WO2019228232A1 - 对话系统间的知识分享方法、对话方法及装置 - Google Patents

对话系统间的知识分享方法、对话方法及装置 Download PDF

Info

Publication number
WO2019228232A1
WO2019228232A1 PCT/CN2019/087808 CN2019087808W WO2019228232A1 WO 2019228232 A1 WO2019228232 A1 WO 2019228232A1 CN 2019087808 W CN2019087808 W CN 2019087808W WO 2019228232 A1 WO2019228232 A1 WO 2019228232A1
Authority
WO
WIPO (PCT)
Prior art keywords
knowledge
shared
dialogue system
answer
dialogue
Prior art date
Application number
PCT/CN2019/087808
Other languages
English (en)
French (fr)
Inventor
葛付江
初敏
郭涛涛
戴中原
Original Assignee
苏州思必驰信息科技有限公司
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 苏州思必驰信息科技有限公司 filed Critical 苏州思必驰信息科技有限公司
Priority to JP2020565948A priority Critical patent/JP7108720B2/ja
Priority to EP19811963.8A priority patent/EP3805954A4/en
Priority to US17/057,722 priority patent/US20210201167A1/en
Publication of WO2019228232A1 publication Critical patent/WO2019228232A1/zh
Priority to JP2022113679A priority patent/JP2022160464A/ja

Links

Images

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06NCOMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N5/00Computing arrangements using knowledge-based models
    • G06N5/02Knowledge representation; Symbolic representation
    • G06N5/022Knowledge engineering; Knowledge acquisition
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F40/00Handling natural language data
    • G06F40/20Natural language analysis
    • G06F40/205Parsing
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F40/00Handling natural language data
    • G06F40/20Natural language analysis
    • G06F40/205Parsing
    • G06F40/216Parsing using statistical methods
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F40/00Handling natural language data
    • G06F40/30Semantic analysis
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F40/00Handling natural language data
    • G06F40/30Semantic analysis
    • G06F40/35Discourse or dialogue representation
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06NCOMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N5/00Computing arrangements using knowledge-based models
    • G06N5/04Inference or reasoning models
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L51/00User-to-user messaging in packet-switching networks, transmitted according to store-and-forward or real-time protocols, e.g. e-mail
    • H04L51/02User-to-user messaging in packet-switching networks, transmitted according to store-and-forward or real-time protocols, e.g. e-mail using automatic reactions or user delegation, e.g. automatic replies or chatbot-generated messages

