WO2014036832A1 - 基于社交网络的混合语言交互处理系统及其方法 - Google Patents

基于社交网络的混合语言交互处理系统及其方法 Download PDF

Info

Publication number
WO2014036832A1
WO2014036832A1 PCT/CN2013/074255 CN2013074255W WO2014036832A1 WO 2014036832 A1 WO2014036832 A1 WO 2014036832A1 CN 2013074255 W CN2013074255 W CN 2013074255W WO 2014036832 A1 WO2014036832 A1 WO 2014036832A1
Authority
WO
WIPO (PCT)
Prior art keywords
data information
social network
module
language form
language
Prior art date
Application number
PCT/CN2013/074255
Other languages
English (en)
French (fr)
Inventor
纪阳
沙晶
吴振宇
徐源
成城
张春红
邱晓峰
程胜
胡清源
Original Assignee
北京邮电大学
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 北京邮电大学 filed Critical 北京邮电大学
Publication of WO2014036832A1 publication Critical patent/WO2014036832A1/zh

Links

Classifications

    • 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
    • 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
    • G06Q30/00Commerce
    • G06Q30/02Marketing; Price estimation or determination; Fundraising
    • 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
    • G06Q50/00Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
    • G06Q50/01Social networking

Definitions

  • the present invention relates to the field of communications, and in particular, to a hybrid network interaction processing system based on social network and a method thereof.
  • SNS Social Network Service
  • Weibo is a kind of social network with a wide application. Weibo has strong openness, real-time and high information sharing. At the same time, it can be easily and conveniently customized by adding friends (attention). Forwarding/commenting enables interaction and information dissemination between users; in addition, since the microblogging social network provides an open web service API, the application development threshold is effectively reduced.
  • the social network platform in the prior art cannot support the above-mentioned "human-object” and “object-object” new social networks, and does not have the corresponding technology to realize autonomous interaction, control and other social interactions between different attribute entities.
  • the main purpose of the present invention is to provide a mixed language interaction processing system based on social network. And a method for solving the problem that the social network existing in the prior art does not support the social interaction function such as autonomous interaction, control, and the like between the different attribute entities, wherein: the social network-based mixed language interaction processing system according to the embodiment of the present invention includes: a first terminal resource management module, configured to collect raw data in a machine language form of the Internet of Things terminal device; a first context-aware inference module, configured to perform inference analysis on the original data according to preset context inference logic, and obtain the obtained The data information in the form of a machine language is sent to the first trigger execution module; the first trigger execution module is configured to start a social network mixed language sending function, and send the data information in the machine language form to the first mixed language mapping parsing module; a language mapping parsing module, configured to parse data information in a machine language form into data information in a corresponding natural language form according to a correspondence between a preset machine language and a
  • the first hybrid interaction module publishes data information in a mixed language form on a social network, and the mixed language form includes a natural language form and a machine language form.
  • the system further includes: a second hybrid interaction module, configured to acquire data information in a natural language form through a social network; and a second mixed language mapping parsing module, configured to acquire a machine language form corresponding to the data information in a natural language form
  • the second context-aware inference module is configured to perform inference analysis on the data information according to the preset context reasoning logic to obtain the inference result;
  • the second trigger execution module is configured to generate the control on the object network terminal device according to the inference result.
  • the second terminal resource management module is configured to control the Internet of Things terminal according to the control instruction.
  • the second mixed language mapping parsing module acquires data information in a machine language form corresponding to the data information in the natural language form through the social network or the first hybrid interaction module.
  • the Internet of Things terminal device registers an account on the social network
  • the first hybrid interaction module issues data information in a natural language form or a mixed language form on the social network with the registered account of the Internet of Things terminal device.
  • a social network-based hybrid language interaction processing method includes: collecting raw data in a machine language form of an Internet of Things terminal device;
  • the original data is reasoned and analyzed to obtain the data information in the form of machine language; the social network mixed language sending function is activated, according to the preset machine language and nature.
  • the correspondence between languages, the data information in the form of machine language is parsed into the corresponding data information in the form of natural language; the data information in the form of natural language is published on the social network.
  • the method further includes: publishing a result of the inference of the mixed language form on the social network, the mixed language form including a natural language form and a machine language form.
  • the method further includes: receiving data information in a natural language form through a social network; acquiring data information in a machine language form corresponding to the data information in a natural language form; and performing inference analysis on the data information according to the preset context reasoning logic, Obtaining the inference result; generating a control instruction for the Internet of Things terminal according to the inference result, and controlling the IoT terminal.
  • the step of acquiring data information in a machine language form corresponding to data information in a natural language form includes: acquiring, by a social network or an Internet of Things terminal device, data information in a machine language form corresponding to data information in a natural language form.
  • the Internet of Things terminal device registers an account on the social network, and publishes data information in a natural language form on the social network with the registered account of the Internet of Things terminal device.
  • the hybrid language interaction processing system constructs a mapping/parsing of natural language and machine language, and combines the data information (natural language or mixed language form) of the Internet of Things terminal device with the context information to be distributed on the social network to realize socialization.
  • the analysis and reaction of network information so as to complete the message interaction and control of socialization between different attribute entities, and realize a new type of social network where "human-object” and "object-object” coexist.
  • FIG. 1 is a block diagram showing the structure of an embodiment of the present invention.
  • FIG. 3 is a flow chart of data reception according to an embodiment of the present invention.
  • FIG. 4 is a flow chart of a hybrid language interaction processing method according to an embodiment of the present invention.
  • FIG. 5 is a flowchart of a hybrid language interaction processing method according to another embodiment of the present invention. detailed description
  • the main idea of the present invention is to collect data information in the machine language form of the Internet of Things terminal device; perform inference analysis on the data information according to preset or user-defined situational reasoning logic, and obtain the inference result in the form of machine language;
  • the social network mixed language sending function parses the inference result of the machine language form into the corresponding natural language form inference result according to the correspondence between the preset machine language and the natural language; and publishes the natural language form on the social network Reasoning results.
  • a social network based hybrid language interaction processing system is provided.
  • the hybrid language interactive processing system 100 is connected to a social network 200 and an Internet of Things terminal device 300.
  • the social network 200 (for example, Weibo) is a information sharing, dissemination, and acquisition platform based on user relationships; and the Internet of Things terminal devices refer to various types of Internet of Things terminal devices, such as gateways, smart terminals, sensors, and controllers. Equipment that can access the Internet and socialize.
  • the social network can support entities with different attributes, that is, the registered users of the social network are no longer just "people", but can be all kinds of Internet of Things terminal devices that can be connected to the Internet, messages issued by various registered users.
  • the content may include not only natural language, but also a machine language corresponding to the natural language or a corresponding machine language link.
  • the natural language that is, the language used for human communication, refers to a language that can be understood and used by ordinary people such as words, images, videos, etc.
  • Chinese is taken as an example for illustration.
  • Machine language which refers to functional modules, terminal devices, machine-processable, understandable language types, including but not limited to code, data, characters, instructions, etc.
  • the hybrid language interaction processing system 100 includes at least: a first terminal resource management module 10, a first context aware inference module 20, a first trigger execution module 30, a first mixed language mapping parsing module 40, and a first A hybrid interaction module 50.
  • the first terminal resource management module 10 is responsible for collecting and managing the original data of the Internet of Things terminal device, and mainly includes: device identification information, device description information, device status information, and the like. Among them, the device description The information includes: device name, type, model, deployment location, etc.; device status information includes current operating status (such as on/off).
  • the first terminal resource management module 10 uniformly identifies, stores, and manages the collected original data. In one embodiment of the invention, the collected raw data is managed by establishing a database (original database).
  • the first terminal resource management module 10 is further responsible for receiving a control instruction sent by the trigger execution module, and completing control of the terminal device according to the control instruction. It should be noted that the data information collected by the first terminal resource management module 10 is in the form of a machine language.
  • the first scenario modeling inference module 20 is configured to extract context data from the original data collected by the first terminal resource management module 10, and establish a context database including context inference logic, and collect data information according to the context inference logic Performing inference analysis, obtaining data information in the form of machine language, and transmitting to the first trigger execution module 30.
  • the context database stores context-inferential logic, including pervasive logic for ordinary users and special logic for specific users. Among them, the context reasoning logic is the condition of judgment, and the context reasoning logic can be system defined or user-defined.
  • the situational reasoning module is also responsible for real-time situational reasoning and correction based on the interaction process and user feedback.
