WO2016201815A1 - Method and apparatus for providing online customer service - Google Patents

Method and apparatus for providing online customer service Download PDF

Info

Publication number
WO2016201815A1
WO2016201815A1 PCT/CN2015/090187 CN2015090187W WO2016201815A1 WO 2016201815 A1 WO2016201815 A1 WO 2016201815A1 CN 2015090187 W CN2015090187 W CN 2015090187W WO 2016201815 A1 WO2016201815 A1 WO 2016201815A1
Authority
WO
WIPO (PCT)
Prior art keywords
session
template
topic
semantic description
semantic
Prior art date
Application number
PCT/CN2015/090187
Other languages
French (fr)
Chinese (zh)
Inventor
刘志阳
Original Assignee
中兴通讯股份有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 中兴通讯股份有限公司 filed Critical 中兴通讯股份有限公司
Publication of WO2016201815A1 publication Critical patent/WO2016201815A1/en

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/90—Details of database functions independent of the retrieved data types
    • G06F16/903—Querying
    • G06F16/90335—Query processing
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION 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
    • G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
    • G06Q50/10—Services

Definitions

  • the present invention relates to the field of communications, and in particular, to a method and apparatus for providing online customer service.
  • the contact center usually provides an online customer service portal in the network hall, the palm office and other systems.
  • the customer enters the online customer service system and chats with the robot in real time to complete the business consulting and business processing requirements.
  • chat bot the customer often finds that after a period of chat, the robot repeatedly confirms the meaning of the problem with the customer, or fails to answer the customer's question or answer the question, resulting in poor customer experience. This happens because the existing online customer service system usually has the following problems:
  • the system uses the simple word segmentation method to extract keywords from the customer's questions, and can't analyze the meaning from the perspective of natural language, which causes the system to misunderstand the customer's problems;
  • the traditional search engine is used to search for the answer based on the keyword, resulting in a low accuracy of the answer.
  • the existing online customer service system usually adopts a question-and-answer session mode when the customer asks questions and automatically answers the system.
  • Each round of questions and answers is irrelevant or low-degree, and does not serve the client's theme according to the session scene. , resulting in incoherent semantics during the conversation process, the system automatically responds with low accuracy and affects the customer experience.
  • the invention provides a method and a device for providing an online customer service, which are used to solve the problem that the existing online customer service system does not serve the subject of the customer consultation according to the session scenario, and the answer to the customer consultation problem is low.
  • a method for providing an online customer service including: after establishing an online customer service, performing semantic analysis on the received session message to obtain a first semantic description; and selecting a preset according to the first semantic description Session template; responds to subsequent received session messages according to the session template and prompts the user with questions set in the session template.
  • the selecting a preset session template according to the first semantic description includes: searching, in a preset session template, a session template corresponding to a topic having the greatest relevance to the first semantic description.
  • the method further includes: before performing semantic analysis on the received session message, separately establishing a corresponding session template for each topic included in each service, where the session template includes a topic-related problem and Data information.
  • the session template responds to the subsequently received session message and prompts the user for the problem set in the session template.
  • the method includes: after performing semantic analysis on the subsequently received session message to obtain a second semantic description, searching for data information of a topic corresponding to the second semantic description in the session template, and prompting the user with topic data corresponding to the second semantic description, Prompts the user to the question in the order in which they are prompted in the session template.
  • the session template responds to the subsequently received session message and prompts the user for the problem set in the session template, including: performing semantic analysis on the subsequently received session message to obtain a second semantic description; if the first semantic description and the second The semantic description indicates that the subsequently received session message is different from the topic to which the previously received session message belongs, and then prompts whether to end the current session process. If the same, the message session continues to be executed according to the session template and prompts the session to the user.
  • the step of setting the problem in the template in the case of receiving the session message ending the session flow, ending the session flow, returning to the step of selecting the preset session template according to the first semantic description, otherwise continuing to perform subsequent response according to the session template response.
  • an apparatus for providing an online customer service including: a first semantic analysis module, configured to perform semantic analysis on the received session message after the online customer service is established, to obtain a first semantic description
  • the selection module is configured to select a preset session template according to the first semantic description
  • the response module is configured to respond to the subsequently received session message according to the session template and prompt the user for the problem set in the session template.
  • the selection module is specifically configured to: search for a session template corresponding to the topic with the highest relevance to the first semantic description in the preset session template.
  • the apparatus further includes: an establishing module, configured to separately establish a corresponding session template for each topic included in each service before performing semantic analysis on the received session message, where the session template includes Topic-related questions and data information.
  • the response module includes: a prompting module, configured to perform semantic analysis on the subsequently received session message, and obtain a second semantic description, and then search for the data information of the topic corresponding to the second semantic description in the session template, The user prompts the topic data corresponding to the second semantic description, and prompts the user to the question according to the prompt order of the problem defined in the session template.
  • a prompting module configured to perform semantic analysis on the subsequently received session message, and obtain a second semantic description, and then search for the data information of the topic corresponding to the second semantic description in the session template, The user prompts the topic data corresponding to the second semantic description, and prompts the user to the question according to the prompt order of the problem defined in the session template.
  • the foregoing response module includes: a second semantic analysis module, configured to perform semantic analysis on the subsequently received session message to obtain a second semantic description; and the second prompting module is configured to: if the first semantic description and the second semantic description indicate If the session message received by the subsequent session is different from the topic of the previously received session message, the message prompts whether to end the current session process. If the message is the same, the session message is subsequently sent according to the session template and the user is prompted to set the session template.
  • a second semantic analysis module configured to perform semantic analysis on the subsequently received session message to obtain a second semantic description
  • the second prompting module is configured to: if the first semantic description and the second semantic description indicate If the session message received by the subsequent session is different from the topic of the previously received session message, the message prompts whether to end the current session process. If the message is the same, the session message is subsequently sent according to the session template and the user is prompted to set the session template.
  • the step of the problem; the execution module is configured to, in the case of receiving the session message ending the session flow, ending the session flow, returning the step of selecting the preset session template according to the first semantic description, otherwise continuing to perform the follow-up according to the session template response The step of receiving the session message and prompting the user for the problem set in the session template.