Definitions

  • the present application relates to the field of artificial intelligence technology, and in particular, to a knowledge sharing method, dialogue method, device, electronic device, and storage medium between dialogue systems.
  • Dialogue systems can be WeChat public accounts, physical robots, web dialogue robots, computers, mobile phones, or other mobile device applications. Each of these forms of dialogue systems may be used in different business areas, such as web dialogue robots, There may be a web dialogue robot for the direction of business cooperation, a web dialogue robot for the direction of technology development, a web dialogue robot for the direction of technical consultation, and the like.
  • the knowledge points contained in the background knowledge bases corresponding to the web conversation robots in different business directions are organized around their respective business directions and fields.
  • the embodiments of the present application provide a knowledge sharing method, a dialogue method, a device, an electronic device, and a storage medium between dialogue systems, which are used to solve at least one of the foregoing technical problems.
  • an embodiment of the present application provides a knowledge sharing method between dialogue systems, which is applied to an electronic device, and the method includes:
  • the first dialogue system receives a knowledge sharing request sent by an external dialogue system, and the knowledge sharing request includes at least characteristic information of knowledge points to be shared of the external dialogue system;
  • the feature information of the knowledge points to be shared is added to the knowledge base of the first dialog system to form a shared knowledge base.
  • an embodiment of the present application provides a dialogue method for a first dialogue system, the first dialogue system sharing knowledge points of at least one external dialogue system, the knowledge base of the first dialogue system and the The knowledge points of at least one external dialogue system shared by the first dialogue system constitute a shared knowledge base, and the method includes:
  • a priority level of each of the multiple answers is determined, and the answer with the highest priority is presented to the user.
  • an embodiment of the present application provides a knowledge sharing device between dialogue systems, including:
  • An information receiving program module configured to receive a knowledge sharing request sent by an external dialogue system, where the knowledge sharing request includes at least characteristic information of knowledge points to be shared of the external dialogue system;
  • An information analysis program module configured to parse the knowledge sharing request to determine characteristic information of the knowledge points to be shared
  • the library update program module is configured to add feature information of the knowledge points to be shared to the knowledge base of the first dialog system to form a shared knowledge base.
  • an embodiment of the present application provides a dialogue device for a first dialogue system.
  • the first dialogue system shares knowledge points of at least one external dialogue system.
  • the knowledge base of the first dialogue system and the The knowledge points of at least one external dialogue system shared by the first dialogue system constitute a shared knowledge base, and the device includes:
  • Semantic parser module for parsing questions submitted by users
  • An answer retrieval program module configured to retrieve an answer corresponding to the question in the shared knowledge base according to the analysis result
  • a first answer program module configured to notify a user that there is no answer corresponding to the question when the retrieval result indicates that there is no answer in the shared knowledge base;
  • a second answer program module configured to present the unique answer to the user when the retrieval result indicates that the unique answer exists in the shared knowledge base
  • a third answering program module is configured to determine a priority level of each of the multiple answers when the retrieval result indicates that there are multiple answers in the shared knowledge base, and present the answer with the highest priority to the user.
  • an embodiment of the present application provides a non-volatile computer-readable storage medium.
  • the storage medium stores one or more programs including execution instructions that can be executed by an electronic device (including but not (Limited to a computer, a server, or a network device, etc.) to read and execute, for performing the knowledge sharing method (or the dialogue method) between any of the above dialogue systems in this application.
  • an electronic device including but not (Limited to a computer, a server, or a network device, etc.
  • an electronic device including: at least one processor, and a memory communicatively connected to the at least one processor, wherein the memory stores instructions executable by the at least one processor, The instructions are executed by the at least one processor, so that the at least one processor can execute a knowledge sharing method (or a dialogue method) between any of the dialogue systems in the present application.
  • an embodiment of the present application further provides a computer program product.
  • the computer program product includes a computer program stored on a non-volatile computer-readable storage medium, and the computer program includes program instructions.
  • the computer is caused to execute the knowledge sharing method (or dialogue method) between any of the above dialogue systems.
  • the embodiment of the present application realizes knowledge sharing between different dialogue systems by sharing knowledge points between dialogue systems, and minimizes the cost, and satisfies the user's cross-domain dialogue needs of the dialogue robot to the greatest extent.
  • FIG. 1 is a flowchart of an embodiment of a knowledge sharing method between dialog systems of the present application
  • FIG. 2 is a flowchart of an embodiment of a dialog method of the present application
  • step S22 in FIG. 2 is a flowchart of an embodiment of step S22 in FIG. 2;
  • FIG. 4 is a principle block diagram of an embodiment of a knowledge sharing device between dialog systems of the present application.
  • FIG. 5 is a functional block diagram of an embodiment of a dialog device of the present application.
  • FIG. 6 is a principle block diagram of an embodiment of an answer retrieval program module in the present application.
  • FIG. 7 is a schematic structural diagram of an embodiment of an electronic device of the present application.
  • program modules include routines, programs, objects, elements, data structures, etc. that perform specific tasks or implement specific abstract data types.
  • the present application can also be practiced in distributed computing environments in which tasks are performed by remote processing devices connected through a communication network.
  • program modules may be located in local and remote computer storage media, including storage devices.
  • module refers to relevant entities applied to a computer, such as hardware, a combination of hardware and software, software or software in execution, and the like.
  • an element may be, but is not limited to, a process running on a processor, a processor, an object, an executable element, an execution thread, a program, and / or a computer.
  • an application program, a script program running on a server, and a server may be components.
  • One or more elements can be in a process and / or thread of execution, and an element can be localized on one computer and / or distributed between two or more computers and can be run by a variety of computer-readable media .
  • An element may also pass based on a signal having one or more data packets, for example, a signal from data that interacts with another element in a local system, a distributed system, and / or interacts with other systems via a signal over a network on the Internet Local and / or remote processes to communicate.
  • an embodiment of the present application provides a method for sharing knowledge between dialog systems, which is applied to an electronic device.
  • the electronic device may be various electronic devices having a display screen and supporting human-computer interaction operations, including but not Limited to desktop computers, laptops, smart phones, tablets, and more.
  • the above electronic device executes the knowledge organization method of the embodiment of the present application, the following steps are implemented:
  • the first dialogue system receives a knowledge sharing request sent by an external dialogue system, and the knowledge sharing request includes at least characteristic information of knowledge points to be shared in the external dialogue system.
  • the knowledge sharing of the dialogue system can be carried out within a certain group, or shared with everyone.
  • a company-wide dialogue system X members of the company have their own dialogue systems A, B, C, etc., dialogue systems A, B, After C shares his knowledge with X, the dialogue in X can include X, A, B, C and other roles.
  • the first dialogue system may be dialogue system X
  • the external dialogue system may include dialogue system A, dialogue system B, and dialogue system C. among them,
  • Each knowledge set consists of a set of knowledge points q.
  • a knowledge point can be a question-answer pair or a triple in a knowledge map.
  • Dialogue system A, dialogue system B, and dialogue system C may share part of their own knowledge points as dialogue knowledge points to be shared with dialogue system X.
  • the feature information of the knowledge points to be shared may be feature vectors or ID information corresponding to the knowledge points to be shared, which is not limited in this application.
  • the corresponding reward of the dialogue system sharing the knowledge point is given according to the frequency of use of the knowledge.
  • the reward is a point value of the dialogue system, which can be expressed as points, grades or corresponding material rewards of the dialogue system.
  • the embodiment of the present application realizes knowledge sharing between different dialogue systems by sharing knowledge points between dialogue systems, so that the first dialogue system can smoothly answer cross-domain questions raised by users, and minimize costs. Under the circumstances, it satisfies the user's cross-domain dialogue needs to the greatest extent.
  • the dialogue system involved in the embodiment of the present application may be one or more of an application program of WeChat public account, webpage dialogue robot or physical robot, computer, mobile phone, or other mobile device.
  • the method based on the embodiment of the present application can share knowledge between dialog systems, and the form of the knowledge can be a question-answer pair, a knowledge base or a knowledge map, and a dialog generation algorithm learned by a machine learning method.
  • dialog system A has its own set of knowledge.
  • Dialogue system A can share its knowledge to dialog system B.
  • Dialogue system B can answer all questions corresponding to the knowledge in dialog system A and dialog system B.
  • the shared knowledge can be carried out in a dialogue with different roles.
  • dialogue system A can share its knowledge with dialogue system B, so that the dialogue in B can have two roles: dialogue system A and dialogue system B (and One end user role), which are displayed in the actual dialogue interface as three roles, of which two are dialogue system roles and one is a user role.
  • the first dialog system further receives configuration information of the external dialog system.
  • the configuration information includes at least avatar setting information of the external dialog system, and is used when the first dialog system answers a knowledge point belonging to the external dialog system.
  • the avatar of the external dialogue system is displayed on the dialogue interface.
  • the configuration information of the external dialogue system is correspondingly shared with the first dialogue system, so that when problems that exceed the scope of the knowledge points of the first dialogue system occur in actual dialogue, not only
  • the answer can be obtained from the knowledge points shared externally, and the avatar information of the external dialogue system that provides the answer can also be presented in the form of a third party, which can to some extent guide the user's current question. It is another category of knowledge in the business direction, which can also play a certain role in guiding and regulating users' knowledge of the field of knowledge.
  • the knowledge sharing request further includes association information of the knowledge points to be shared, which is used to determine a priority level of the knowledge points to be shared, and the association information includes: the knowledge points to be shared Number of adoptions a, and / or
  • Integral b of the external dialogue system corresponding to the knowledge points to be shared, and / or
  • the related information further includes a shared type of the knowledge points to be shared, and the shared types include: sharing to a specific dialogue system type (c 1 ), sharing to a specific dialogue system group type (c 2 ), and sharing to all External dialog system type (c 3 ).
  • the weighted sum of the function values of the parameters a, b, and c 1 -c 3 can be used as the comprehensive score w of the knowledge points to be shared, for determining the priority level of the knowledge points to be shared. Sharing knowledge points has a higher priority, for example,
  • the value of w is obtained through machine learning algorithms such as logistic regression and neural networks, such as
  • n 1 , n 2 , m 1 , m 2 , and m 3 can be obtained by training and fitting a large number of user data.
  • Whether the answer was adopted can be determined by whether the subsequent conversation continues after the recommendation of the answer during the dialogue or further dialogue behavior occurs for the answer or the answer provider. It is determined that each time the acceptance count is increased by 1, the more the number of acceptances is, the greater the probability that the user is likely to be accepted by the user. Therefore, the larger the corresponding determination is adopted, the higher the priority of the answer.