  • the first trigger execution module 30 receives the data information of the first context modeling inference module 20, and then determines to start the social network mixed language sending function according to the data information, that is, the program for publishing the terminal data information to the social network. Next, the first trigger execution module 30 sends the data information to the first mixed language mapping parsing module 40.
  • the first mixed language mapping parsing module 40 is configured to map the data information in the machine language form to the data information in the natural language form according to the correspondence between the preset machine language and the natural language (machine language-natural language).
  • the natural language ontology may be preserved by establishing a corpus to support a mapping relationship between machine language and natural language. After the first terminal resource management module 10 establishes the original database according to the original data of the collection, the first mixed language mapping parsing module 40 can set the corresponding natural language according to the original data in the original database, and save the natural by establishing a corpus. Language ontology.
  • a first hybrid interaction module 50 configured to publish the data information in the natural language form on a social network; or publish data information in a mixed language form on a social network, where the mixed language includes a natural language and corresponds to the natural language Data information in the form of machine language/machine language links.
  • the first hybrid interaction module 50 provides an interface between the entire system and the social network, system and social Data interaction between the networks is performed by the first hybrid interaction module 50.
  • the social network platform is used as a natural language publishing platform (similar to the human-only microblog now), and the receiving end entity needs to register information and other addressing modes in the microblog platform by the sending end entity.
  • Obtain the network address of the sender entity directly send the "natural language-machine language" parsing request to the sender entity, and implement the interaction between the entities through a certain authentication process to obtain the corresponding machine language. This way of data interaction can be called implicit mode.
  • a natural language can be released to a social network platform, and a corresponding machine language can also be released.
  • the social network platform (background) stores entities about "natural language.” - The correspondence of machine language.
  • the receiving entity After receiving the natural language issued by the sender, the receiving entity must send a resolution request to the social network platform, return the corresponding machine language to the receiving end through the social networking platform, or transit the request through the social network platform, thereby realizing the parsing of the natural language.
  • This data interaction mode can be referred to as a display mode. In the display mode, the sender entity and the receiver entity do not directly perform data interaction.
  • the mixed language interactive processing system publishes the data information (natural language or mixed language) of the Internet of Things terminal device on the social network for sharing and acquisition by other users (users or users).
  • the data processing procedure of the mixed language interactive processing system of the present invention is exemplified below.
  • the temperature sensor detects the data information of "the ambient temperature is 35 degrees”
  • the first terminal resource management module collects the data information in the machine language form.
  • the first context-aware inference module infers and analyzes the data information according to the preset situational reasoning logic, and obtains the inference result of "the temperature is greater than 35 degrees, the environment is overheated, and the air conditioner needs to be opened”, and the inference result is submitted to the first trigger execution module.
  • the first trigger execution module receives the foregoing inference result, determines to start the social network mixed language sending function, and sends the inference result to the first mixed language mapping parsing module.
  • the first mixed language mapping parsing module generates the corresponding natural language form data according to the corresponding relationship between the preset machine language and the natural language, and the inference result of the machine language form "the temperature is greater than 35 degrees, the environment is overheated, and the air conditioner needs to be opened". Information, for example, can be generated "It's so hot today, even the house has become a steamer.” Then, the first hybrid interaction module publishes and shares data in the natural language form or the mixed language form on the social network for other users to obtain. After the above-described data information is released, the mixed-language interactive processing system of the present invention continues the process of receiving data.
  • the second hybrid interaction module is configured to acquire data information in a natural language form through a social network, where the data information in the natural language form is published on the social network in the name of the Internet of Things terminal device (registration information of the Internet of Things terminal in the social network) Data information. Then, the second hybrid interaction module sends the data information to the second mixed language mapping parsing module.
  • the second mixed language mapping parsing module is configured to acquire data information in a machine language form corresponding to the data information in the natural language form.
  • the machine language data acquisition here includes two ways:
  • the second context-aware inference module is configured to perform a reasoning analysis on the data information according to the preset context reasoning logic, and notify the second trigger execution module of the obtained reasoning result.
  • the second trigger execution module is configured to generate a control instruction for the Internet of Things terminal according to the inference result, and send the control instruction to the second terminal resource management module.
  • the second terminal resource management module is configured to control the Internet of Things terminal device according to the control instruction.
  • the status information of the Internet of Things terminal device that executes the control instruction may also be released on the social network as a response to the previously released information, thereby realizing the "object-object” or "person-object” between interactive.
  • the first terminal resource management module and the second terminal resource management module, the first context-aware inference module and the second context-aware inference module, the first trigger execution module, and the second trigger execution module First mixed language mapping parsing module and second mixed language mapping
  • the shot analysis module, the first hybrid interaction module and the second hybrid interaction module may be combined, and the structure and function of the module after the integration are not described herein.
  • FIG. 2 it is a data publishing process according to an embodiment of the present invention, which specifically includes:
  • Step S202 the Internet of Things terminal temperature sensor detects that the indoor ambient temperature is greater than 35 degrees, and transmits related data represented by the machine language to the resource management module;
  • Step S204 After receiving the message, the resource management module submits the terminal data to the situation modeling reasoning module for reasoning analysis;
  • Step S206 the context modeling inference module invokes the context database knowledge, infers that the temperature is too hot, needs to open the air conditioning result, and submits the inference result of the machine language representation to the trigger execution module;
  • Step S208 triggering the execution module to start after receiving the message a mixed language sending program, and submitting the inference result of the inference machine language representation to the mixed language mapping parsing module;
  • Step S210 the mixed language mapping parsing module calls the corpus to generate the natural language "Today is hot, even the house has become a steamer", and combined with the raw data represented by the machine language to further generate a mixed language representation of the release information, and the message Send to the hybrid interaction module;
  • Step S212 The hybrid interaction module sends the message to the social network platform, where the sensor resource data is published on the social network in a natural language form or a mixture of natural language and machine language.
  • Step S302 The hybrid interaction module receives a message from the social network platform, "Today is so hot, even the house has become a steamer", the message It can be implemented by the specified sending (@ ) of the other party, or it can be implemented by the periodic message search itself, for example, through social relationships such as microblog messages published by users or friends who are interested in it.
  • Step S304 the hybrid module submits the message represented by the natural language to the mixed language mapping parsing module for parsing;
  • Step S306 the hybrid language mapping parsing module invokes the corpus, parses the natural language into a corresponding machine language, and submits the parsing result to the context modeling inference module for reasoning;
  • Step S310 the message satisfies the condition for triggering the opening of the air conditioning event, so the triggering execution module sends the execution instruction of the air conditioner to the terminal resource management module after receiving the message;
  • Step S312 After receiving the execution instruction, the terminal resource management module turns on the air conditioner; after the air conditioner is turned on, feeds back its own state (temperature, sweep, humidity, etc.) to the terminal resource management module.
  • a hybrid language interaction processing method for the Internet of Things is also provided. Referring to FIG. 4, at least the following steps are included:
  • Step S402 collecting data information in a machine language form of the Internet of Things terminal device
  • Step S404 Perform inference analysis on the data information according to preset context reasoning logic, and obtain a reasoning result in a machine language form;
  • Step S406 Start a social network mixed language sending function, and parse the inference result of the machine language form into a corresponding natural language form inference result according to a correspondence between the preset machine language and the natural language;
  • Step S408 publishing the reasoning result of the natural language form on the social network.
  • data information in the form of a machine language corresponding to the data information in the natural language form is simultaneously distributed on the social network.
  • step S408 the method of the present invention further includes:
  • Step S512 receiving data information in a natural language form through a social network
  • Step S514 acquiring data information in a machine language form corresponding to the data information in the natural language form.
  • display mode there are two types: display mode and implicit mode.
  • implicit mode For details, please refer to the previous descriptions in this document, and we will not repeat them here.
  • Step S516 performing inference analysis on the data information according to the preset context reasoning logic, and obtaining a prediction result
  • Step S518, generating a control instruction for the Internet of Things terminal according to the inference result, and controlling the IoT terminal according to the control instruction.
  • the hybrid language interaction processing system constructs mapping/analysis of natural language and machine language, and combines situation information to data information of the Internet of Things terminal device.
  • Natural language or mixed language form is published on social networks to realize the analysis and reaction of social network information, thus completing the social message interaction and control between different attribute entities, realizing a new type of "human-object", "Society-object” coexisting social network.