  • the solution of the embodiment of the present invention selects a preset session template for executing the online customer service according to the semantic analysis result of the received session message, so that the online service can be performed around the topic that the user needs to consult, and the online customer service can be more Accurately answer questions from users and improve the user experience.
  • FIG. 1 is a flowchart of a method for providing online customer service according to Embodiment 1 of the present invention
  • FIG. 2 is a flowchart of a method for providing online customer service according to Embodiment 2 of the present invention
  • Embodiment 3 is a flow chart of a customer consulting problem through an online customer service in Embodiment 2 of the present invention
  • FIG. 4 is a structural block diagram of an apparatus for providing online customer service according to Embodiment 3 of the present invention.
  • the present invention provides a method and device for providing online customer service, the following The invention will be further described in detail by way of examples. It is understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
  • a customer's chat process is often a semantic coherent conversation process around a topic. This process generally consists of opening, topic introduction, topic expansion, and topic ending.
  • the opening statement is usually a robot throwing a greeting.
  • the topic introduction can be the customer's consultation on the product or service.
  • the client continues to ask questions according to the topic, and the chat robot responds step by step.
  • the end of the topic ends with the client or the chat robot.
  • this embodiment proposes to perform online customer service according to a preset topic template according to a topic introduced by the user.
  • FIG. 1 is a flowchart of a method for providing online customer service according to Embodiment 1 of the present invention. As shown in FIG. 1, the method includes the following processing:
  • Step 101 After establishing an online customer service, perform semantic analysis on the received session message to obtain a first semantic description.
  • the method may further include:
  • a session template is created for each topic included in each service.
  • the session template includes prompt questions related to the topic and data information.
  • the data information of the topic is used to respond to the query raised by the user.
  • the session template may be used.
  • the specified data information is set in the corresponding question, that is, when the user asks a specific question, the corresponding data information can be found in the template to solve the problem for the user.
  • Step 102 Select a preset session template according to the first semantic description.
  • selecting a session template that is set in advance according to the topic may include: searching, in a preset session template, a session template corresponding to a topic having the greatest relevance to the first semantic description.
  • Step 103 Respond to the subsequently received session message according to the session template and prompt the user for the question set in the session template.
  • the data information prompts the user with the topic data corresponding to the second semantic description, and prompts the user to the question according to the problem prompt order defined in the template. If the first semantic description and the second semantic description indicate that the subsequently received session message is different from the topic to which the previously received session message belongs, a prompt message is sent to prompt whether to end the current session process, and if the same, the session template is continued.
  • FIG. 2 is a flowchart of a method for providing online customer service according to Embodiment 2 of the present invention. As shown in FIG. 2, the method mainly includes The following steps:
  • Step S101 maintaining the semantic network data and the topic session template (that is, the session template described in Embodiment 1), and generating a corresponding data record;
  • Step S102 using a semantic analysis method to parse the meaning of the chat topic introduced by the client;
  • Step S103 searching for a session template with a high degree of relevance to the meaning of the chat topic from the configured topic session template according to the meaning of the chat topic of the client;
  • Step S104 Guide the client to expand the chat topic according to the adapted session template, and interact with the client step by step until the topic ends.
  • a plurality of topic session templates are developed for each service of the contact center (which may be manually created or automatically summarized by the system through continuous self-learning), and the chat robot performs semantic analysis on the chat topic introduced by the client to match a topic session with the highest relevance.
  • Template expand the topic according to the steps of the session template, step by step guide the customer chat until the topic ends.
  • the client can confirm with the client whether to end the topic. If the client confirms the end, the topic is terminated first, and then according to the new topic. Repeat steps C and D, otherwise continue this topic interaction.
  • step S301 the system consults the 4G mobile phone price range desired by the user. For example, "Hello! How many yuan do you want to buy a 4G phone?"
  • step S302 the customer replies to the price range that can be received. For example, "I want to buy a mobile phone of about 2,000 yuan";
  • step S303 the system searches for the model of the mobile phone that satisfies the user's condition, and displays the details of each mobile phone to the user.
  • the display format is not limited, and may be a description of a text and a description, and may be a webpage link including a detailed introduction.
  • step S304 the system continues to ask the customer which mobile phone to look at. For example, “Hello! Which mobile phone do you want to buy?";
  • step S305 after the customer knows the details, the answer is interested in one of the mobile phones. For example, "I might want a XX (where XX refers to a specific model phone)";
  • step S306 the system continues to verify some other problems for the mobile phone (for example, determining the color requirement of the customer. For example, "The mobile phone case has three colors of white, blue, and red, which color do you like?";
  • step S307 the customer gives an answer. For example, "I want to white this!”;
  • step S308 the system asks the user whether to purchase the mobile phone immediately. If the customer is willing to purchase, the process jumps to S310, otherwise the process jumps to step 309.
  • step S309 the customer decides to place an order, and the system automatically jumps to the online shopping mall.
  • the mobile phone automatically enters the customer's online shopping cart, and the user continues to perform the operation of purchasing the mobile phone.
  • step S310 the chat topic ends. Subsequently, the customer may end the entire online service or introduce other chat topics.
  • the embodiment further provides an apparatus for implementing a method for providing online customer service.
  • the apparatus includes: a semantic network module, a semantic analysis module, a semantic search engine, a topic session template configuration module, and an automatic response module.
  • the Semantic Web module is the basis of semantic analysis and semantic search. It is used to configure and maintain the contact domain domain ontology, the general knowledge ontology and the topic session scene template ontology, and the semantic relationship between ontology.
  • the semantic analysis module has two main functions. The first is to analyze the semantics of the chat theme introduced by the customer through word segmentation, feature extraction and grammar analysis, and convert it into a semantic description that can be recognized by the semantic search engine. Among them, the semantic analysis of the chat text input by the customer is transformed into a semantic description that the semantic search engine can recognize.
  • the semantic search module has two main functions. One is to search for a topical session template with strong relevance according to the semantic description of the chat subject analysis and processing. Second, during the development of the chat topic, if the customer's problem can be found by searching the knowledge ontology (such as a text description, an introduction article), the semantic search engine is used to find the matching answer, and the candidate answers are extracted and sorted. Give users feedback on the best answers and suggestions.