  • Each dialogue system has points based on the historical performance of serving the user. The better the historical performance, the more points are obtained. The more points are set in this embodiment, the corresponding belongs to the dialogue. The system's answer has higher priority.
  • the priority level of the knowledge points is determined according to the source of the knowledge points.
  • the priorities are, from high to low, as follows: the first dialogue system's own knowledge points, the knowledge points shared to a specific dialogue system, and the knowledge shared to a specific dialogue system group. Points to share knowledge points with all external dialogue systems. The greater the sharing range of knowledge points, the lower the priority weight.
  • the multiple answers can be sorted according to priority, so as to achieve optimal selection for the user.
  • Recommended answer action Specifically, the answer with the highest priority can be directly displayed in the dialog interface, or several answers with higher priority can be simultaneously displayed in the dialog interface for users to choose.
  • the purpose of setting the priority is mainly to solve the problem of using duplicate knowledge, avoid blind sharing, and improve the user experience.
  • the sharing function provided by the dialogue system is equivalent to allowing the original private knowledge to be shared, reducing the threshold for establishing a dialogue system and improving the ability of the dialogue system. Provide a certain social ability to the dialogue system. After knowledge sharing, when the knowledge point is asked again, it can be displayed in the form of a multi-person dialogue.
  • the present application further provides a dialog method for a first dialog system, which may be expressed as a software product or a hardware product (electronic device) equipped with the software product.
  • the dialogue system shares knowledge points of at least one external dialogue system, and the knowledge base of the first dialogue system and the knowledge points of at least one external dialogue system shared by the first dialogue system constitute a shared knowledge base.
  • S21 Analyze the question submitted by the user; specifically, semantically understand the question submitted by the user, and determine the intention of the user question for searching for a corresponding answer.
  • determining the respective priority levels of the plurality of answers includes:
  • association information of a knowledge point corresponding to each answer includes: a number of times a knowledge point corresponding to each answer is adopted, and / or
  • the points of the external dialogue system corresponding to the knowledge points corresponding to each answer, and / or
  • the attributes include local knowledge point types and shared types, and the shared types include: sharing to a specific dialogue system type, sharing to a specific dialogue system group type, and sharing to All external dialog system types;
  • the dialogue method in the embodiment of the present application is based on the shared knowledge base between the first dialogue system and the external dialogue system, and realizes the mutual complementation of knowledge points between dialogue systems, so that users can obtain knowledge points across business fields using a dialogue system.
  • the user's needs for knowledge in different business areas are satisfied under a dialogue system.
  • retrieving an answer corresponding to the question in the shared knowledge base according to the analysis result includes:
  • the distance between the external dialogue system and the domain of the first dialogue system is determined by comparing the similarities between the domains of the external dialogue system and the first dialogue system, respectively.
  • the standard of the answer to the question avoids the waste of computing resources caused by blind search in the shared database, and also improves the search efficiency. It can provide reference answers for the user's questions in the shortest time, ensuring the fluency of this dialogue process. To improve the user experience.
  • the dialog method of the present application further includes: updating the associated information of the knowledge points corresponding to the unique answer, specifically:
  • the points of the dialogue system corresponding to the knowledge points corresponding to the unique answer are updated.
  • the real-time update of the number of adopted knowledge points and the corresponding scoring of the dialogue system according to the actual use situation ensure the accuracy of the data, thereby providing a reliable data source for evaluating the priority level of the knowledge points, which helps to improve Recommend the accuracy of answers for users.
  • an embodiment of the present application further provides a knowledge sharing device 400 between dialogue systems, which is used for any kind of dialogue system for implementing knowledge sharing in the dialogue system, including:
  • An information receiving program module 410 is configured to receive a knowledge sharing request sent by an external dialogue system, where the knowledge sharing request includes at least characteristic information of knowledge points to be shared of the external dialogue system;
  • An information parsing program module 420 configured to parse the knowledge sharing request to determine characteristic information of the knowledge points to be shared;
  • the library update program module 430 is configured to add feature information of the knowledge points to be shared to the knowledge base of the first dialog system to form a shared knowledge base.
  • an embodiment of the present application further provides a dialogue device 500 for a first dialogue system.
  • the first dialogue system shares knowledge points of at least one external dialogue system.
  • the knowledge base and knowledge points of at least one external dialogue system shared by the first dialogue system constitute a shared knowledge base, and the device includes:
  • An answer retrieval program module 520 configured to retrieve an answer corresponding to the question in the shared knowledge base according to the analysis result
  • a first answer program module 530 configured to notify the user that there is no answer corresponding to the question when the retrieval result indicates that there is no answer in the shared knowledge base;
  • a second answer program module 540 configured to present the unique answer to the user when the retrieval result indicates that the unique answer exists in the shared knowledge base
  • the third answering program module 550 is configured to determine a priority level of each of the multiple answers when the retrieval result indicates that there are multiple answers in the shared knowledge base, and present the answer with the highest priority to the user.
  • the answer retrieval program module 520 includes:
  • a similarity determining unit 521 is configured to determine at least one similarity between the at least one external dialogue system and the first dialogue system respectively according to a domain corresponding to the first dialogue system and the at least one external dialogue system, respectively. degree;
  • the first priority determining unit 522 is configured to determine a priority level of a knowledge point in the shared knowledge base that belongs to the at least one external dialogue system. The greater the value of the at least one similarity is, the higher the degree of similarity is. The higher the knowledge point priority level of the corresponding knowledge point of the external system;
  • a second priority determining unit 523 configured to determine that a priority level of a knowledge point in the shared knowledge base corresponding to a knowledge point of the first dialog system is higher than a knowledge point of the knowledge point of the at least one external system Level
  • the answer retrieval unit 524 is configured to retrieve an answer corresponding to the question in the shared knowledge base in the order of the priority level of the knowledge points from high to low.
  • an embodiment of the present application further provides a computer program product.
  • the computer program product includes a computer program stored on a non-volatile computer-readable storage medium.
  • the computer program includes program instructions. When the program instructions are executed by a computer, the computer is caused to execute the knowledge sharing method (or dialogue method) between any of the above dialogue systems.
  • an embodiment of the present application further provides an electronic device, including: at least one processor, and a memory communicatively connected to the at least one processor, wherein the memory stores a memory that can be used by the at least one processor. Instructions executed by a processor, the instructions being executed by the at least one processor, so that the at least one processor can execute a knowledge sharing method (or a dialogue method) between dialogue systems.
  • the embodiment of the present application further provides a storage medium on which a computer program is stored, characterized in that, when the program is executed by a processor, the computer is caused to execute the knowledge between any of the above-mentioned dialog systems. Sharing method (or conversation method).
  • the device in the embodiment of the present application can be used to execute the method in the embodiment of the present application, and the technical effect achieved by the method in the embodiment of the present application can be achieved accordingly, and details are not described herein again.
  • related functional modules may be implemented by a hardware processor.
  • FIG. 7 is a schematic diagram of a hardware structure of an electronic device that executes a knowledge sharing method (or a dialogue method) between dialog systems according to another embodiment of the present application. As shown in FIG. 7, the device includes:
  • One or more processors 710 and a memory 720 are taken as an example in FIG. 7.
  • the apparatus for performing the knowledge sharing method (or the dialogue method) between the dialogue systems may further include an input device 730 and an output device 740.
  • the processor 710, the memory 720, the input device 730, and the output device 740 may be connected through a bus or in other manners. In FIG. 7, the connection through the bus is taken as an example.
  • the memory 720 is a non-volatile computer-readable storage medium, and can be used to store non-volatile software programs, non-volatile computer executable programs, and modules, such as a knowledge sharing method between dialog systems in the embodiments of the present application. (Or dialog method) corresponding program instructions / modules.
  • the processor 710 executes various functional applications and data processing of the server by running the non-volatile software programs, instructions, and modules stored in the memory 720, that is, the knowledge sharing method (or dialogue) between the dialogue systems of the method embodiments described above is implemented. method).
  • the memory 720 may include a storage program area and a storage data area, where the storage program area may store an operating system and an application program required for at least one function; the storage data area may store information stored in a knowledge sharing device (or a dialogue device) between dialogue systems. Use the created data, etc.
  • the memory 720 may include a high-speed random access memory, and may further include a non-volatile memory, such as at least one magnetic disk storage device, a flash memory device, or other non-volatile solid-state storage device.
  • the memory 720 may optionally include a memory remotely disposed with respect to the processor 710, and these remote memories may be connected to a knowledge sharing device (or a dialogue device) between dialogue systems through a network. Examples of the above network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.
  • the input device 730 may receive inputted numeric or character information and generate signals related to user settings and function control of a knowledge sharing device (or a dialog device) between dialog systems.
  • the output device 740 may include a display device such as a display screen.
  • the one or more modules are stored in the memory 720, and when executed by the one or more processors 710, execute a knowledge sharing method (or a dialog method) between dialog systems in any of the above method embodiments.
  • the above product can execute the method provided in the embodiment of the present application, and has corresponding function modules and beneficial effects of executing the method.
  • the above product can execute the method provided in the embodiment of the present application, and has corresponding function modules and beneficial effects of executing the method.
  • the electronic devices in the embodiments of the present application exist in various forms, including but not limited to:
  • Mobile communication equipment This type of equipment is characterized by mobile communication functions, and its main goal is to provide voice and data communication.
  • Such terminals include: smart phones (such as iPhone), multimedia phones, feature phones, and low-end phones.
  • Ultra-mobile personal computer equipment This type of equipment belongs to the category of personal computers, has computing and processing functions, and generally has mobile Internet access characteristics.
  • Such terminals include: PDA, MID and UMPC devices, such as iPad.
  • Portable entertainment equipment This type of equipment can display and play multimedia content. These devices include audio and video players (such as iPods), handheld game consoles, e-books, as well as smart toys and portable car navigation devices.
  • the composition of the server includes a processor, hard disk, memory, and system bus.
  • the server is similar to a general-purpose computer architecture. , Reliability, security, scalability, manageability and other aspects of higher requirements.
  • the device embodiments described above are only schematic, and the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, may be located One place, or it can be distributed across multiple network elements. Some or all of the modules may be selected according to actual needs to achieve the objective of the solution of this embodiment.