Landscapes

  • Business, Economics & Management (AREA)
  • Engineering & Computer Science (AREA)
  • Strategic Management (AREA)
  • Economics (AREA)
  • Theoretical Computer Science (AREA)
  • Marketing (AREA)
  • Human Resources & Organizations (AREA)
  • Physics & Mathematics (AREA)
  • General Business, Economics & Management (AREA)
  • General Physics & Mathematics (AREA)
  • Development Economics (AREA)
  • Accounting & Taxation (AREA)
  • Game Theory and Decision Science (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Finance (AREA)
  • Tourism & Hospitality (AREA)
  • Operations Research (AREA)
  • Health & Medical Sciences (AREA)
  • General Health & Medical Sciences (AREA)
  • Primary Health Care (AREA)
  • Computing Systems (AREA)
  • Quality & Reliability (AREA)
  • Information Transfer Between Computers (AREA)

Abstract

本发明公开了一种基于社交网络的混合语言交互处理系统及其方法,该系统包括:第一终端资源管理模块,采集物联网终端设备的机器语言形式的原始数据;第一情境感知推理模块,根据预设的情境推理逻辑对原始数据进行推理分析,并将得到的机器语言形式的数据信息发送至第一触发执行模块;第一触发执行模块,启动社交网络混合语言发送功能,将机器语言形式的数据信息发送至第一混合语言映射解析模块;第一混合语言映射解析模块,根据机器语言与自然语言之间的对应关系,将机器语言形式的数据信息解析为对应的自然语言形式的数据信息;第一混合交互模块,在社交网络上发布自然语言形式的数据信息。本发明完成了不同属性实体间社会化的消息交互及控制。

