301中，将与各个备选设备相关联的连接权重进行排序。
在步骤S302中，将最大的连接权重对应的备选设备作为待启动设备。
在本实施例中，通过知识图谱中的连接权重，从备选设备中选择用户高频选择的设备作为待启动设备。
在确定好待启动设备后，启动该设备。
在本公开一优选的实施例中，在执行步骤S101之间，需要预先离线构建知识图谱。图5示出了本公开实施例中构建知识图谱的流程示意图。如图5所示， 构建知识图谱的方法主要包括步骤S401至步骤S404。
在步骤S401中，确定待调度的设备列表，并将设备列表中的每个设备记录在知识图谱中。
在步骤S402中，将设备列表中所有设备的功能汇总成功能列表，并将功能列表中的每个功能记录在知识图谱中。
在步骤S403中，针对每个功能，将该功能与具有该功能的设备之间通过路径连接。
在步骤S404中，针对每个功能，获取用户在面对该功能时选择设备的记录，根据记录，确定该功能与具有该功能的各个设备之间的连接权重，并基于该连接权重调整该功能与具有该功能的各个设备之间的路径长度。
以图2所示的知识图谱为例，首先确定待调度的设备列表，列表中包括风扇、空调、冰箱、空调扇、除湿机和加湿器这六个设备。将这六个设备的功能汇总可得到功能列表，功能列表中包括降温功能、除湿功能和加湿功能。
将设备列表中的设备以及功能列表中的功能都保存在知识图谱中。除此之外，将有关联的功能和设备用路径连接起来，路径的长短表征连接权重。路径越短表示连接权重越大，此时用户面对该功能时选择该设备的概率就越大。相反地，路径越长表示连接权重越小，此时用户面对该功能时选择该设备的概率就越小，由此就构建出了包括功能、设备和路径的知识图谱。
应用本实施例所述的资源调度方法，在用户指定的意图设备无法启动的情况下，根据构建好的知识图谱中功能与设备的关联关系来合理确定备选设备，进而从备选设备中确定与启动待启动设备。可见，本公开实施例解决了相关技术中资源无法个性化调度的问题，提升系统个性化程度，满足不同资源的调度需求，提升用户体验感。此外，系统根据实体设备间的相互关系以及用户对实体的喜好程度建立知识图谱，进行资源调度，使得调度结果更趋向于用户的真实选择，有助于进一步提升用户体验感。
实施例二
下面结合图2说明本公开实施例构建知识图谱以及基于知识图谱进行资源调度的方法。
首先，基于电器设备之间(也就是实体)的属性建立连接，例如空调具有制 冷、除湿的功能，除湿机也具有除湿功能，那么基于除湿的功能，使得空调和除湿机以除湿这个属性相连接。同时，通过用户日常对于设备的精准控制，比如打开除湿机，那么云服务器记录下除湿机的使用记录，有时用户也打开了空调的除湿功能，云服务器也同样记录了用户使用数据。那么基于该历史数据以及设备间的属性关系，在属性到实体间建立权重值，权重值根据用户对设备的使用频率的增加而增加。简单的说就是这个连线(路径)间的长度。权重值越大，路径线越短。通过这种关系，就建立了对应的用户知识图谱。
用户通过系统资源的使用，在云端服务器记录了用户的每次操作所需资源的喜好，例如用户1常用资源1，系统通过对用户指令的解析，将用户指令放入NLP语义解析器中，例如：我要打开空调，语义解析器通过解析用户语义，需找空调。但是当空调出现无法打开，断电断网的情况出现时，用户就不能很好的控制空调。即用户不能对对应资源进行操作。出现这种问题时，系统通过所建立的知识图谱，资源的属性或功能，例如系统利用空调的属性，如制冷、降温、除湿的功能，寻找用户家庭设备中具有降温功能的电器作为替代品，如风扇，那么系统通过知识图谱找到后向用户反馈对应的信息，如空调无法控制，是否开启风扇，待用户选择后再次通过NLP语义解析器对用户指令进行理解，如果说是，则打开风扇。即当资源1出现问题时，系统通过知识图谱查找对应功能的其他设备，如资源5，向用户推荐，得到用户反馈后进行资源的调度。
实施例三
本实施例提供了一种资源调度系统。图6示出了本公开实施例资源调度系统的结构示意图。如图6所示，本实施例的资源调度系统主要包括指令获取模块、意图确定模块、存储模块、查询模块和启动模块。
具体地，指令获取模块，配置为获取用户输入的控制指令。意图确定模块，配置为基于控制指令，确定意图设备。存储模块，配置为存储有构建的知识图谱，知识图谱中保存有多个功能以及与各个功能相关联的设备。查询模块，配置为在意图设备无法启动的情况下，针对意图设备的每个功能，从构建的知识图谱中查询与该功能相关联的备选设备。启动模块，配置为从所有备选设备中确定并启动待启动设备。
本实施例还提供了一种计算机可读存储介质，其中存储有计算机程序，计算 机程序被处理器执行时实现如实施例一或实施例二的资源调度方法。
应用本实施例，在用户指定的意图设备无法启动的情况下，根据构建好的知识图谱中功能与设备的关联关系来合理确定备选设备，进而从备选设备中确定与启动待启动设备。可见，本公开实施例解决了相关技术中资源无法个性化调度的问题，提升系统个性化程度，满足不同资源的调度需求，提升用户体验感。此外，系统根据实体设备间的相互关系以及用户对实体的喜好程度建立知识图谱，进行资源调度，使得调度结果更趋向于用户的真实选择，有助于进一步提升用户体验感。
虽然本公开所公开的实施方式如上，但所述的内容只是为了便于理解本公开而采用的实施方式，并非用以限定本公开。任何本公开所属技术领域内的技术人员，在不脱离本公开所公开的精神和范围的前提下，可以在实施的形式上及细节上作任何的修改与变化，但本公开的保护范围，仍须以所附的权利要求书所界定的范围为准。