  • the knowledge ontology such as a text description, an introduction article
  • Topic Session Configuration Module Sets to configure and maintain several topic session templates for various services in the contact center. It can be manually edited and modified, or it can be automatically summarized by the system training method, such as system analysis and association mining of human-machine historical conversation data, extracting features, automatically constructing and perfecting topic session templates, and analyzing the system training methods. Template manuals can be added or deleted. If you want to make the degree of judgment more and more accurate, you need to constantly improve the theme session template.
  • Auto-responder module During each step of the topic unfolding process, the system responds to the customer's outgoing message, for example, the chat text.
  • the response method is not limited to one type, and may be to return a text description to display an article. Link, call an interface service and return the result of the call, or prompt the user to leave a message, transfer to a human service, and so on.
  • the embodiment provides a device for providing online customer service, and the device is used to implement the method for providing online customer service as set forth in Embodiment 1 and Embodiment 2. It should be noted that the device and the device provided in Embodiment 2 are provided. Although the structure is different, they are all used to implement the same method. Therefore, the device may include any module for providing an online customer service device in Embodiment 2, and FIG. 4 is a structural block diagram of a device for providing online customer service according to Embodiment 3 of the present invention. As shown in FIG. 4, the device 40 includes the following components:
  • the first semantic analysis module 41 is configured to perform semantic analysis on the received session message after the online customer service is established, to obtain a first semantic description
  • the selecting module 42 is configured to select a preset session template according to the first semantic description
  • the response module 43 is configured to respond to the subsequently received session message according to the session template and prompt the user for the question set in the session template.
  • the selection module 42 is specifically configured to: search for a session template corresponding to the topic with the highest relevance to the first semantic description in the preset session template.
  • the apparatus 40 may further include: an establishing module, configured to separately establish a corresponding session template for each topic included in each service before the semantic analysis of the received session message, in the session template Includes prompt questions and data information related to the topic.
  • an establishing module configured to separately establish a corresponding session template for each topic included in each service before the semantic analysis of the received session message, in the session template Includes prompt questions and data information related to the topic.
  • the response module 43 may specifically include: a prompting module, configured to: after performing semantic analysis on the subsequently received session message, and obtaining the second semantic description, searching for the topic data corresponding to the second semantic description in the session template The information prompts the user with the topic data corresponding to the second semantic description, and prompts the user according to the prompt order of the questions defined in the session template.
  • a prompting module configured to: after performing semantic analysis on the subsequently received session message, and obtaining the second semantic description, searching for the topic data corresponding to the second semantic description in the session template The information prompts the user with the topic data corresponding to the second semantic description, and prompts the user according to the prompt order of the questions defined in the session template.
  • the foregoing response module 43 may further include: a second semantic analysis module, configured to perform semantic analysis on the subsequently received session message to obtain a second semantic description; and the second prompting module is configured to: if the first semantic description and the second The semantic description indicates that the subsequently received session message is different from the topic that the previously received session message belongs to, and a prompt message is sent to prompt whether to end the current session process. If the same, the session message is continuously executed according to the session template.
  • a second semantic analysis module configured to perform semantic analysis on the subsequently received session message to obtain a second semantic description
  • the second prompting module is configured to: if the first semantic description and the second The semantic description indicates that the subsequently received session message is different from the topic that the previously received session message belongs to, and a prompt message is sent to prompt whether to end the current session process. If the same, the session message is continuously executed according to the session template.
  • the execution module is configured to, in the case of receiving the session message ending the session flow, end the session flow, and return to the step of selecting the preset session template according to the first semantic description, Otherwise, the steps of responding to the subsequently received session message according to the session template and prompting the user for the problem set in the session template are continued.
  • the contact center online customer service not only has a more accurate understanding and automatic response of the customer's chat text, but also the chat process is more like a natural communication between people, not between the person and the machine.
  • the mechanical question and answer greatly enhances the user experience of the contact center online customer service and brings a better experience to the customer.
  • a preset session template for executing an online customer service is selected according to a semantic analysis result of the received session message, so that the online service can be performed around a topic that the user needs to consult, and Online customer service can answer user consultation questions more accurately and improve user experience.

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Business, Economics & Management (AREA)
  • Databases & Information Systems (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • Data Mining & Analysis (AREA)
  • Tourism & Hospitality (AREA)
  • Computational Linguistics (AREA)
  • Health & Medical Sciences (AREA)
  • Economics (AREA)
  • General Health & Medical Sciences (AREA)
  • Human Resources & Organizations (AREA)
  • Marketing (AREA)
  • Primary Health Care (AREA)
  • Strategic Management (AREA)
  • General Business, Economics & Management (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)

Abstract

The present invention provides a method and apparatus for providing an online customer service, for use of resolving the problem of low accuracy of answering questions consulted by a customer due to the fact that the existing online customer service system does not provide, according to a scenario of a conversion, a service focusing on a subject consulted by the customer. The method comprises: after an online customer service is established, performing semantic analysis on a received conversation message, so as to obtain a first semantic description; selecting a preset conversation template according to the first semantic description; and responding to a received follow-up conversation message according to the conversation template, and reminding a user of a question set in the conversation template. The solution improves the accuracy rate of answering questions of customers.

Description

一种提供在线客服的方法及装置Method and device for providing online customer service 技术领域Technical field
本发明涉及通讯领域，特别是涉及一种提供在线客服的方法及装置。The present invention relates to the field of communications, and in particular, to a method and apparatus for providing online customer service.