Description

基于社交网络的混合语言交互处理系统及其方法 技术领域
本发明涉及通讯领域, 尤其涉及一种基于社交网络的混合语言交互处理 系统及其方法。
背景技术
社交网络( Social Network Service, 简称 SNS )是一种人类世界信息互通 的平台, 通过形成互联网中虚拟社交关系, 实现人与人的信息共享和交互。 目前, 微博是应用比较广泛的一种社交网络, 微博具有很强的开放性、 实时 性、 信息共享程度高; 同时, 可以通过加好友(关注) 简单方便的定制化的 获取信息, 通过转发 /评论实现用户之间的交互和信息传播; 另外, 由于微博 类社交网络提供开放的 web service API, 有效降低了应用开发门槛。
近年来, 物联网传感器设备、 智能终端设备的广泛部署促使各种基于釆 集数据的泛在应用进入人们的生活中。 人们不仅仅对传感器釆集的原始数据 存在兴趣, 对于现实世界的实体(设备, 地理位置, 其他人等) 以及这些实 体的状态 (是否空闲, 拥挤程度, 动作等) 同样十分关注。 因此人们更加地 依赖于具备情境感知能力的服务及设备。
然而, 当前的物联网实体, 特别是传感器等设备釆集到的数据、 实体状 态信息等数据大都以机器语言形态存在, 人们难以理解并加以利用; 并且, 当前的物联网设备部署和利用大都存在于封闭的系统中, 交互协议和交互语 言以私有为主, 导致互通性和开放程度不强, 外界难以访问并进行二次开发。
因此, 现有技术中的社交网络平台无法支持上述 "人 -物" 、 "物 -物"构 成的新型社交网络, 不具备相应的技术以实现不同属性实体间自主的交互、 控制等社会化交互功能。
发明内容
本发明的主要目的在于提供一种基于社交网络的混合语言交互处理系统 及其方法,以解决现有技术存在的社交网络不支持不同属性实体间自主交互、 控制等社会化交互功能的问题, 其中: 根据本发明实施例的基于社交网络的混合语言交互处理系统包括: 第一 终端资源管理模块, 用于釆集物联网终端设备的机器语言形式的原始数据; 第一情境感知推理模块, 用于根据预设的情境推理逻辑对原始数据进行推理 分析, 并将得到的机器语言形式的数据信息发送至第一触发执行模块; 第一 触发执行模块, 用于启动社交网络混合语言发送功能, 将机器语言形式的数 据信息发送至第一混合语言映射解析模块; 第一混合语言映射解析模块, 用 于根据预设的机器语言与自然语言之间的对应关系, 将机器语言形式的数据 信息解析为对应的自然语言形式的数据信息; 第一混合交互模块, 用于在社 交网络上发布自然语言形式的数据信息。
其中, 第一混合交互模块在社交网络上发布混合语言形式的数据信息, 混合语言形式包括自然语言形式和机器语言形式。
其中, 该系统还包括: 第二混合交互模块, 用于通过社交网络获取自然 语言形式的数据信息; 第二混合语言映射解析模块, 用于获取与自然语言形 式的数据信息相对应的机器语言形式的数据信息; 第二情境感知推理模块, 用于根据预设的情境推理逻辑对数据信息进行推理分析, 得到推理结果; 第 二触发执行模块, 用于根据推理结果生成对于物联网终端设备的控制指令; 第二终端资源管理模块, 用于根据控制指令对物联网终端进行控制。 其中, 第二混合语言映射解析模块通过社交网络或第一混合交互模块, 获取与自然语言形式的数据信息相对应的机器语言形式的数据信息。 其中, 物联网终端设备在社交网络上注册有账户, 第一混合交互模块在 社交网络上以物联网终端设备的注册账户发布自然语言形式或混合语言形式 的数据信息。
根据本发明实施例的基于社交网络的混合语言交互处理方法包括: 釆集 物联网终端设备的机器语言形式的原始数据;
根据预设的情境推理逻辑对原始数据进行推理分析, 得到机器语言形式 的数据信息; 启动社交网络混合语言发送功能, 根据预设的机器语言与自然 语言之间的对应关系, 将机器语言形式的数据信息解析为对应的自然语言形 式的数据信息; 在社交网络上发布自然语言形式的数据信息。
其中, 该方法还包括: 在社交网络上发布混合语言形式的推理结果, 混 合语言形式包括自然语言形式和机器语言形式。
其中, 该方法还包括: 通过社交网络接收自然语言形式的数据信息; 获 取与自然语言形式的数据信息相对应的机器语言形式的数据信息; 根据预设 的情境推理逻辑对数据信息进行推理分析, 得到推理结果; 根据推理结果生 成对于物联网终端的控制指令, 并对物联网终端进行控制。
其中, 获取与自然语言形式的数据信息相对应的机器语言形式的数据信 息的步骤, 包括: 通过社交网络或物联网终端设备获取与自然语言形式的数 据信息相对应的机器语言形式的数据信息。
其中, 物联网终端设备在社交网络上注册有账户, 在社交网络上以物联 网终端设备的注册账户发布自然语言形式的数据信息。
根据本发明的技术方案, 混合语言交互处理系统构建自然语言与机器语 言的映射 /解析, 结合情境信息将物联网终端设备的数据信息 (自然语言或混 合语言形式)在社交网络上发布, 实现社交网络信息的解析和反应, 从而完 成不同属性实体间社会化的消息交互及控制, 实现了一种新型的 "人 -物" , "物 -物" 共存的社交网络。 附图说明
此处所说明的附图用来提供对本发明的进一步理解, 构成本申请的一部 分, 本发明的示意性实施例及其说明用于解释本发明, 并不构成对本发明的 不当限定。 在附图中:
图 1是本发明实施例的结构框图;
图 2是本发明实施例的数据发布的流程图;
图 3是本发明实施例的数据接收的流程图;
图 4是本发明一个实施例的混合语言交互处理方法的流程图;
图 5是本发明另一实施例的混合语言交互处理方法的流程图。 具体实施方式
本发明的主要思想在于, 釆集物联网终端设备的机器语言形式的数据信 息; 根据预设或用户自定义的情境推理逻辑对所述数据信息进行推理分析, 得到机器语言形式的推理结果; 启动社交网络混合语言发送功能, 根据预设 的机器语言与自然语言之间的对应关系, 将机器语言形式的推理结果解析为 对应的自然语言形式的推理结果; 在社交网络上发布所述自然语言形式的推 理结果。
为使本发明的目的、 技术方案和优点更加清楚, 以下结合附图及具体实 施例, 对本发明作进一步地详细说明。 根据本发明的实施例, 提供了一种基于社交网络的混合语言交互处理系 统。