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • General Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Artificial Intelligence (AREA)
  • Computational Linguistics (AREA)
  • General Physics & Mathematics (AREA)
  • Evolutionary Computation (AREA)
  • Software Systems (AREA)
  • Mathematical Physics (AREA)
  • Computing Systems (AREA)
  • Data Mining & Analysis (AREA)
  • General Health & Medical Sciences (AREA)
  • Health & Medical Sciences (AREA)
  • Audiology, Speech & Language Pathology (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Probability & Statistics with Applications (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
  • Machine Translation (AREA)

Abstract

本申请公开一种对话系统间的知识分享方法,包括:第一对话系统接收外部对话系统所发送的知识分享请求,所述知识分享请求至少包括所述外部对话系统的待分享知识点的特征信息;解析所述知识分享请求以确定所述待分享知识点的特征信息;将所述待分享知识点的特征信息增加至所述第一对话系统的知识库中以形成分享知识库。本申请实施例通过对话系统之间分享知识点的方式实现了不同对话系统之间的知识的共享,在成本最小化的情况下,最大程度的满足了用户对于对话机器人的跨领域的对话需求。

Description

对话系统间的知识分享方法、对话方法及装置
本申请要求在2018年5月30日提交中国专利局、申请号为201810536554.1、发明名称为“对话系统间的知识分享方法、对话方法及装置”的中国专利申请的优先权,该申请的全部内容通过引用结合在本申请中。
技术领域
本申请涉及人工智能技术领域,尤其涉及一种对话系统间的知识分享方法、对话方法、装置、电子设备及存储介质。
背景技术
对话系统可以是微信公众号、实体机器人、网页对话机器人、电脑、手机或其它移动设备的应用程序,其中上述每一种形式的对话系统都可能用在不同的业务领域,例如,网页对话机器人,可以有用于商务合作方向的网页对话机器人、用于技术开发方向的网页对话机器人、用于技术咨询方向的网页对话机器人等。不同业务方向的网页对话机器人所对应的后台知识库中所包含的知识点都是围绕各自的业务方向及领域来组织的,这样可以使得到的网页对话机器人成为所属业务方向的“专家”,但是所存在的缺陷是当用户在与某一网页对话机器人的聊天过程中提出的问题超出了当前某一网页对话机器人的知识库所涵盖的知识点范畴时,就无法回答该问题,或者出现所答非所问的情况,影响用户体验。
现有技术中为解决该问题,整合了尽可能多的业务方向的知识点,构成了一个综合知识库,并以此训练构建了可以回答尽可能多的业务方向的问题的超级网页对话机器人。
然而,发明人在实现本申请的过程中发现,现有技术中实际上无形中增加了构建网页对话机器人的成本,也造成了资源浪费。原因在于实际应用中,用户在与某一业务方向网页对话机器人进行对话的过程中是很少提及超出该业务方向的知识点范畴的问题的,因此为了避免这种小概率事件 的发生去整合所有可能的业务领域的知识点,然而利用率却不高,造成了资源的浪费。此外,由于需要更大量的数据进行训练,一方面加大了训练难度,另一方面增加了训练成本。
发明内容
本申请实施例提供一种对话系统间的知识分享方法、对话方法、装置、电子设备及存储介质,用于至少解决上述技术问题之一。
第一方面,本申请实施例提供一种对话系统间的知识分享方法,应用于电子设备,所述方法包括:
第一对话系统接收外部对话系统所发送的知识分享请求,所述知识分享请求至少包括所述外部对话系统的待分享知识点的特征信息;
解析所述知识分享请求以确定所述待分享知识点的特征信息;
将所述待分享知识点的特征信息增加至所述第一对话系统的知识库中以形成分享知识库。
第二方面,本申请实施例提供一种对话方法,用于第一对话系统,所述第一对话系统共享有至少一个外部对话系统的知识点,所述第一对话系统的知识库与所述第一对话系统所共享的至少一个外部对话系统的知识点构成共享知识库,所述方法包括:
解析用户所提交的问题;
根据所述解析结果在所述共享知识库中检索对应于所述问题的答案;
当检索结果表明所述共享知识库中不存在答案时,通知用户不存在对应于所述问题的答案;
当检索结果表明所述共享知识库中存在唯一答案时,将所述唯一答案呈现给用户;
当检索结果表明所述共享知识库中存在多个答案时,确定所述多个答案各自的优先级等级,并将优先级最高的答案呈现给用户。
第三方面,本申请实施例提供一种对话系统间的知识分享装置,包括:
信息接收程序模块,用于接收外部对话系统所发送的知识分享请求,所述知识分享请求至少包括所述外部对话系统的待分享知识点的特征信息;
信息解析程序模块,用于解析所述知识分享请求以确定所述待分享知识点的特征信息;
库更新程序模块,用于将所述待分享知识点的特征信息增加至所述第一对话系统的知识库中以形成分享知识库。
第四方面,本申请实施例提供一种对话装置,用于第一对话系统,所述第一对话系统共享有至少一个外部对话系统的知识点,所述第一对话系统的知识库与所述第一对话系统所共享的至少一个外部对话系统的知识点构成共享知识库,所述装置包括:
语义解析程序模块,用于解析用户所提交的问题;
答案检索程序模块,用于根据所述解析结果在所述共享知识库中检索对应于所述问题的答案;
第一应答程序模块,用于当检索结果表明所述共享知识库中不存在答案时,通知用户不存在对应于所述问题的答案;
第二应答程序模块,用于当检索结果表明所述共享知识库中存在唯一答案时,将所述唯一答案呈现给用户;
第三应答程序模块,用于当检索结果表明所述共享知识库中存在多个答案时,确定所述多个答案各自的优先级等级,并将优先级最高的答案呈现给用户。
第五方面,本申请实施例提供一种非易失性计算机可读存储介质,所述存储介质中存储有一个或多个包括执行指令的程序,所述执行指令能够被电子设备(包括但不限于计算机,服务器,或者网络设备等)读取并执行,以用于执行本申请上述任一项对话系统间的知识分享方法(或者对话方法)。
第六方面,提供一种电子设备,其包括:至少一个处理器,以及与所述至少一个处理器通信连接的存储器,其中,所述存储器存储有可被所述至少一个处理器执行的指令,所述指令被所述至少一个处理器执行,以使所述至少一个处理器能够执行本申请上述任一项对话系统间的知识分享方法(或者对话方法)。
第七方面,本申请实施例还提供一种计算机程序产品,所述计算机程序产品包括存储在非易失性计算机可读存储介质上的计算机程序,所述计 算机程序包括程序指令,当所述程序指令被计算机执行时,使所述计算机执行上述任一项对话系统间的知识分享方法(或者对话方法)。
本申请实施例通过对话系统之间分享知识点的方式实现了不同对话系统之间的知识的共享,在成本最小化的情况下,最大程度的满足了用户对于对话机器人的跨领域的对话需求。
附图说明
为了更清楚地说明本申请实施例的技术方案,下面将对实施例描述中所需要使用的附图作一简单地介绍,显而易见地,下面描述中的附图是本申请的一些实施例,对于本领域普通技术人员来讲,在不付出创造性劳动的前提下,还可以根据这些附图获得其他的附图。
图1为本申请的对话系统间的知识分享方法的一实施例的流程图;
图2为本申请的对话方法的一实施例的流程图;
图3为图2中的步骤S22的一实施例的流程图;
图4为本申请的对话系统间的知识分享装置的一实施例的原理框图;
图5为本申请的对话装置的一实施例的原理框图;
图6为本申请中的答案检索程序模块的一实施例的原理框图;
图7为本申请的电子设备的一实施例的结构示意图。
具体实施方式
为使本申请实施例的目的、技术方案和优点更加清楚,下面将结合本申请实施例中的附图,对本申请实施例中的技术方案进行清楚、完整地描述,显然,所描述的实施例是本申请一部分实施例,而不是全部的实施例。基于本申请中的实施例,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,都属于本申请保护的范围。
需要说明的是,在不冲突的情况下,本申请中的实施例及实施例中的特征可以相互组合。
本申请可以在由计算机执行的计算机可执行指令的一般上下文中描述,例如程序模块。一般地,程序模块包括执行特定任务或实现特定抽象数据类型的例程、程序、对象、元件、数据结构等等。也可以在分布式计 算环境中实践本申请,在这些分布式计算环境中,由通过通信网络而被连接的远程处理设备来执行任务。在分布式计算环境中,程序模块可以位于包括存储设备在内的本地和远程计算机存储介质中。