Claims (10)

  1. 一种资源调度方法，包括：
    获取用户输入的控制指令；
    基于所述控制指令，确定意图设备；
    在所述意图设备无法启动的情况下，针对所述意图设备的每个功能，从构建的知识图谱中查询与该功能相关联的备选设备，所述知识图谱中保存有多个功能以及与各个功能相关联的设备；
    从所有备选设备中确定并启动待启动设备。
  2. 根据权利要求1所述的方法，其中，所述控制指令为语音指令，
    基于所述控制指令，确定意图设备，包括：对所述语音指令进行语义解析，并基于语义解析结果确定所述意图设备。
  3. 根据权利要求1所述的方法，其中，从所有备选设备中确定待启动设备，包括：
    向用户展示记录有所有备选设备的提示信息；
    接收用户针对所述提示信息的反馈信息，所述反馈信息指示用户从所有备选设备中选择的设备；
    根据所述反馈信息确定所述待启动设备。
  4. 根据权利要求1所述的方法，其中，所述知识图谱中还保存有功能与设备之间的连接权重，所述连接权重随着用户在面对该功能时选择该设备的概率的增加而增加，
    从所有备选设备中确定待启动设备，包括：
    将与各个备选设备相关联的连接权重进行排序；
    将最大的连接权重对应的备选设备作为所述待启动设备。
  5. 根据权利要求4所述的方法，其中，所述功能与设备之间的连接权重的大小通过连接该功能和设备的路径长度体现。
  6. 根据权利要求5所述的方法，其中，所述功能与设备之间的连接权重越大，连接该功能与该设备的路径越短。
  7. 根据权利要求1至6中任一项所述的方法，其中，在获取用户输入的控制指令之前，所述方法还包括构建所述知识图谱，其包括：
    确定待调度的设备列表，并将所述设备列表中的每个设备记录在所述知识图谱中；
    将所述设备列表中所有设备的功能汇总成功能列表，并将所述功能列表中的每个功能记录在所述知识图谱中；
    针对每个功能，将该功能与具有该功能的设备之间通过路径连接。
  8. 根据权利要求7所述的方法，其中，构建所述知识图谱，还包括：
    针对每个功能，执行以下步骤：
    获取用户在面对该功能时选择设备的记录；
    根据所述记录，确定该功能与具有该功能的各个设备之间的连接权重并基于该连接权重调整该功能与具有该功能的各个设备之间的路径长度。
  9. 一种资源调度系统，包括：
    指令获取模块，配置为获取用户输入的控制指令；
    意图确定模块，配置为基于所述控制指令，确定意图设备；
    存储模块，配置为存储有构建的知识图谱，所述知识图谱中保存有多个功能以及与各个功能相关联的设备；
    查询模块，配置为在所述意图设备无法启动的情况下，针对所述意图设备的每个功能，从构建的知识图谱中查询与该功能相关联的备选设备；
    启动模块，配置为从所有备选设备中确定并启动待启动设备。
  10. 一种计算机可读存储介质，其中存储有计算机程序，所述计算机程序被处理器执行时实现如权利要求1至8中任一项所述的资源调度方法。
PCT/CN2020/100339 2019-08-12 2020-07-06 资源调度方法及系统、计算机可读存储介质 WO2021027437A1 (zh)