参考图 1 , 该混合语言交互处理系统 100与社交网络 200和物联网终端 设备 300连接。 所述社交网络 200 (例如微博) , 是一个基于用户关系的信 息分享、传播以及获取平台; 所述物联网终端设备是指各类物联网终端设备, 如网关、 智能终端、 传感器、 控制器等能够接入互联网并进行社会化交互的 设备。 社交网络能够支持不同属性的实体, 也就是说, 所述社交网络的注册 用户不再仅仅是 "人" , 而可以是各类可以连接到互联网的物联网终端设备, 各类注册用户发布的消息内容不仅可以包括自然语言, 还可以包括自然语言 对应的机器语言或者对应的机器语言链接。
所述自然语言, 即用于人类交流的语言, 是指文字、 图像、 视频等普通 人可理解、 使用的语言, 在本文中以中文为例进行说明。 而机器语言, 是指 面向功能模块、 终端设备的, 机器可处理、 可理解的语言类型, 包括但不限 于代码、 数据、 字符、 指令等。
如图 1所示, 所述混合语言交互处理系统 100至少包括: 第一终端资源 管理模块 10、 第一情境感知推理模块 20、 第一触发执行模块 30、 第一混合 语言映射解析模块 40、 第一混合交互模块 50。
第一终端资源管理模块 10 ,负责釆集并管理物联网终端设备的原始数据, 主要包括: 设备标识信息、 设备描述信息、 设备状态信息等。 其中, 设备描 述信息包括: 设备名称、 类型、 型号、 部署位置等信息; 设备状态信息包括 当前运行状态 (如开 /关)等。 第一终端资源管理模块 10对釆集到的原始数 据进行统一标识、 存储、 管理。 在本发明的一个实施例中, 通过建立数据库 (原始数据库) 的方式管理釆集到的原始数据。 此外, 第一终端资源管理模 块 10还负责接收触发执行模块发送的控制指令,并根据该控制指令完成对终 端设备的控制。 需要说明, 第一终端资源管理模块 10釆集的数据信息为机器 语言形式。
第一情境建模推理模块 20, 负责从第一终端资源管理模块 10釆集的原 始数据中提取情境数据, 并建立包括情境推理逻辑的情境数据库, 根据所述 情境推理逻辑对釆集的数据信息进行推理分析, 得到机器语言形式的数据信 息, 并发送至第一触发执行模块 30。 情境数据库存储情境推理逻辑, 包含针 对普通用户的普适逻辑和针对特定用户的特殊逻辑。 其中, 情境推理逻辑就 是判断的条件, 情境推理逻辑可以是系统定义或用户自定义的。 情境推理模 块还负责根据交互过程及用户反馈, 实现情境实时推理和修正。
第一触发执行模块 30接收第一情境建模推理模块 20的数据信息, 然后 根据该数据信息判断启动社交网络混合语言发送功能, 也就是开启将终端数 据信息发布至社交网络的程序。接着, 第一触发执行模块 30将所述数据信息 发送至第一混合语言映射解析模块 40。
第一混合语言映射解析模块 40, 用于根据预设的机器语言与自然语言之 间 (机器语言 -自然语言)的对应关系, 将机器语言形式的数据信息映射为自 然语言形式的数据信息。 在本发明的一个实施例中, 可以通过建立语料库来 保存自然语言本体, 以支持机器语言与自然语言之间的映射关系。 当第一终 端资源管理模块 10根据釆集的原始数据建立原始数据库后,第一混合语言映 射解析模块 40就可以根据原始数据库中的原始数据设置对应的自然语言,并 通过建立语料库来保存这些自然语言本体。
第一混合交互模块 50, 用于在社交网络上发布所述自然语言形式的数据 信息; 或者在社交网络上发布混合语言形式的数据信息, 所述混合语言就是 包括自然语言和与该自然语言对应的机器语言 /机器语言链接形式的数据信 息。 第一混合交互模块 50提供整个系统与社交网络之间的接口, 系统与社交 网络之间的数据交互都通过第一混合交互模块 50进行。
在只发布自然语言的情况下, 社交网络平台作为自然语言发布平台 (类 似现在仅面向人的微博) , 接收端实体需要通过发送端实体在微博平台中的 注册信息及其他寻址方式, 获得发送端实体的网络地址, 直接向发送端实体 发送 "自然语言 -机器语言" 的解析请求, 通过一定的认证过程, 实现实体间 的交互, 获得对应的机器语言。 这种数据交互方式可以称为隐式模式,
在发布混合语言的情况下, 根据本发明的一个实施例, 向社交网络平台 发布自然语言的同时, 还可以将其对应的机器语言进行发布, 社交网络平台 (后台)存储各实体关于 "自然语言 -机器语言" 的对应关系。 接收端实体接 收到发送端发布的自然语言后, 必须向社交网络平台发出解析请求, 通过社 交网络平台将对应的机器语言返回接收端,或者通过社交网络平台中转请求, 从而实现自然语言的解析。 这种数据交互方式可以称为显示模式, 在显示模 式中, 发送端实体与接收端实体不直接进行数据交互。
通过上述模块的处理, 混合语言交互处理系统将物联网终端设备的数据 信息 (自然语言或混合语言)在社交网络上发布, 供其他用户 (人或物的用 户)分享、 获取。
下面举例说明本发明的混合语言交互处理系统的数据处理过程。 温度传 感器检测到 "环境温度为 35度" 的数据信息, 第一终端资源管理模块釆集到 该机器语言形式的数据信息。 第一情境感知推理模块根据预设的情境推理逻 辑对数据信息进行推理分析, 得出 "温度大于 35度环境过热, 需要开空调" 的推理结果, 并将推理结果提交至第一触发执行模块。 第一触发执行模块接 收到上述推理结果, 判断启动社交网络混合语言发送功能, 并将所述推理结 果发送至第一混合语言映射解析模块。 第一混合语言映射解析模块根据预设 的机器语言与自然语言之间的对应关系, 将机器语言形式的 "温度大于 35度 环境过热, 需要开空调" 的推理结果生成对应的自然语言形式的数据信息, 例如可以生成 "今天好热啊, 连屋里都成了蒸笼了" 。 接着, 第一混合交互 模块将上述自然语言形式或者是混合语言形式的数据信息在社交网络上发 布、 分享数据, 供其他用户获取。 在经过上述的数据信息发布后, 本发明的混合语言交互处理系统继续进 行接收数据的处理过程。
根据本发明实施例的混合语言交互处理系统, 还包括:
第二混合交互模块用于通过社交网络获取自然语言形式的数据信息, 该 自然语言形式的数据信息是以物联网终端设备的名义(物联网终端在社交网 络的注册信息)在社交网络上发布的数据信息。 接着, 第二混合交互模块将 该数据信息发送至第二混合语言映射解析模块。