在本申请中,“模块”、“装置”、“系统”等指应用于计算机的相关实体,如硬件、硬件和软件的组合、软件或执行中的软件等。详细地说,例如,元件可以、但不限于是运行于处理器的过程、处理器、对象、可执行元件、执行线程、程序和/或计算机。还有,运行于服务器上的应用程序或脚本程序、服务器都可以是元件。一个或多个元件可在执行的过程和/或线程中,并且元件可以在一台计算机上本地化和/或分布在两台或多台计算机之间,并可以由各种计算机可读介质运行。元件还可以根据具有一个或多个数据包的信号,例如,来自一个与本地系统、分布式系统中另一元件交互的,和/或在因特网的网络通过信号与其它系统交互的数据的信号通过本地和/或远程过程来进行通信。
最后,还需要说明的是,在本文中,诸如第一和第二等之类的关系术语仅仅用来将一个实体或者操作与另一个实体或操作区分开来,而不一定要求或者暗示这些实体或操作之间存在任何这种实际的关系或者顺序。而且,术语“包括”、“包含”,不仅包括那些要素,而且还包括没有明确列出的其他要素,或者是还包括为这种过程、方法、物品或者设备所固有的要素。在没有更多限制的情况下,由语句“包括……”限定的要素,并不排除在包括所述要素的过程、方法、物品或者设备中还存在另外的相同要素。
如图1所示,本申请的实施例提供一种对话系统间的知识分享方法,应用于电子设备,该电子设备可以是具有显示屏并且支持人机交互操作的各种电子设备,包括但不限于台式计算机、膝上型便携式计算机、智能手机、平板电脑等等。当以上电子设备执行本申请实施例的知识组织方法使实现以下步骤:
S11、第一对话系统接收外部对话系统所发送的知识分享请求,所述知识分享请求至少包括所述外部对话系统的待分享知识点的特征信息。
对话系统的知识分享可以在一定群组范围内进行,也可以分享给所有人,一个公司范围的对话系统X,公司的成员有各自的对话系统A,B,C 等,对话系统A,B,C把自己的知识分享给X后,在X中的对话就可以包含X,A,B,C等角色。
本实施例中,第一对话系统可以为对话系统X,外部对话系统可以包括对话系统A、对话系统B、对话系统C。其中,
对话系统X有自己的知识集合K x={q x1,q x2,…q xn}
对话系统A有自己的知识集合K a={q a1,q a2,…q ar}
对话系统B有自己的知识集合K b={q b1,q b2,…q bs}
对话系统C有自己的知识集合K c={q c1,q c2,…q ct}
每一个知识集合由一组知识点q构成,一个知识点可以是一个问答对或一个知识图谱中的三元组。
对话系统A,对话系统B,对话系统C可以分别把自己的部分知识点作为待分享知识点分享给对话系统X。例如,对话系统A,对话系统B,对话系统C分别把自己的知识子集合P a,P b,P c分享给对话系统X,对话系统X的知识集合变成K x'={K x,P a,P b,P c},每个知识点关联自己所属的原始对话系统(例如,知识子集合P a中的知识点关联至对话系统A),这样对话系统X就可以回答所有知识集合K x'中的知识点所对应的问题。
待分享知识点的特征信息可以是对应于待分享知识点的特征向量或者ID信息,本申请对此不作限定。
根据知识的使用频率给分享该知识点的对话系统对应的奖励,奖励是该对话系统的一个分值,可以表现为该对话系统的积分、等级或对应的物质奖励。
S12、解析所述知识分享请求以确定所述待分享知识点的特征信息。
S13、将所述待分享知识点的特征信息增加至所述第一对话系统的知识库中以形成分享知识库。
本申请实施例通过对话系统之间分享知识点的方式实现了不同对话系统之间的知识的共享,使得第一对话系统能够对用户所提出的跨领域的问题进行顺利解答,在成本最小化的情况下,最大程度的满足了用户对于对话机器人的跨领域的对话需求。本申请实施例中所涉及到的对话系统可以是以微信公众号、网页对话机器人或者实体机器人、电脑、手机或其它移动设备的应用程序中的一种或者多种。
基于本申请实施例的方法可以在对话系统之间分享知识,知识的形式可以是问答对、知识库或知识图谱、用机器学习方法学习得到的对话生成算法。例如,对话系统A有自己的知识集合,对话系统A可以把自己的知识分享给对话系统B,对话系统B就可以回答对话系统A和对话系统B中所有的知识对应的问题。
分享后的知识可以以不同角色的对话进行,例如,对话系统A可以把自己的知识分享给对话系统B,这样在B中的对话就可以有对话系统A,对话系统B两个角色(还有一个终端用户角色),在实际的对话界面中显示为三个角色,其中,两个是对话系统角色,一个是用户角色。
在一些实施例中,第一对话系统还接收外部对话系统的配置信息,所述配置信息至少包括外部对话系统的头像设置信息,以用于当第一对话系统回答属于外部对话系统的知识点的问题时在对话界面显示该外部对话系统的头像。本实施例中,在进行知识点分享时,相应的将外部对话系统的配置信息分享给第一对话系统,从而能够在实际对话中当出现超出了第一对话系统知识点范畴的问题时,不仅能够从外部分享的知识点中获取答案,而且还可以以第三方的形式呈现出提供该答案的外部对话系统的头像信息,在一定程度上能够起到指导用户当前所提及的问题实际上已经是另一个业务方向的知识范畴了,这对于规范用户对知识所属领域的认知也能够起到一定的指导及规范作用。
在一些实施例中,所述知识分享请求中还包括所述待分享知识点的关联信息,用于确定所述待分享知识点的优先级等级,所述关联信息包括:所述待分享知识点被采纳的次数a,和/或
所述待分享知识点多对应的外部对话系统的积分b,和/或
所述关联信息还包括所述待分享知识点的被分享类型,所述被分享类型包括:分享给特定对话系统类型(c 1)、分享给特定对话系统群类型(c 2)和分享给所有外部对话系统类型(c 3)。
本实施例中可以基于参数a、b和c 1-c 3各自的函数值的加权和作为待分享知识点的综合得分w,以用于确定待分享知识点的优先级等级,w越大待分享知识点的优先级等级越高,例如,
w=n 1f 1(a)+n 2f 2(b)+n 3f 3(m 1*c 1+m 2*c 2+m 3*c 3),其中,n 1-n 3分别取值为0或1,m 1-m 3中一个取值为1,其它两个取值为0。或者
w的取值通过逻辑回归、神经网络等机器学习算法获得,比如
w=1/(1+e -z),其中z=n 1a+n 2b+m 1c 1+m 2c 2+m 3c 3。其中的参数n 1,n 2,m 1,m 2,m 3的取值可以通过大量用户数据来训练拟合得到。
其中,答案(即,待分享知识点)被采纳的次数:答案被是否被采纳可以通过对话过程中推荐了这个答案后后续对话是否继续或针对这个答案或答案提供者有进一步的对话行为发生来确定,每一次被采纳计数增加1,被采纳次数越多,表明该大概被用户所接收采纳的概率越大,因此,相应的确定被采纳次数越大,该答案的优先级越高。
答案被采纳的对话系统的积分,每一个对话系统根据服务于用户的历史表现都存在积分,历史表现越好,所得到的积分越多,本实施例中设定积分多,相应的属于该对话系统的答案的优先级越高。
根据知识点的来源确定知识点的优先级等级,优先级从高到低依次为:第一对话系统自己的知识点,定向分享给特定对话系统的知识点,定向分享给特定对话系统群的知识点,分享给所有外部对话系统的知识点。知识点的分享范围越大,优先级权重越低。
在本申请的实施例中,由于存在多个外部对话系统同时向一个第一对话系统分享知识点的情况,这就导致用户所提的一个问题,可能会有多个答案命中,并且该多个答案对应于多个对话系统(第一对话系统和/或外部对话系统)多个知识点,这时基于本实施例中的方法就能够对多个答案按照优先级进行排序,以实现向用户择优推荐答案的操作。具体的可以直接将优先级最高的答案呈现在对话界面中,或者将优先级较高的几个答案同时呈现在对话界面中供用户选择。设置优先级的目的主要是为了解决重复知识的使用问题,避免盲目分享,提升了用户体验。
对话系统提供分享功能相当于让原来私有的知识可以分享,降低建立对话系统的门槛,提高对话系统的能力。给对话系统提供一定的社交能力,知识分享之后当该知识点再被问到后就可以已多人对话的形式展示。
在一些实施例中本申请还提供一种对话方法,用于第一对话系统,该 对话系统可以表现为一种软件产品或者是搭载了该软件产品的硬件产品(电子设备),所述第一对话系统共享有至少一个外部对话系统的知识点,所述第一对话系统的知识库与所述第一对话系统所共享的至少一个外部对话系统的知识点构成共享知识库。
如图2所示,在一些实施例中,当第一对话系统执行本申请的对话方法时实现以下步骤:
S21、解析用户所提交的问题;具体地,对用户所提交的问题进行语义理解,确定用户问题的意图,以用于搜索相应的答案。
S22、根据所述解析结果在所述共享知识库中检索对应于所述问题的答案;
S23、当检索结果表明所述共享知识库中不存在答案时,通知用户不存在对应于所述问题的答案;
S24当检索结果表明所述共享知识库中存在唯一答案时,将所述唯一答案呈现给用户;
S25、当检索结果表明所述共享知识库中存在多个答案时,确定所述多个答案各自的优先级等级,并将优先级最高的答案呈现给用户。其中,所述确定所述多个答案各自的优先级等级包括:
确定所述多个答案中的每一个答案所对应的知识点;
获取所述每一个答案所对应的知识点的关联信息,所述关联信息包括:所述每一个答案所对应的知识点被采纳的次数,和/或
所述每一个答案所对应的知识点对应的外部对话系统的积分,和/或
所述每一个答案所对应的知识点的属性,所述属性包括本地知识点类型和被分享类型,所述被分享类型包括:分享给特定对话系统类型、分享给特定对话系统群类型和分享给所有外部对话系统类型;
根据所述关联信息确定所述多个答案各自的优先级等级。
本申请实施例的对话方法以第一对话系统与外部对话系统之间的共享知识库为基础,实现了对话系统间知识点的相互补充,使得用户使用一个对话系统即可跨业务领域获取知识点,在最小化对话系统构建成本,避免资源浪费的情况下,在一个对话系统之下就满足了用户对不同业务领域知识的需求。