Priority Applications (1)

Application Number Priority Date Filing Date Title
US17/623,677 US20220276897A1 (en) 2019-08-12 2020-07-06 Resource Scheduling Method and System, and Computer-Readable Storage Medium

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201910739694.3A CN110457489A (zh) 2019-08-12 2019-08-12 资源调度方法及系统、计算机可读存储介质
CN201910739694.3 2019-08-12

Publications (1)

Publication Number Publication Date
WO2021027437A1 true WO2021027437A1 (zh) 2021-02-18

Family

ID=68486006

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2020/100339 WO2021027437A1 (zh) 2019-08-12 2020-07-06 资源调度方法及系统、计算机可读存储介质

Country Status (3)

Country Link
US (1) US20220276897A1 (zh)
CN (1) CN110457489A (zh)
WO (1) WO2021027437A1 (zh)

Families Citing this family (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110457489A (zh) * 2019-08-12 2019-11-15 珠海格力电器股份有限公司 资源调度方法及系统、计算机可读存储介质
CN111104504A (zh) * 2019-12-25 2020-05-05 天津中科智能识别产业技术研究院有限公司 一种基于自然语言处理和知识图谱的对话方法
CN113032576B (zh) * 2019-12-25 2022-06-03 杭州海康威视数字技术股份有限公司 知识图谱管理方法、装置、设备及存储介质
CN111459037B (zh) * 2020-02-27 2021-09-14 珠海格力电器股份有限公司 智能家居系统控制方法、装置、电子设备及可读存储介质

Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105611023A (zh) * 2015-12-17 2016-05-25 北京奇虎科技有限公司 一种备选电话号码推送方法和装置
US20160189028A1 (en) * 2014-12-31 2016-06-30 Verizon Patent And Licensing Inc. Systems and Methods of Using a Knowledge Graph to Provide a Media Content Recommendation
CN105864976A (zh) * 2016-04-18 2016-08-17 广东美的环境电器制造有限公司 家电设备的联动控制方法和系统、服务器
CN106713083A (zh) * 2016-11-24 2017-05-24 海信集团有限公司 基于知识图谱的智能家居设备控制方法、装置及系统
CN207802669U (zh) * 2018-02-08 2018-08-31 深圳市艾特网能技术有限公司 一种智能一体化机柜
CN109754804A (zh) * 2019-02-21 2019-05-14 珠海格力电器股份有限公司 一种语音控制方法、装置、存储介质及智能家居系统
CN109995620A (zh) * 2017-12-30 2019-07-09 成都标制文化创意有限公司 一种控制智能家居设备的方法和智能家居设备
CN110457489A (zh) * 2019-08-12 2019-11-15 珠海格力电器股份有限公司 资源调度方法及系统、计算机可读存储介质