第二混合语言映射解析模块用于获取与自然语言形式的数据信息相对应 的机器语言形式的数据信息。 具体地, 根据之前数据发布的方式, 此处的机 器语言数据获取包括两种方式:
( 1 )通过发送端实体在社交网络平台中的注册信息及其他寻址方式, 获 得发送端实体的网络地址, 直接向发送端实体发送 "自然语言 -机器语言" 的 解析请求, 通过一定的认证过程, 实现实体间的交互, 完成机器语言数据的 获取。
( 2 )通过向社交网络平台发出解析请求, 通过社交网络平台将对应的机 器语言返回接收端, 或者通过社交网络平台中转请求, 从而实现自然语言的 解析, 完成机器语言数据的获取。
第二情境感知推理模块用于根据预设的情境推理逻辑对数据信息进行推 理分析, 并将得到的推理结果通知第二触发执行模块。 第二触发执行模块用 于根据推理结果生成对于物联网终端的控制指令, 并发送至第二终端资源管 理模块。 第二终端资源管理模块, 用于根据控制指令对物联网终端设备进行 控制。
在上述处理之后, 还可以将执行控制指令的物联网终端设备的状态信息 在社交网络上发布, 作为对于之前发布信息的响应, 从而实现了 "物 -物" 或 "人 -物" 之间的互动。
在本发明的一个实施例中, 上述的第一终端资源管理模块与第二终端资 源管理模块、 第一情境感知推理模块与第二情境感知推理模块、 第一触发执 行模块与第二触发执行模块、 第一混合语言映射解析模块与第二混合语言映 射解析模块、 第一混合交互模块与第二混合交互模块可以合一设置, 合一设 置后的模块的结构和功能此处不再赘述。
下面以温度传感器为例详细描述本发明系统的处理流程。 参考图 2 , 是 本发明实施例的数据发布流程, 具体包括:
步骤 S202 , 物联网终端温度传感器检测到室内环境温度大于 35度, 并 将机器语言表示的相关数据传输给资源管理模块;
步骤 S204 , 资源管理模块收到消息后将终端数据提交给情境建模推理模 块进行推理分析;
步骤 S206 , 情境建模推理模块调用情境数据库知识, 推理出温度太热, 需要开空调的结果, 并将机器语言表示的推理结果提交给触发执行模块; 步骤 S208 , 触发执行模块收到消息后启动混合语言发送程序, 并将推理 机器语言表示的推理结果提交给混合语言映射解析模块;
步骤 S210 , 混合语言映射解析模块调用语料库生成自然语言 "今天好热 啊, 连屋里都成了蒸笼了" , 并结合机器语言表示的原始数据进一步生成混 合语言表示的发布信息, 并将此消息发送给混合交互模块;
步骤 S212 , 混合交互模块将消息发送到社交网络平台, 此时传感器资源 数据以自然语言形式或者是自然语言和机器语言混合的形式发布在社交网络 上。
下面继续参考图 3 , 是本发明实施例的数据接收流程, 具体包括: 步骤 S302 , 混合交互模块从社交网络平台接收到消息 "今天好热啊, 连 屋里都成了蒸笼了" , 这个消息可以通过对方的指定发送(@ ) 实现, 也可 以自己进行周期性消息搜索实现, 例如, 通过自己关注的用户或者好友用户 发布的微博消息等社交关系实现。
步骤 S304, 混合模块将自然语言表示的消息提交给混合语言映射解析模 块进行解析;
步骤 S306 , 混合语言映射解析模块调用语料库, 将自然语言解析为对应 的机器语言, 并将解析结果提交给情境建模推理模块进行推理; 步骤 S308, 情境建模推理模块调用情境数据库, 推理出温度大于 35度 导致环境过热, 需要开空调, 并将推理结果提交给触发执行模块;
步骤 S310, 此消息满足触发开空调事件的条件, 因此触发执行模块收到 消息后发送开空调的执行指令至终端资源管理模块;
步骤 S312, 终端资源管理模块收到执行指令后打开空调; 空调打开后将 自己的状态 (温度, 扫风, 湿度等)反馈给终端资源管理模块。
之后, 执行如步骤 S202-S212的流程, 此处不赘述。
通过上述处理,融合了计算机人机交互技术和社交网络社会化交互思想, 通过微博发布平台及微博 API对各类物联网终端设备数据和状态进行开放, 并借助微博关注、 评论、 转发等功能实现 "人 -物" 、 "物 -物" 的自然语言交 互。
根据本发明实施例,还提供了一种用于物联网的混合语言交互处理方法, 参考图 4, 至少包括以下步骤:
步骤 S402, 釆集物联网终端设备的机器语言形式的数据信息;
步骤 S404, 根据预设的情境推理逻辑对所述数据信息进行推理分析, 得 到机器语言形式的推理结果;
步骤 S406, 启动社交网络混合语言发送功能, 根据预设的机器语言与自 然语言之间的对应关系, 将机器语言形式的推理结果解析为对应的自然语言 形式的推理结果;
步骤 S408, 在社交网络上发布所述自然语言形式的推理结果。
此外, 同时在社交网络上发布与所述自然语言形式的数据信息相对应的 机器语言形式的数据信息。
参考图 5 , 在步骤 S408之后, 本发明的方法还包括:
步骤 S512, 通过社交网络接收自然语言形式的数据信息;
步骤 S514, 获取与自然语言形式的数据信息相对应的机器语言形式的数 据信息。 根据数据发布形式的不同, 此处可以分为显示模式和隐式模式两种 方式, 具体请参见本文之前的描述, 此处不再赘述。
步骤 S516, 根据预设的情境推理逻辑对数据信息进行推理分析, 得到推 理结果;
步骤 S518, 根据推理结果生成对于物联网终端的控制指令, 并根据控制 指令对物联网终端进行控制。
综上所述, 根据本发明的上述实施例, 混合语言交互处理系统构建自然 语言与机器语言的映射 /解析, 结合情境信息将物联网终端设备的数据信息
(自然语言或混合语言形式)在社交网络上发布, 实现社交网络信息的解析 和反应, 从而完成不同属性实体间的社会化消息交互及控制, 实现了一种新 型的 "人 -物" , "物 -物" 共存的社交网络。
以上所述仅为本发明的实施例而已, 并不用于限制本发明, 对于本领域 的技术人员来说, 本发明可以有各种更改和变化。 凡在本发明的精神和原则 之内, 所作的任何修改、 等同替换、 改进等, 均应包含在本发明的权利要求 范围之内。
10
P1310092CN-PCT-BUPT