如图3所示,在一些实施例中,所述根据所述解析结果在所述共享知识库中检索对应于所述问题的答案包括:
S221、根据所述第一对话系统与所述至少一个外部对话系统所分别对应的领域确定所述至少一个外部对话系统分别与所述第一对话系统之间的至少一个相似度;
S222、确定所述共享知识库中属于所述至少一个外部对话系统的知识点的知识点优先级等级,所述至少一个相似度中的值越大的相似度所对应的外部系统的知识点的知识点优先级等级越高
S223、确定所述共享知识库中的对应于所述第一对话系统的知识点的知识点优先级等级高于所述至少一个外部系统的知识点的知识点优先级等级;
S224、按照知识点优先级等级从高到低的顺序在所述共享知识库中检索对应于所述问题的答案。
通过比较外部对话系统与第一对话系统分别所述领域之间的相似度,确定了出了外部对话系统与第一对话系统的领域之间的远近关系,界定了用于搜索对应于用户所提问题的答案的标准,避免了在共享数据库中盲目搜索所导致的计算资源的浪费,同时也提高了搜索效率,能够在最短时间内为用户的提问提供参考答案,保证了这个对话过程的流畅性,提升了用户体验。
在一些实施例中,本申请的对话方法还包括:更新对应于所述唯一答案的知识点的关联信息,具体地:
更新对应于所述唯一答案的知识点被采纳的次数,和/或
更新对应于所述唯一答案的知识点对应的对话系统的积分。
通过实时的根据实际使用情况更新知识点被采纳的次数以及相应的对对话系统的评分,保证了数据的准确性,从而为评价知识点的优先级等级提供了可靠的数据源,有助于提升为用户推荐答案的准确性。
需要说明的是,对于前述的各方法实施例,为了简单描述,故将其都 表述为一系列的动作合并,但是本领域技术人员应该知悉,本申请并不受所描述的动作顺序的限制,因为依据本申请,某些步骤可以采用其他顺序或者同时进行。其次,本领域技术人员也应该知悉,说明书中所描述的实施例均属于优选实施例,所涉及的动作和模块并不一定是本申请所必须的。
在上述实施例中,对各个实施例的描述都各有侧重,某个实施例中没有详述的部分,可以参见其他实施例的相关描述。
如图4所示,本申请的实施例还提供一种对话系统间的知识分享装置400,用于任意一种对话系统以用于实现对话系统进行知识分享,包括:
信息接收程序模块410,用于接收外部对话系统所发送的知识分享请求,所述知识分享请求至少包括所述外部对话系统的待分享知识点的特征信息;
信息解析程序模块420,用于解析所述知识分享请求以确定所述待分享知识点的特征信息;
库更新程序模块430,用于将所述待分享知识点的特征信息增加至所述第一对话系统的知识库中以形成分享知识库。
如图5所示,本申请的实施例还提供一种对话装置500,用于第一对话系统,所述第一对话系统共享有至少一个外部对话系统的知识点,所述第一对话系统的知识库与所述第一对话系统所共享的至少一个外部对话系统的知识点构成共享知识库,所述装置包括:
语义解析程序模块510,用于解析用户所提交的问题;
答案检索程序模块520,用于根据所述解析结果在所述共享知识库中检索对应于所述问题的答案;
第一应答程序模块530,用于当检索结果表明所述共享知识库中不存在答案时,通知用户不存在对应于所述问题的答案;
第二应答程序模块540,用于当检索结果表明所述共享知识库中存在唯一答案时,将所述唯一答案呈现给用户;
第三应答程序模块550,用于当检索结果表明所述共享知识库中存在 多个答案时,确定所述多个答案各自的优先级等级,并将优先级最高的答案呈现给用户。
如图6所示,本申请的一些实施例中,所述答案检索程序模块520包括:
相似度确定单元521,用于根据所述第一对话系统与所述至少一个外部对话系统所分别对应的领域确定所述至少一个外部对话系统分别与所述第一对话系统之间的至少一个相似度;
第一优先级确定单元522,用于确定所述共享知识库中属于所述至少一个外部对话系统的知识点的知识点优先级等级,所述至少一个相似度中的值越大的相似度所对应的外部系统的知识点的知识点优先级等级越高;
第二优先级确定单元523,用于确定所述共享知识库中的对应于所述第一对话系统的知识点的知识点优先级等级高于所述至少一个外部系统的知识点的知识点优先级等级;
答案检索单元524,用于按照知识点优先级等级从高到低的顺序在所述共享知识库中检索对应于所述问题的答案。
在一些实施例中,本申请实施例还提供一种计算机程序产品,所述计算机程序产品包括存储在非易失性计算机可读存储介质上的计算机程序,所述计算机程序包括程序指令,当所述程序指令被计算机执行时,使所述计算机执行上述任一项对话系统间的知识分享方法(或者对话方法)。
在一些实施例中,本申请实施例还提供一种电子设备,其包括:至少一个处理器,以及与所述至少一个处理器通信连接的存储器,其中,所述存储器存储有可被所述至少一个处理器执行的指令,所述指令被所述至少一个处理器执行,以使所述至少一个处理器能够执行对话系统间的知识分享方法(或者对话方法)。
在一些实施例中,本申请实施例还提供一种存储介质,其上存储有计算机程序,其特征在于,该程序被处理器执行时,使所述计算机执行上述任一项对话系统间的知识分享方法(或者对话方法)。
上述本申请实施例中的装置可用于执行本申请实施例中的方法,并相应的达到上述本申请实施例的方法所达到的技术效果,这里不再赘述。本 申请实施例中可以通过硬件处理器(hardware processor)来实现相关功能模块。
图7是本申请另一实施例提供的执行对话系统间的知识分享方法(或者对话方法)的电子设备的硬件结构示意图,如图7所示,该设备包括:
一个或多个处理器710以及存储器720,图7中以一个处理器710为例。
执行对话系统间的知识分享方法(或者对话方法)的设备还可以包括:输入装置730和输出装置740。
处理器710、存储器720、输入装置730和输出装置740可以通过总线或者其他方式连接,图7中以通过总线连接为例。
存储器720作为一种非易失性计算机可读存储介质,可用于存储非易失性软件程序、非易失性计算机可执行程序以及模块,如本申请实施例中的对话系统间的知识分享方法(或者对话方法)对应的程序指令/模块。处理器710通过运行存储在存储器720中的非易失性软件程序、指令以及模块,从而执行服务器的各种功能应用以及数据处理,即实现上述方法实施例对话系统间的知识分享方法(或者对话方法)。
存储器720可以包括存储程序区和存储数据区,其中,存储程序区可存储操作系统、至少一个功能所需要的应用程序;存储数据区可存储根据对话系统间的知识分享装置(或者对话装置)的使用所创建的数据等。此外,存储器720可以包括高速随机存取存储器,还可以包括非易失性存储器,例如至少一个磁盘存储器件、闪存器件、或其他非易失性固态存储器件。在一些实施例中,存储器720可选包括相对于处理器710远程设置的存储器,这些远程存储器可以通过网络连接至对话系统间的知识分享装置(或者对话装置)。上述网络的实例包括但不限于互联网、企业内部网、局域网、移动通信网及其组合。
输入装置730可接收输入的数字或字符信息,以及产生与对话系统间的知识分享装置(或者对话装置)的用户设置以及功能控制有关的信号。输出装置740可包括显示屏等显示设备。
所述一个或者多个模块存储在所述存储器720中,当被所述一个或者 多个处理器710执行时,执行上述任意方法实施例中的对话系统间的知识分享方法(或者对话方法)。
上述产品可执行本申请实施例所提供的方法,具备执行方法相应的功能模块和有益效果。未在本实施例中详尽描述的技术细节,可参见本申请实施例所提供的方法。
本申请实施例的电子设备以多种形式存在,包括但不限于:
(1)移动通信设备:这类设备的特点是具备移动通信功能,并且以提供话音、数据通信为主要目标。这类终端包括:智能手机(例如iPhone)、多媒体手机、功能性手机,以及低端手机等。
(2)超移动个人计算机设备:这类设备属于个人计算机的范畴,有计算和处理功能,一般也具备移动上网特性。这类终端包括:PDA、MID和UMPC设备等,例如iPad。
(3)便携式娱乐设备:这类设备可以显示和播放多媒体内容。该类设备包括:音频、视频播放器(例如iPod),掌上游戏机,电子书,以及智能玩具和便携式车载导航设备。
(4)服务器:提供计算服务的设备,服务器的构成包括处理器、硬盘、内存、系统总线等,服务器和通用的计算机架构类似,但是由于需要提供高可靠的服务,因此在处理能力、稳定性、可靠性、安全性、可扩展性、可管理性等方面要求较高。
(5)其他具有数据交互功能的电子装置。
以上所描述的装置实施例仅仅是示意性的,其中所述作为分离部件说明的单元可以是或者也可以不是物理上分开的,作为单元显示的部件可以是或者也可以不是物理单元,即可以位于一个地方,或者也可以分布到多个网络单元上。可以根据实际的需要选择其中的部分或者全部模块来实现本实施例方案的目的。
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各实施方式可借助软件加通用硬件平台的方式来实现,当然也可以通过硬件。基于这样的理解,上述技术方案本质上或者说对相关技术做出贡献的部分可以以软件产品的形式体现出来,该计算机软件产品可以存储在计算机可读存储介质中,如ROM/RAM、磁碟、光盘等,包括若干指令用以使 得一台计算机设备(可以是个人计算机,服务器,或者网络设备等)执行各个实施例或者实施例的某些部分所述的方法。
最后应说明的是:以上实施例仅用以说明本申请的技术方案,而非对其限制;尽管参照前述实施例对本申请进行了详细的说明,本领域的普通技术人员应当理解:其依然可以对前述各实施例所记载的技术方案进行修改,或者对其中部分技术特征进行等同替换;而这些修改或者替换,并不使相应技术方案的本质脱离本申请各实施例技术方案的精神和范围。