背景技术Background technique
随着竞争激烈化、客户普及率饱和化，电信运营商陷入收入增长放缓、利润下滑的危机。受此影响，运营商联络中心逐渐减少人力成本，加大对自助服务的建设，引导客户更多地通过自助服务和人机交互的方式满足需求，提升服务效率和服务质量。联络中心通常在网厅、掌厅等***提供在线客服入口，客户进入在线客服***与机器人实时聊天，完成业务咨询、业务办理等需求。客户在和聊天机器人交互过程当中，经常发现聊了一段时间后，机器人要么跟客户反复确认问题的含义，要么无法回答客户的问题或者答非所问，造成客户体验不佳。出现这种情况是由于现有的在线客服***通常存在如下问题：With the fierce competition and the saturation of customer penetration, telecom operators are caught in a crisis of slower revenue growth and lower profits. Affected by this, the operator contact center gradually reduced the labor cost, increased the construction of self-service, and guided customers to meet the demand through self-service and human-computer interaction to improve service efficiency and service quality. The contact center usually provides an online customer service portal in the network hall, the palm office and other systems. The customer enters the online customer service system and chats with the robot in real time to complete the business consulting and business processing requirements. During the interaction with the chat bot, the customer often finds that after a period of chat, the robot repeatedly confirms the meaning of the problem with the customer, or fails to answer the customer's question or answer the question, resulting in poor customer experience. This happens because the existing online customer service system usually has the following problems:
***对客户的提问采用简单的分词等方法提取关键字，不能从自然语言的角度分析含义，造成***对客户的问题理解偏差；The system uses the simple word segmentation method to extract keywords from the customer's questions, and can't analyze the meaning from the perspective of natural language, which causes the system to misunderstand the customer's problems;
***回答客户的问题时，采用传统的搜索引擎根据关键字搜索答案，造成答案准确率低。When the system answers the customer's question, the traditional search engine is used to search for the answer based on the keyword, resulting in a low accuracy of the answer.
可见，现有的在线客服***在客户提问和***自动回答时，通常采用一问一答式的会话方式，每轮问答之间无关性或者关联度低，没有根据会话场景围绕客户的主题进行服务，导致会话过程中语义不连贯，***自动回答准确率低，影响客户体验。It can be seen that the existing online customer service system usually adopts a question-and-answer session mode when the customer asks questions and automatically answers the system. Each round of questions and answers is irrelevant or low-degree, and does not serve the client's theme according to the session scene. , resulting in incoherent semantics during the conversation process, the system automatically responds with low accuracy and affects the customer experience.
发明内容Summary of the invention
本发明提供一种提供在线客服的方法及装置，用以解决目前现有在线客服***没有根据会话场景围绕客户咨询的主题进行服务，导致回答客户咨询问题准确率较低的问题。The invention provides a method and a device for providing an online customer service, which are used to solve the problem that the existing online customer service system does not serve the subject of the customer consultation according to the session scenario, and the answer to the customer consultation problem is low.
根据本发明的一个实施例，提供了一种提供在线客服的方法，包括：在建立在线客服后，对接收到的会话消息进行语义分析，得到第一语义描述；根据第一语义描述选择预先设置的会话模板；按照会话模板响应后续接收到的会话消息并向用户提示会话模板中设置的问题。According to an embodiment of the present invention, a method for providing an online customer service is provided, including: after establishing an online customer service, performing semantic analysis on the received session message to obtain a first semantic description; and selecting a preset according to the first semantic description Session template; responds to subsequent received session messages according to the session template and prompts the user with questions set in the session template.
其中，根据第一语义描述选择预先设置的会话模板，包括：在预先设置的会话模板中查找与第一语义描述关联性最大的话题对应的会话模板。The selecting a preset session template according to the first semantic description includes: searching, in a preset session template, a session template corresponding to a topic having the greatest relevance to the first semantic description.
在本发明的实施例中，上述方法还包括：在对接收到的会话消息进行语义分析之前，为每种业务包含的各话题分别建立对应的会话模板，会话模板中包括与话题相关的问题以及数据信息。In an embodiment of the present invention, the method further includes: before performing semantic analysis on the received session message, separately establishing a corresponding session template for each topic included in each service, where the session template includes a topic-related problem and Data information.
其中，按照会话模板响应后续接收到的会话消息并向用户提示会话模板中设置的问题， 包括：在对后续接收到的会话消息进行语义分析得到第二语义描述之后，在会话模板中查找与第二语义描述对应的话题的数据信息，向用户提示与第二语义描述对应的话题数据，按照会话模板中定义的问题的提示顺序向用户提示问题。The session template responds to the subsequently received session message and prompts the user for the problem set in the session template. The method includes: after performing semantic analysis on the subsequently received session message to obtain a second semantic description, searching for data information of a topic corresponding to the second semantic description in the session template, and prompting the user with topic data corresponding to the second semantic description, Prompts the user to the question in the order in which they are prompted in the session template.
其中，按照会话模板响应后续接收到的会话消息并向用户提示会话模板中设置的问题，包括：对后续接收到的会话消息进行语义分析，得到第二语义描述；若第一语义描述与第二语义描述表明后续接收到的会话消息与之前接收到的会话消息所属的话题不同，则提示是否结束当前会话流程，若相同，则继续执行按照会话模板响应后续接收到的会话消息并向用户提示会话模板中设置的问题的步骤；在接收到结束会话流程的会话消息的情况下，结束会话流程，返回到根据第一语义描述选择预先设置的会话模板的步骤，否则继续执行按照会话模板响应后续接收到的会话消息并向用户提示会话模板中设置的问题的步骤。The session template responds to the subsequently received session message and prompts the user for the problem set in the session template, including: performing semantic analysis on the subsequently received session message to obtain a second semantic description; if the first semantic description and the second The semantic description indicates that the subsequently received session message is different from the topic to which the previously received session message belongs, and then prompts whether to end the current session process. If the same, the message session continues to be executed according to the session template and prompts the session to the user. The step of setting the problem in the template; in the case of receiving the session message ending the session flow, ending the session flow, returning to the step of selecting the preset session template according to the first semantic description, otherwise continuing to perform subsequent response according to the session template response The steps to the session message and prompt the user for the problem set in the session template.
根据本发明的另一个实施例，提供了一种提供在线客服的装置，包括：第一语义分析模块，设置为在建立在线客服后，对接收到的会话消息进行语义分析，得到第一语义描述；选择模块，设置为根据第一语义描述选择预先设置的会话模板；响应模块，设置为按照会话模板响应后续接收到的会话消息并向用户提示会话模板中设置的问题。According to another embodiment of the present invention, an apparatus for providing an online customer service is provided, including: a first semantic analysis module, configured to perform semantic analysis on the received session message after the online customer service is established, to obtain a first semantic description The selection module is configured to select a preset session template according to the first semantic description; the response module is configured to respond to the subsequently received session message according to the session template and prompt the user for the problem set in the session template.