Claims

权 利 要 求 书
1、 一种基于社交网络的混合语言交互处理系统, 其特征在于, 包括: 第一终端资源管理模块, 用于釆集物联网终端设备的机器语言形式的原 始数据;
第一情境感知推理模块, 用于根据预设的情境推理逻辑对所述原始数据 进行推理分析, 并将得到的机器语言形式的数据信息发送至第一触发执行模 块;
第一触发执行模块, 用于启动社交网络混合语言发送功能, 将所述机器 语言形式的数据信息发送至第一混合语言映射解析模块;
第一混合语言映射解析模块, 用于根据预设的机器语言与自然语言之间 的对应关系, 将机器语言形式的数据信息解析为对应的自然语言形式的数据 信息;
第一混合交互模块, 用于在社交网络上发布所述自然语言形式的数据信 息。
2、根据权利要求 1所述的系统, 其特征在于, 所述第一混合交互模块在 社交网络上发布混合语言形式的数据信息, 所述混合语言形式包括自然语言 形式和机器语言形式。
3、 根据权利要求 1或 2所述的系统, 其特征在于, 还包括:
第二混合交互模块, 用于通过所述社交网络获取所述自然语言形式的数 据信息;
第二所述混合语言映射解析模块, 用于获取与所述自然语言形式的数据 信息相对应的机器语言形式的数据信息;
第二所述情境感知推理模块, 用于根据预设的情境推理逻辑对所述数据 信息进行推理分析, 得到推理结果;
第二所述触发执行模块, 用于根据推理结果生成对于物联网终端设备的 控制指令; 第二所述终端资源管理模块, 用于根据所述控制指令对物联网终端进行 控制。
4、根据权利要求 3所述的系统, 其特征在于, 所述第二所述混合语言映 射解析模块通过所述社交网络或所述第一混合交互模块, 获取与所述自然语 言形式的数据信息相对应的机器语言形式的数据信息。
5、根据权利要求 2所述的系统, 其特征在于, 所述物联网终端设备在所 述社交网络上注册有账户, 所述第一混合交互模块在社交网络上以所述物联 网终端设备的注册账户发布所述自然语言形式或所述混合语言形式的数据信 息。
6、 一种基于社交网络的混合语言交互处理方法, 其特征在于, 包括: 釆集物联网终端设备的机器语言形式的原始数据;
根据预设的情境推理逻辑对所述原始数据进行推理分析, 得到机器语言 形式的数据信息;
启动社交网络混合语言发送功能, 根据预设的机器语言与自然语言之间 的对应关系, 将机器语言形式的数据信息解析为对应的自然语言形式的数据 信息;
在社交网络上发布所述自然语言形式的数据信息。
7、 根据权利要求 6所述的方法, 其特征在于, 还包括:
在社交网络上发布混合语言形式的推理结果, 所述混合语言形式包括自 然语言形式和机器语言形式。
8、 根据权利要求 6或 7所述的方法, 其特征在于, 还包括:
通过所述社交网络接收所述自然语言形式的数据信息;
获取与所述自然语言形式的数据信息相对应的机器语言形式的数据信 息;
根据预设的情境推理逻辑对所述数据信息进行推理分析,得到推理结果; 根据推理结果生成对于所述物联网终端的控制指令, 并对所述物联网终 端进行控制。
9、根据权利要求 8所述的方法, 其特征在于, 所述获取与所述自然语言 形式的数据信息相对应的机器语言形式的数据信息的步骤, 包括:
通过所述社交网络或所述物联网终端设备获取与所述自然语言形式的数 据信息相对应的机器语言形式的数据信息。
10、 根据权利要求 6所述的方法, 其特征在于, 所述物联网终端设备在 所述社交网络上注册有账户, 在社交网络上以所述物联网终端设备的注册账 户发布所述自然语言形式的数据信息。
PCT/CN2013/074255 2012-09-06 2013-04-16 基于社交网络的混合语言交互处理系统及其方法 WO2014036832A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201210328436.4A CN102857506B (zh) 2012-09-06 2012-09-06 基于社交网络的混合语言交互处理系统及其方法
CN201210328436.4 2012-09-06