Family Cites Families (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6226627B1 (en) * 1998-04-17 2001-05-01 Fuji Xerox Co., Ltd. Method and system for constructing adaptive and resilient software
JP2017024340A (ja) * 2015-07-24 2017-02-02 株式会社リコー 課金システム、課金方法及び課金プログラム
CN106125567B (zh) * 2016-08-30 2019-11-08 广东美的制冷设备有限公司 一种基于用户指令识别家电的方法、系统和装置
CN107346115B (zh) * 2017-06-29 2020-12-08 浙江大华技术股份有限公司 一种智能设备的控制方法、控制终端和智能设备
US10536294B2 (en) * 2017-07-17 2020-01-14 Midea America Corp. Computer-based platform for quality management of home devices
CN108923969A (zh) * 2018-06-29 2018-11-30 海尔优家智能科技（北京）有限公司 一种设备联动的方法、装置、设备及计算机可读存储介质
US11068960B2 (en) * 2019-05-29 2021-07-20 Walmart Apollo, Llc Methods and apparatus for item substitution

Patent Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20160189028A1 (en) * 2014-12-31 2016-06-30 Verizon Patent And Licensing Inc. Systems and Methods of Using a Knowledge Graph to Provide a Media Content Recommendation
CN105611023A (zh) * 2015-12-17 2016-05-25 北京奇虎科技有限公司 一种备选电话号码推送方法和装置
CN105864976A (zh) * 2016-04-18 2016-08-17 广东美的环境电器制造有限公司 家电设备的联动控制方法和系统、服务器
CN106713083A (zh) * 2016-11-24 2017-05-24 海信集团有限公司 基于知识图谱的智能家居设备控制方法、装置及系统
CN109995620A (zh) * 2017-12-30 2019-07-09 成都标制文化创意有限公司 一种控制智能家居设备的方法和智能家居设备
CN207802669U (zh) * 2018-02-08 2018-08-31 深圳市艾特网能技术有限公司 一种智能一体化机柜
CN109754804A (zh) * 2019-02-21 2019-05-14 珠海格力电器股份有限公司 一种语音控制方法、装置、存储介质及智能家居系统
CN110457489A (zh) * 2019-08-12 2019-11-15 珠海格力电器股份有限公司 资源调度方法及系统、计算机可读存储介质

Also Published As

Publication number Publication date
US20220276897A1 (en) 2022-09-01
CN110457489A (zh) 2019-11-15

Similar Documents

Publication Publication Date Title
WO2021027437A1 (zh) 资源调度方法及系统、计算机可读存储介质
KR102480949B1 (ko) 디지털 음성 어시스턴트 컴퓨팅 디바이스들 사이에서 신호 프로세싱 조정
US11991258B2 (en) Methods, systems, and media for managing notifications directed to multiple applications installed on a user device
EP3583770B1 (en) Providing composite graphical assistant interfaces for controlling various connected devices
JP5134203B2 (ja) ネットワーク内の装置によって利用可能になったタスクを順位付ける方法及びシステム
US10185534B2 (en) Control method, controller, and recording medium
US9898534B2 (en) Automatically adapting a user interface
US20220060348A1 (en) Method for dynamically building environment conditioning rule list in smart home, and device
CN111095399A (zh) 助理应用的语音用户接口快捷方式
WO2021232797A1 (zh) 一种控制家居设备的方法、装置、计算机设备及存储介质
US11537417B2 (en) Task delegation and cooperation for automated assistants
WO2020199672A1 (zh) 空调器及其控制方法、控制装置、存储介质和处理器
CN110618614A (zh) 一种智能家居的控制方法、装置、存储介质及机器人
WO2021196610A1 (zh) 一种设备推荐方法、装置、电子设备及可读存储介质
US11177044B2 (en) Summarily conveying smart appliance statuses
EP4310418A2 (en) Method, device and medium for controlling water chiller in central air conditioning system
WO2023221357A1 (zh) 设备控制方法和相关装置
WO2022268136A1 (zh) 一种进行语音控制的终端设备及服务器
KR20150110877A (ko) 사용자 기반 홈 오토메이션 수행 방법 및 장치와 이를 지원하는 방법 및 장치
WO2020238174A1 (zh) 电子设备控制方法及装置、计算机可读存储介质
US11971944B1 (en) Suppression filter for presenting content in a network document
TW202044163A (zh) 使用者操作介面推薦方法
Coutand et al. Determining the behavior of mobile location-aware services based on user action history

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: 20852041

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 20852041

Country of ref document: EP

Kind code of ref document: A1