其中，上述选择模块具体设置为：在预先设置的会话模板中查找与第一语义描述关联性最大的话题对应的会话模板。The selection module is specifically configured to: search for a session template corresponding to the topic with the highest relevance to the first semantic description in the preset session template.
在本发明的实施例中，上述装置还包括：建立模块，设置为在对接收到的会话消息进行语义分析之前，为每种业务包含的各话题分别建立对应的会话模板，会话模板中包括与话题相关的问题以及数据信息。In an embodiment of the present invention, the apparatus further includes: an establishing module, configured to separately establish a corresponding session template for each topic included in each service before performing semantic analysis on the received session message, where the session template includes Topic-related questions and data information.
其中，上述响应模块，包括：提示模块，设置为在对后续接收到的会话消息进行语义分析，得到第二语义描述之后，在会话模板中查找与第二语义描述对应的话题的数据信息，向用户提示与第二语义描述对应的话题数据，按照会话模板中定义的问题的提示顺序向用户提示问题。The response module includes: a prompting module, configured to perform semantic analysis on the subsequently received session message, and obtain a second semantic description, and then search for the data information of the topic corresponding to the second semantic description in the session template, The user prompts the topic data corresponding to the second semantic description, and prompts the user to the question according to the prompt order of the problem defined in the session template.
其中，上述响应模块包括：第二语义分析模块，设置为对后续接收到的会话消息进行语义分析，得到第二语义描述；第二提示模块，设置为若第一语义描述与第二语义描述表明后续接收到的会话消息与之前接收到的会话消息所属的话题不同，则提示是否结束当前会话流程，若相同，则继续执行按照会话模板响应后续接收到的会话消息并向用户提示会话模板中设置的问题的步骤；执行模块，设置为在接收到结束会话流程的会话消息的情况下，结束会话流程，返回根据第一语义描述选择预先设置的会话模板的步骤，否则继续执行按照会话模板响应后续接收到的会话消息并向用户提示会话模板中设置的问题的步骤。The foregoing response module includes: a second semantic analysis module, configured to perform semantic analysis on the subsequently received session message to obtain a second semantic description; and the second prompting module is configured to: if the first semantic description and the second semantic description indicate If the session message received by the subsequent session is different from the topic of the previously received session message, the message prompts whether to end the current session process. If the message is the same, the session message is subsequently sent according to the session template and the user is prompted to set the session template. The step of the problem; the execution module is configured to, in the case of receiving the session message ending the session flow, ending the session flow, returning the step of selecting the preset session template according to the first semantic description, otherwise continuing to perform the follow-up according to the session template response The step of receiving the session message and prompting the user for the problem set in the session template.
本发明实施例的方案，按照对接收到的会话消息的语义分析结果来选择用于执行在线客服的预先设置的会话模板，使得在线服务可以围绕用户所需咨询的话题进行，且在线客服可以更加准确的回答用户咨询的问题，提高了用户体验。The solution of the embodiment of the present invention selects a preset session template for executing the online customer service according to the semantic analysis result of the received session message, so that the online service can be performed around the topic that the user needs to consult, and the online customer service can be more Accurately answer questions from users and improve the user experience.
附图说明 DRAWINGS
图1是本发明实施例1的提供在线客服的方法的流程图；1 is a flowchart of a method for providing online customer service according to Embodiment 1 of the present invention;
图2是本发明实施例2的提供在线客服的方法的流程图；2 is a flowchart of a method for providing online customer service according to Embodiment 2 of the present invention;
图3是本发明实施例2中客户通过在线客服咨询问题的流程图；3 is a flow chart of a customer consulting problem through an online customer service in Embodiment 2 of the present invention;
图4是本发明实施例3的提供在线客服的装置的结构框图。4 is a structural block diagram of an apparatus for providing online customer service according to Embodiment 3 of the present invention.
具体实施方式detailed description
为了解决现有技术中的在线客服方式并非根据围绕客户咨询的主题进行服务，导致回答客户咨询问题准确率较低的问题，本发明提供了一种提供在线客服的方法及装置，以下结合附图以及实施例，对本发明进行进一步详细说明。应当理解，此处所描述的具体实施例仅仅用以解释本发明，并不限定本发明。In order to solve the problem that the online customer service mode in the prior art is not based on the subject of customer consultation, resulting in a low accuracy rate of answering the customer consultation question, the present invention provides a method and device for providing online customer service, the following The invention will be further described in detail by way of examples. It is understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
实施例1Example 1
在线客服时，客户的一次聊天过程往往是围绕某一主题展开的语义连贯的会话过程，这个过程一般由开场白、话题引入、话题展开、话题结束几个环节构成。开场白通常是机器人抛出问候语，话题引入可以是客户针对产品或服务的咨询，话题展开阶段客户继续根据主题发问，聊天机器人一步步做出相应的响应，话题结束阶段以客户或者聊天机器人一方结束服务，基于此，本实施例提出了按照用户引入的话题依照预先设置的话题模板来进行在线客户服务。In online customer service, a customer's chat process is often a semantic coherent conversation process around a topic. This process generally consists of opening, topic introduction, topic expansion, and topic ending. The opening statement is usually a robot throwing a greeting. The topic introduction can be the customer's consultation on the product or service. During the topic development phase, the client continues to ask questions according to the topic, and the chat robot responds step by step. The end of the topic ends with the client or the chat robot. Service, based on this, this embodiment proposes to perform online customer service according to a preset topic template according to a topic introduced by the user.
图1是本发明实施例1提供在线客服的方法的流程图，如图1所示，该方法包括如下处理：1 is a flowchart of a method for providing online customer service according to Embodiment 1 of the present invention. As shown in FIG. 1, the method includes the following processing:
步骤101：在建立在线客服后，对接收到的会话消息进行语义分析，得到第一语义描述；Step 101: After establishing an online customer service, perform semantic analysis on the received session message to obtain a first semantic description.