Publications (1)

Publication Number Publication Date
WO2014036832A1 true WO2014036832A1 (zh) 2014-03-13

Family

ID=47403700

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2013/074255 WO2014036832A1 (zh) 2012-09-06 2013-04-16 基于社交网络的混合语言交互处理系统及其方法

Country Status (2)

Country Link
CN (1) CN102857506B (zh)
WO (1) WO2014036832A1 (zh)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2016189084A1 (en) * 2015-05-26 2016-12-01 Abb Schweiz Ag Social media and industrial internet of things

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102857506B (zh) * 2012-09-06 2015-11-25 纪阳 基于社交网络的混合语言交互处理系统及其方法
US20170061011A1 (en) * 2015-08-25 2017-03-02 International Mobile Iot Corp Server and data search method

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102457483A (zh) * 2010-10-25 2012-05-16 中兴通讯股份有限公司 一种视频监控前端及其向物联网上报数据的方法
CN102495611A (zh) * 2011-12-23 2012-06-13 郭松 安防消防电梯综合安全管理系统
CN102613897A (zh) * 2012-01-11 2012-08-01 浙江大学 一种具有信息交互功能的智能饮水机及其信息交互方法
CN102629249A (zh) * 2012-02-24 2012-08-08 王博 一种提供网络信息的方法及系统
CN102857506A (zh) * 2012-09-06 2013-01-02 纪阳 基于社交网络的混合语言交互处理系统及其方法

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102457483A (zh) * 2010-10-25 2012-05-16 中兴通讯股份有限公司 一种视频监控前端及其向物联网上报数据的方法
CN102495611A (zh) * 2011-12-23 2012-06-13 郭松 安防消防电梯综合安全管理系统
CN102613897A (zh) * 2012-01-11 2012-08-01 浙江大学 一种具有信息交互功能的智能饮水机及其信息交互方法
CN102629249A (zh) * 2012-02-24 2012-08-08 王博 一种提供网络信息的方法及系统
CN102857506A (zh) * 2012-09-06 2013-01-02 纪阳 基于社交网络的混合语言交互处理系统及其方法

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2016189084A1 (en) * 2015-05-26 2016-12-01 Abb Schweiz Ag Social media and industrial internet of things

Also Published As

Publication number Publication date
CN102857506A (zh) 2013-01-02
CN102857506B (zh) 2015-11-25

Similar Documents

Publication Publication Date Title
US11677812B2 (en) Lightweight IoT information model
KR101973298B1 (ko) M2m-iot 서비스의 발행 및 발견
CN105453047B (zh) 提供物联网(iot)适配服务的系统和方法
KR101812398B1 (ko) Iot 시스템들에 대한 서비스로서의 데이터 주석화
KR101467173B1 (ko) M2m 네트워크의 리소스 관리 방법 및 리소스 관리 장치
WO2015081786A1 (zh) 订阅通知的实现方法和装置
EP3861706B1 (en) Framework for dynamic brokerage and management of topics and data at the service layer
EP3378217A1 (en) Cross-resource subscription for m2m service layer
WO2017128178A1 (zh) 一种信息交互方法、客户端和装置
CN103326929B (zh) 一种消息传输方法和装置
Diwan et al. A framework for modeling and verifying IoT communication protocols
Elmangoush et al. Application-derived communication protocol selection in M2M platforms for smart cities
WO2014036832A1 (zh) 基于社交网络的混合语言交互处理系统及其方法
KR102500594B1 (ko) 통신 네트워크에서의 서비스 계층 메시지 템플릿들
Azzara et al. The icsi m2m middleware for iot-based intelligent transportation systems
Hu et al. A distributed and efficient system architecture for smart home
US20100049804A1 (en) Instant Messaging
Li et al. Efficient oneM2M protocol conversion platform based on NB-IoT access
Auzias Programming Support for a Delay-Tolerant Web of Things
CN102387203B (zh) 一种物联网中基于soap协议的多播应用方法
CN105554048A (zh) 消息推送方法和装置

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: 13835528

Country of ref document: EP

Kind code of ref document: A1

DPE1 Request for preliminary examination filed after expiration of 19th month from priority date (pct application filed from 20040101)
NENP Non-entry into the national phase

Ref country code: DE

32PN Ep: public notification in the ep bulletin as address of the adressee cannot be established

Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205 DATED 23/07/2015)

122 Ep: pct application non-entry in european phase

Ref document number: 13835528

Country of ref document: EP

Kind code of ref document: A1