Claims (10)

  1. 一种对话系统间的知识分享方法,包括:
    第一对话系统接收外部对话系统所发送的知识分享请求,所述知识分享请求至少包括所述外部对话系统的待分享知识点的特征信息;
    解析所述知识分享请求以确定所述待分享知识点的特征信息;
    将所述待分享知识点的特征信息增加至所述第一对话系统的知识库中以形成分享知识库。
  2. 根据权利要求1所述的方法,其中,所述知识分享请求中还包括所述待分享知识点的关联信息,用于确定所述待分享知识点的优先级等级,所述关联信息包括:
    所述待分享知识点被采纳的次数,和/或
    所述待分享知识点对应的外部对话系统的积分,和/或
    所述待分享知识点的被分享类型,所述被分享类型包括:分享给特定对话系统类型、分享给特定对话系统群类型和分享给所有外部对话系统类型。
  3. 一种对话方法,用于第一对话系统,所述第一对话系统共享有至少一个外部对话系统的知识点,所述第一对话系统的知识库与所述第一对话系统所共享的至少一个外部对话系统的知识点构成共享知识库,所述方法包括:
    解析用户所提交的问题得到语义解析结果;
    根据所述解析结果在所述共享知识库中检索对应于所述问题的答案;
    当检索结果表明所述共享知识库中不存在答案时,通知用户不存在对应于所述问题的答案;
    当检索结果表明所述共享知识库中存在唯一答案时,将所述唯一答案呈现给用户;
    当检索结果表明所述共享知识库中存在多个答案时,确定所述多个答案各自的优先级等级,并将优先级最高的答案呈现给用户。
  4. 根据权利要求3所述的方法,其中,所述根据所述解析结果在所述共享知识库中检索对应于所述问题的答案包括:
    根据所述第一对话系统与所述至少一个外部对话系统所分别对应的领域确定所述至少一个外部对话系统分别与所述第一对话系统之间的至少一个相似度;
    确定所述共享知识库中属于所述至少一个外部对话系统的知识点的知识点优先级等级,所述至少一个相似度中的值越大的相似度所对应的外部系统的知识点的知识点优先级等级越高;
    确定所述共享知识库中的对应于所述第一对话系统的知识点的知识点优先级等级高于所述至少一个外部系统的知识点的知识点优先级等级;
    按照知识点优先级等级从高到低的顺序在所述共享知识库中检索对应于所述问题的答案。
  5. 根据权利要求3所述的方法,其中,所述确定所述多个答案各自的优先级等级包括:
    确定所述多个答案中的每一个答案所对应的知识点;
    获取所述每一个答案所对应的知识点的关联信息,所述关联信息包括:所述每一个答案所对应的知识点被采纳的次数,和/或
    所述每一个答案所对应的知识点对应的外部对话系统的积分,和/或
    所述每一个答案所对应的知识点的属性,所述属性包括本地知识点类型和被分享类型,所述被分享类型包括:分享给特定对话系统类型、分享给特定对话系统群类型和分享给所有外部对话系统类型;
    根据所述关联信息确定所述多个答案各自的优先级等级。
  6. 根据权利要求5所述的方法,其中,还包括:更新对应于所述唯一答案的知识点的关联信息,具体地:
    更新对应于所述唯一答案的知识点被采纳的次数,和/或
    更新对应于所述唯一答案的知识点对应的对话系统的积分。
  7. 一种对话系统间的知识分享装置,包括:
    信息接收程序模块,用于接收外部对话系统所发送的知识分享请求,所述知识分享请求至少包括所述外部对话系统的待分享知识点的特征信息;
    信息解析程序模块,用于解析所述知识分享请求以确定所述待分享知识点的特征信息;
    库更新程序模块,用于将所述待分享知识点的特征信息增加至所述第一对话系统的知识库中以形成分享知识库。
  8. 一种对话装置,用于第一对话系统,所述第一对话系统共享有至少一个外部对话系统的知识点,所述第一对话系统的知识库与所述第一对话系统所共享的至少一个外部对话系统的知识点构成共享知识库,所述装置包括:
    语义解析程序模块,用于解析用户所提交的问题;
    答案检索程序模块,用于根据所述解析结果在所述共享知识库中检索对应于所述问题的答案;
    第一应答程序模块,用于当检索结果表明所述共享知识库中不存在答案时,通知用户不存在对应于所述问题的答案;
    第二应答程序模块,用于当检索结果表明所述共享知识库中存在唯一答案时,将所述唯一答案呈现给用户;
    第三应答程序模块,用于当检索结果表明所述共享知识库中存在多个答案时,确定所述多个答案各自的优先级等级,并将优先级最高的答案呈现给用户。
  9. 一种电子设备,其包括:至少一个处理器,以及与所述至少一个处理器通信连接的存储器,其中,所述存储器存储有可被所述至少一个处理器执行的指令,所述指令被所述至少一个处理器执行,以使所述至少一个处理器能够执行权利要求1-6中任意一项所述方法的步骤。
  10. 一种存储介质,其上存储有计算机程序,其特征在于,该程序被 处理器执行时实现权利要求1-6中任意一项所述方法的步骤。
PCT/CN2019/087808 2018-05-30 2019-05-21 对话系统间的知识分享方法、对话方法及装置 WO2019228232A1 (zh)

Priority Applications (4)

Application Number Priority Date Filing Date Title
JP2020565948A JP7108720B2 (ja) 2018-05-30 2019-05-21 対話システム間の知識共有方法、対話方法、知識共有装置、対話装置、電子機器及び記憶媒体
EP19811963.8A EP3805954A4 (en) 2018-05-30 2019-05-21 KNOWLEDGE SHARING METHOD BETWEEN DIALOGUE SYSTEMS, AND DIALOGUE METHOD AND APPARATUS
US17/057,722 US20210201167A1 (en) 2018-05-30 2019-05-21 Method of knowledge sharing among dialogue systems, dialogue method and device
JP2022113679A JP2022160464A (ja) 2018-05-30 2022-07-15 対話システム間の知識共有方法、対話方法、知識共有装置、対話装置、電子機器及び記憶媒体

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201810536554.1 2018-05-30
CN201810536554.1A CN108763494B (zh) 2018-05-30 2018-05-30 对话系统间的知识分享方法、对话方法及装置

Publications (1)

Publication Number Publication Date
WO2019228232A1 true WO2019228232A1 (zh) 2019-12-05

Family

ID=64004031

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2019/087808 WO2019228232A1 (zh) 2018-05-30 2019-05-21 对话系统间的知识分享方法、对话方法及装置

Country Status (5)

Country Link
US (1) US20210201167A1 (zh)
EP (1) EP3805954A4 (zh)
JP (2) JP7108720B2 (zh)
CN (1) CN108763494B (zh)
WO (1) WO2019228232A1 (zh)

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112131338A (zh) * 2020-06-05 2020-12-25 支付宝(杭州)信息技术有限公司 一种建立问答对的方法及装置
CN113094485A (zh) * 2021-04-08 2021-07-09 思必驰科技股份有限公司 知识库优化方法及系统
CN116187446A (zh) * 2023-05-04 2023-05-30 中国人民解放军国防科技大学 基于自适应注意力机制的知识图谱补全方法、装置和设备
CN116628160A (zh) * 2023-05-24 2023-08-22 中南大学 一种基于多知识库的任务型对话方法、系统及介质