在该步骤101之前，该方法还可以包括：Before the step 101, the method may further include:
为每种业务包含的各话题分别建立对应的会话模板，会话模板中包括与话题相关的提示问题以及数据信息，其中，话题的数据信息用于响应用户提出的疑问，具体的，可以在会话模板中设置指定的数据信息对应指定的疑问，即，在用户提出指定疑问时，即可在模板中查找相应的数据信息来为用户解疑。A session template is created for each topic included in each service. The session template includes prompt questions related to the topic and data information. The data information of the topic is used to respond to the query raised by the user. Specifically, the session template may be used. The specified data information is set in the corresponding question, that is, when the user asks a specific question, the corresponding data information can be found in the template to solve the problem for the user.
步骤102：根据第一语义描述选择预先设置的会话模板；Step 102: Select a preset session template according to the first semantic description.
优选地，根据话题选择预先设置的会话模板，具体可以包括：在预先设置的会话模板中查找与第一语义描述关联性最大的话题对应的会话模板。Preferably, selecting a session template that is set in advance according to the topic may include: searching, in a preset session template, a session template corresponding to a topic having the greatest relevance to the first semantic description.
步骤103：按照会话模板响应后续接收到的会话消息并向用户提示会话模板中设置的问题。Step 103: Respond to the subsequently received session message according to the session template and prompt the user for the question set in the session template.
在对后续接收到的会话消息进行语义分析，得到第二语义描述；在对后续接收到的会话消息进行语义分析，得到第二语义描述之后，在会话模板中查找与第二语义描述对应的话题 数据信息，向用户提示与第二语义描述对应的话题数据，按照模板中定义的问题提示顺序向用户提示问题。若第一语义描述与第二语义描述表明后续接收到的会话消息与之前接收到的会话消息所属的话题不同，则发出提示消息，提示是否结束当前会话流程，若相同，则继续执行按照会话模板响应后续接收到的会话消息并向用户提示会话模板中设置的问题的步骤；在接收到结束会话流程的会话消息的情况下，结束会话流程，返回到根据第一语义描述选择预先设置的会话模板的步骤，否则继续执行按照会话模板响应后续接收到的会话消息并向用户提示会话模板中设置的问题的步骤。Performing semantic analysis on the subsequently received session message to obtain a second semantic description; after performing semantic analysis on the subsequently received session message to obtain a second semantic description, searching for a topic corresponding to the second semantic description in the session template The data information prompts the user with the topic data corresponding to the second semantic description, and prompts the user to the question according to the problem prompt order defined in the template. If the first semantic description and the second semantic description indicate that the subsequently received session message is different from the topic to which the previously received session message belongs, a prompt message is sent to prompt whether to end the current session process, and if the same, the session template is continued. Responding to the subsequently received session message and prompting the user for the problem set in the session template; in the case of receiving the session message ending the session flow, ending the session flow, returning to selecting the pre-set session template according to the first semantic description The step of otherwise continuing to perform the step of responding to the subsequently received session message in accordance with the session template and prompting the user for the problem set in the session template.
实施例2Example 2
以下通过公开更多的技术细节来对本发明提供的在线客服的处理方法进行进一步说明，图2是本发明实施例2的提供在线客服的方法的流程图，如图2所示，该方法主要包括如下步骤：The following is a further description of the online customer service processing method provided by the present invention by disclosing more technical details. FIG. 2 is a flowchart of a method for providing online customer service according to Embodiment 2 of the present invention. As shown in FIG. 2, the method mainly includes The following steps:
步骤S101，维护语义网数据和主题会话模板(即实施例1中记载的会话模板)，生成相应的数据记录；Step S101, maintaining the semantic network data and the topic session template (that is, the session template described in Embodiment 1), and generating a corresponding data record;
步骤S102，采用语义分析方法解析客户引入的聊天话题含义；Step S102, using a semantic analysis method to parse the meaning of the chat topic introduced by the client;
步骤S103，根据客户的聊天话题含义，从配置的主题会话模板中，寻找出与聊天话题含义关联度高的会话模板；Step S103, searching for a session template with a high degree of relevance to the meaning of the chat topic from the configured topic session template according to the meaning of the chat topic of the client;
步骤S104，根据适配的会话模板引导客户展开聊天话题，一步一步与客户聊天交互，直至话题结束。Step S104: Guide the client to expand the chat topic according to the adapted session template, and interact with the client step by step until the topic ends.
具体地，针对联络中心每种业务制定若干主题会话模板(可以是人工方式创建或者***通过不断自我学习自动总结)，聊天机器人对客户引入的聊天话题进行语义分析，匹配一个关联性最高的主题会话模板，将话题展开按照会话模板的步骤，一步一步引导客户聊天，直至话题结束。Specifically, a plurality of topic session templates are developed for each service of the contact center (which may be manually created or automatically summarized by the system through continuous self-learning), and the chat robot performs semantic analysis on the chat topic introduced by the client to match a topic session with the highest relevance. Template, expand the topic according to the steps of the session template, step by step guide the customer chat until the topic ends.
特别地，步骤D中如果交互过程未结束时，客户试图尝试切换到新的聊天话题时，可以和客户确认是否结束本次话题，如果客户确认结束，则先结束本次话题，再根据新话题重复步骤C和步骤D，否则继续本次话题交互。In particular, if the client tries to switch to a new chat topic in step D, the client can confirm with the client whether to end the topic. If the client confirms the end, the topic is terminated first, and then according to the new topic. Repeat steps C and D, otherwise continue this topic interaction.
以下以客户询问和购买4G手机为例，对本实施例的在线提供在线客服的方法进行举例说明书：The following is an example of a method for online online customer service in this embodiment by taking a customer inquiry and purchasing a 4G mobile phone as an example:
当对接收到的客户输入的消息进行语义分析确认客户有购买4G手机的意向，并且搜索到匹配的主题会话模型后，话题展开详细过程如图3所示：When the semantic analysis of the received customer input message confirms that the customer has the intention to purchase the 4G mobile phone, and searches for the matching topic session model, the topic expansion detailed process is as shown in FIG. 3:
步骤S301，***咨询用户期望的4G手机价格范围。比如，“您好！请问您想买多少元左右的4G手机？”；In step S301, the system consults the 4G mobile phone price range desired by the user. For example, "Hello! How many yuan do you want to buy a 4G phone?"
步骤S302，客户答复可接收的价格区间。比如，“我想买2千元左右的手机”； In step S302, the customer replies to the price range that can be received. For example, "I want to buy a mobile phone of about 2,000 yuan";
步骤S303，***搜索出满足用户条件的手机型号，并将各个手机详情展示给用户，展示形式不限，可以是一段图文并茂的描述，可以是包含详情介绍的网页链接等。In step S303, the system searches for the model of the mobile phone that satisfies the user's condition, and displays the details of each mobile phone to the user. The display format is not limited, and may be a description of a text and a description, and may be a webpage link including a detailed introduction.
步骤S304，***继续询问客户看中哪一款手机。比如，“您好！请问您想购买以上哪一款手机？”；In step S304, the system continues to ask the customer which mobile phone to look at. For example, "Hello! Which mobile phone do you want to buy?";
步骤S305，客户了解详情后，回答对其中的某款手机感兴趣。比如，“我可能想要一款XX(这里的XX代指具体的某型号手机)”；In step S305, after the customer knows the details, the answer is interested in one of the mobile phones. For example, "I might want a XX (where XX refers to a specific model phone)";
步骤S306，***针对这款手机继续核实一些其他问题(例如，确定客户的颜色需求。比如，“这款手机外壳有白色、蓝色、红色三种颜色，您喜欢哪一种颜色？”；In step S306, the system continues to verify some other problems for the mobile phone (for example, determining the color requirement of the customer. For example, "The mobile phone case has three colors of white, blue, and red, which color do you like?";
步骤S307，客户给出答案。比如，“就要白色这款吧！”；In step S307, the customer gives an answer. For example, "I want to white this!";
步骤S308，***询问用户是否立即购买手机，如果客户愿意购买，则跳转到S310，否则跳转到步骤309。In step S308, the system asks the user whether to purchase the mobile phone immediately. If the customer is willing to purchase, the process jumps to S310, otherwise the process jumps to step 309.
步骤S309，客户决定下单，***自动跳转到网上商城，此款手机自动进入客户的网上购物车，用户继续执行购买手机操作。In step S309, the customer decides to place an order, and the system automatically jumps to the online shopping mall. The mobile phone automatically enters the customer's online shopping cart, and the user continues to perform the operation of purchasing the mobile phone.
步骤S310，本次聊天话题结束。后续，客户可能结束整个在线服务，或者引入其他聊天话题。In step S310, the chat topic ends. Subsequently, the customer may end the entire online service or introduce other chat topics.
本实施例还提供了用于实现提供在线客服的方法的装置，如图3所示，该装置包括：语义网模块，语义分析模块，语义搜索引擎，主题会话模板配置模块，自动应答模块。The embodiment further provides an apparatus for implementing a method for providing online customer service. As shown in FIG. 3, the apparatus includes: a semantic network module, a semantic analysis module, a semantic search engine, a topic session template configuration module, and an automatic response module.
语义网模块，是语义分析和语义搜索的依据，用于配置和维护联络中心领域知识本体、通用知识本体和主题会话场景模板本体，以及本体之间的语义关系。The Semantic Web module is the basis of semantic analysis and semantic search. It is used to configure and maintain the contact domain domain ontology, the general knowledge ontology and the topic session scene template ontology, and the semantic relationship between ontology.
语义分析模块，主要有两个作用，一是对客户引入的聊天主题通过分词、特征提取、语法分析等手段进行语义分析，转化成语义搜索引擎所能识别的语义描述；二是聊天话题展开过程当中，对客户输入的聊天文本进行语义分析，转化成语义搜索引擎所能识别的语义描述。The semantic analysis module has two main functions. The first is to analyze the semantics of the chat theme introduced by the customer through word segmentation, feature extraction and grammar analysis, and convert it into a semantic description that can be recognized by the semantic search engine. Among them, the semantic analysis of the chat text input by the customer is transformed into a semantic description that the semantic search engine can recognize.
语义搜索模块，主要有两个作用，一是根据***对聊天主题分析处理后的语义描述，搜索出关联性强的主题会话模板。二是聊天话题展开过程当中，如果发现客户的问题可以通过搜索知识本体(比如一段文字说明、一篇介绍文章)给予回答，则通过语义搜索引擎寻找相匹配的答案，并抽取和排序候选答案，给用户反馈最佳答案和建议。The semantic search module has two main functions. One is to search for a topical session template with strong relevance according to the semantic description of the chat subject analysis and processing. Second, during the development of the chat topic, if the customer's problem can be found by searching the knowledge ontology (such as a text description, an introduction article), the semantic search engine is used to find the matching answer, and the candidate answers are extracted and sorted. Give users feedback on the best answers and suggestions.
主题会话配置模块：设置为配置和维护联络中心各种业务对应的若干主题会话模板。可以是人工方式编辑和修改模板，也可以是***训练的方式自动总结，比如***对人机历史会话数据分析与关联挖掘，提取特征，自动构建和完善主题会话模板，***训练方式给出的分析模板人工可以增加或删除，如果要使判定度越来越精确，则需要不断完善主题会话模板。Topic Session Configuration Module: Sets to configure and maintain several topic session templates for various services in the contact center. It can be manually edited and modified, or it can be automatically summarized by the system training method, such as system analysis and association mining of human-machine historical conversation data, extracting features, automatically constructing and perfecting topic session templates, and analyzing the system training methods. Template manuals can be added or deleted. If you want to make the degree of judgment more and more accurate, you need to constantly improve the theme session template.
自动应答模块：在话题展开过程的每个步骤过中，***对客户的发出的消息，例如，聊天文本做出相应的应答。应答方式不限于一种，可以是返回一段文字说明，显示一篇文章的 链接，调用一个接口服并返回调用结果，或者提示用户留言，转人工服务等等。Auto-responder module: During each step of the topic unfolding process, the system responds to the customer's outgoing message, for example, the chat text. The response method is not limited to one type, and may be to return a text description to display an article. Link, call an interface service and return the result of the call, or prompt the user to leave a message, transfer to a human service, and so on.