Families Citing this family (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108763494B (zh) * 2018-05-30 2020-02-21 苏州思必驰信息科技有限公司 对话系统间的知识分享方法、对话方法及装置
CN109543019A (zh) * 2018-11-27 2019-03-29 苏州思必驰信息科技有限公司 用于车辆的对话服务方法及装置
CN110008326B (zh) * 2019-04-01 2020-11-03 苏州思必驰信息科技有限公司 会话系统中的知识摘要生成方法及系统
CN110096583B (zh) * 2019-05-09 2021-05-14 思必驰科技股份有限公司 多领域对话管理系统及其构建方法
CN114968034A (zh) * 2022-04-24 2022-08-30 上海传英信息技术有限公司 交互方法、智能终端及存储介质
CN114756693B (zh) * 2022-06-15 2022-11-01 科大讯飞(苏州)科技有限公司 对话生成方法、相关设备及可读存储介质
CN117271290B (zh) * 2023-11-20 2024-02-20 北京智谱华章科技有限公司 一种公平高效的多对话系统测评系统及方法
CN117573985B (zh) * 2024-01-16 2024-04-05 四川航天职业技术学院(四川航天高级技工学校) 一种应用于智能化在线教育系统的信息推送方法及系统

Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101437018A (zh) * 2007-11-14 2009-05-20 英业达股份有限公司 条件式知识即时分享系统及其方法
CN102667765A (zh) * 2009-09-08 2012-09-12 诺基亚公司 用于有选择地共享语义信息集合的方法和装置
CN103493045A (zh) * 2011-01-18 2014-01-01 谷歌公司 对在线问题的自动回答
CN105592082A (zh) * 2015-12-18 2016-05-18 北京微播易科技股份有限公司 内容分享系统和内容分享方法
CN106663130A (zh) * 2016-06-29 2017-05-10 深圳狗尾草智能科技有限公司 一种问答方法、系统和机器人
CN106844506A (zh) * 2016-12-27 2017-06-13 竹间智能科技(上海)有限公司 一种人工智能对话的知识检索方法及知识库自动完善方法
US20180025726A1 (en) * 2016-07-22 2018-01-25 International Business Machines Corporation Creating coordinated multi-chatbots using natural dialogues by means of knowledge base
CN107767253A (zh) * 2017-09-28 2018-03-06 山东浪潮商用系统有限公司 一种税务信息管理平台、方法及系统
CN108763494A (zh) * 2018-05-30 2018-11-06 苏州思必驰信息科技有限公司 对话系统间的知识分享方法、对话方法及装置

Family Cites Families (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH1031637A (ja) * 1996-07-17 1998-02-03 Matsushita Electric Ind Co Ltd エージェント通信装置
US6499024B1 (en) * 1999-08-24 2002-12-24 Stream International, Inc. Method and system for development of a knowledge base system
JPWO2002099695A1 (ja) 2001-05-31 2004-09-16 ソニー株式会社 情報処理装置、および情報処理方法、並びにプログラム
JP2004295484A (ja) 2003-03-27 2004-10-21 Nippon Telegr & Teleph Corp <Ntt> 問題解決方法、問題解決装置、問題解決プログラム及びそのプログラムを記録した記録媒体
JP2005108067A (ja) 2003-10-01 2005-04-21 Ok Web Inc Q&aシステム及び情報提供方法
JP2006085283A (ja) 2004-09-14 2006-03-30 Nec Corp 知識提供プログラム、知識提供装置、知識提供lanシステム、及び知識提供システム
JP5068358B2 (ja) 2010-09-21 2012-11-07 ヤフー株式会社 回答者抽出装置及びその方法
JP2012248161A (ja) 2011-05-31 2012-12-13 Oki Electric Ind Co Ltd 対話型検索システム及びプログラム、並びに、対話シナリオ生成システム及びプログラム
US8787553B2 (en) 2011-09-22 2014-07-22 At&T Intellectual Property I, L.P. Implementing a network of intelligent virtual service agents to provide personalized automated responses

Patent Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101437018A (zh) * 2007-11-14 2009-05-20 英业达股份有限公司 条件式知识即时分享系统及其方法
CN102667765A (zh) * 2009-09-08 2012-09-12 诺基亚公司 用于有选择地共享语义信息集合的方法和装置
CN103493045A (zh) * 2011-01-18 2014-01-01 谷歌公司 对在线问题的自动回答
CN105592082A (zh) * 2015-12-18 2016-05-18 北京微播易科技股份有限公司 内容分享系统和内容分享方法
CN106663130A (zh) * 2016-06-29 2017-05-10 深圳狗尾草智能科技有限公司 一种问答方法、系统和机器人
US20180025726A1 (en) * 2016-07-22 2018-01-25 International Business Machines Corporation Creating coordinated multi-chatbots using natural dialogues by means of knowledge base
CN106844506A (zh) * 2016-12-27 2017-06-13 竹间智能科技(上海)有限公司 一种人工智能对话的知识检索方法及知识库自动完善方法
CN107767253A (zh) * 2017-09-28 2018-03-06 山东浪潮商用系统有限公司 一种税务信息管理平台、方法及系统
CN108763494A (zh) * 2018-05-30 2018-11-06 苏州思必驰信息科技有限公司 对话系统间的知识分享方法、对话方法及装置

Cited By (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112131338A (zh) * 2020-06-05 2020-12-25 支付宝(杭州)信息技术有限公司 一种建立问答对的方法及装置
CN112131338B (zh) * 2020-06-05 2024-02-09 支付宝(杭州)信息技术有限公司 一种建立问答对的方法及装置
CN113094485A (zh) * 2021-04-08 2021-07-09 思必驰科技股份有限公司 知识库优化方法及系统
CN116187446A (zh) * 2023-05-04 2023-05-30 中国人民解放军国防科技大学 基于自适应注意力机制的知识图谱补全方法、装置和设备
CN116187446B (zh) * 2023-05-04 2023-07-04 中国人民解放军国防科技大学 基于自适应注意力机制的知识图谱补全方法、装置和设备
CN116628160A (zh) * 2023-05-24 2023-08-22 中南大学 一种基于多知识库的任务型对话方法、系统及介质
CN116628160B (zh) * 2023-05-24 2024-04-19 中南大学 一种基于多知识库的任务型对话方法、系统及介质

Also Published As

Publication number Publication date
CN108763494A (zh) 2018-11-06
CN108763494B (zh) 2020-02-21
JP2021519472A (ja) 2021-08-10
EP3805954A1 (en) 2021-04-14
JP2022160464A (ja) 2022-10-19
US20210201167A1 (en) 2021-07-01
EP3805954A4 (en) 2022-04-06
JP7108720B2 (ja) 2022-07-28

Similar Documents

Publication Publication Date Title
WO2019228232A1 (zh) 对话系统间的知识分享方法、对话方法及装置
JP7288446B2 (ja) ヒューマンマシン対話方法および電子デバイス
US11012569B2 (en) Insight based routing for help desk service
US10878011B2 (en) Cognitive ranking of terms used during a conversation
US20160335537A1 (en) Neural networking system and methods
CN107741976B (zh) 智能应答方法、装置、介质和电子设备
US11553048B2 (en) Method and apparatus, computer device and medium
US11347940B2 (en) Asynchronous role-playing system for dialog data collection
US10885080B2 (en) Cognitive ranking of terms used during a conversation
US10666803B2 (en) Routing during communication of help desk service
US11176472B2 (en) Chat delta prediction and cognitive opportunity system
JP7488871B2 (ja) 対話推薦方法、装置、電子機器、記憶媒体ならびにコンピュータプログラム
CN108322317A (zh) 一种账号识别关联方法及服务器
US11886816B2 (en) Bot dialog manager
US20220245158A1 (en) Recommendation and prediction engines for virtual and hybrid events
CN111461188A (zh) 一种目标业务控制方法、装置、计算设备及存储介质
US11875127B2 (en) Query response relevance determination
US20220336053A1 (en) Inherited machine learning model
US10831749B2 (en) Expert discovery using user query navigation paths
US20240087187A1 (en) Systems and methods for visualizing problem areas of an automated question answering system
US20220277326A1 (en) Performance and quality improvements for a market research platform
US20240177013A1 (en) Machine-learning model that recommends virtual experiences based on graphs and clustering
CN117708292A (zh) 应用于大语言模型的答案反馈方法、装置
CN116821295A (zh) 智能问答方法、装置、计算机设备及介质
CN115292608A (zh) 内容资源推荐方法、神经网络的训练方法、装置和设备

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: 19811963

Country of ref document: EP

Kind code of ref document: A1

ENP Entry into the national phase

Ref document number: 2020565948

Country of ref document: JP

Kind code of ref document: A

NENP Non-entry into the national phase

Ref country code: DE

ENP Entry into the national phase

Ref document number: 2019811963

Country of ref document: EP

Effective date: 20210111