实施例3Example 3
本实施例提供了一种提供在线客服的装置，该装置用于实现上述实施例1以及实施例2中提出的提供在线客服的方法，需要说明的是，该装置与实施例2中提供的装置虽然结构不同，但均用于实现相同的方法，故，该装置可以包括上述实施例2中的提供在线客服装置的任意模块，图4是本发明实施例3的提供在线客服的装置的结构框图，如图4所示，该装置40包括如下组成部分：The embodiment provides a device for providing online customer service, and the device is used to implement the method for providing online customer service as set forth in Embodiment 1 and Embodiment 2. It should be noted that the device and the device provided in Embodiment 2 are provided. Although the structure is different, they are all used to implement the same method. Therefore, the device may include any module for providing an online customer service device in Embodiment 2, and FIG. 4 is a structural block diagram of a device for providing online customer service according to Embodiment 3 of the present invention. As shown in FIG. 4, the device 40 includes the following components:
第一语义分析模块41，设置为在建立在线客服后，对接收到的会话消息进行语义分析，得到第一语义描述；The first semantic analysis module 41 is configured to perform semantic analysis on the received session message after the online customer service is established, to obtain a first semantic description;
选择模块42，设置为根据第一语义描述选择预先设置的会话模板；The selecting module 42 is configured to select a preset session template according to the first semantic description;
响应模块43，设置为按照会话模板响应后续接收到的会话消息并向用户提示会话模板中设置的问题。The response module 43 is configured to respond to the subsequently received session message according to the session template and prompt the user for the question set in the session template.
其中，上述选择模块42具体设置为：在预先设置的会话模板中查找与第一语义描述关联性最大的话题对应的会话模板。The selection module 42 is specifically configured to: search for a session template corresponding to the topic with the highest relevance to the first semantic description in the preset session template.
在本发明的实施例中，上述装置40还可以包括：建立模块，设置为在对接收到的会话消息进行语义分析之前，为每种业务包含的各话题分别建立对应的会话模板，会话模板中包括与话题相关的提示问题以及数据信息。In an embodiment of the present invention, the apparatus 40 may further include: an establishing module, configured to separately establish a corresponding session template for each topic included in each service before the semantic analysis of the received session message, in the session template Includes prompt questions and data information related to the topic.
其中，上述响应模块43，具体可以包括：提示模块，设置为在对后续接收到的会话消息进行语义分析，得到第二语义描述之后，在会话模板中查找与第二语义描述对应的话题的数据信息，向用户提示与第二语义描述对应的话题数据，按照会话模板中定义的问题的提示顺序向用户提示问题。The response module 43 may specifically include: a prompting module, configured to: after performing semantic analysis on the subsequently received session message, and obtaining the second semantic description, searching for the topic data corresponding to the second semantic description in the session template The information prompts the user with the topic data corresponding to the second semantic description, and prompts the user according to the prompt order of the questions defined in the session template.
此外，上述响应模块43还可以包括：第二语义分析模块，设置为对后续接收到的会话消息进行语义分析，得到第二语义描述；第二提示模块，设置为若第一语义描述与第二语义描述表明后续接收到的会话消息与之前接收到的会话消息所属的话题不同，则发出提示消息，提示是否结束当前会话流程，若相同，则继续执行按照会话模板响应后续接收到的会话消息并向用户提示会话模板中设置的问题的步骤；执行模块，设置为在接收到结束会话流程的会话消息的情况下，结束会话流程，返回到根据第一语义描述选择预先设置的会话模板的步骤，否则继续执行按照会话模板响应后续接收到的会话消息并向用户提示会话模板中设置的问题的步骤。In addition, the foregoing response module 43 may further include: a second semantic analysis module, configured to perform semantic analysis on the subsequently received session message to obtain a second semantic description; and the second prompting module is configured to: if the first semantic description and the second The semantic description indicates that the subsequently received session message is different from the topic that the previously received session message belongs to, and a prompt message is sent to prompt whether to end the current session process. If the same, the session message is continuously executed according to the session template. a step of prompting the user for the question set in the session template; the execution module is configured to, in the case of receiving the session message ending the session flow, end the session flow, and return to the step of selecting the preset session template according to the first semantic description, Otherwise, the steps of responding to the subsequently received session message according to the session template and prompting the user for the problem set in the session template are continued.
本发明实施例提供的技术方案，联络中心在线客服不仅对客户的聊天文本的含义理解和自动应答更加精准，而且聊天过程更像是人与人之间的自然交流，而不是人与机器之间的机械式问答，大大提升联络中心在线客服的用户体验效果，给客户带来更好的体验。According to the technical solution provided by the embodiment of the present invention, the contact center online customer service not only has a more accurate understanding and automatic response of the customer's chat text, but also the chat process is more like a natural communication between people, not between the person and the machine. The mechanical question and answer greatly enhances the user experience of the contact center online customer service and brings a better experience to the customer.
尽管为示例目的，已经公开了本发明的优选实施例，本领域的技术人员将意识到各种改 进、增加和取代也是可能的，因此，本发明的范围应当不限于上述实施例。Although preferred embodiments of the present invention have been disclosed for illustrative purposes, those skilled in the art will recognize various modifications. Further, additions and substitutions are also possible, and therefore, the scope of the invention should not be limited to the above embodiments.
工业实用性Industrial applicability
基于本发明实施例提供的上述技术方案，按照对接收到的会话消息的语义分析结果来选择用于执行在线客服的预先设置的会话模板，使得在线服务可以围绕用户所需咨询的话题进行，且在线客服可以更加准确的回答用户咨询的问题，提高了用户体验。 Based on the foregoing technical solution provided by the embodiment of the present invention, a preset session template for executing an online customer service is selected according to a semantic analysis result of the received session message, so that the online service can be performed around a topic that the user needs to consult, and Online customer service can answer user consultation questions more accurately and improve user experience.

Claims (10)

  1. 一种提供在线客服的方法，包括：A method of providing online customer service, including:
    在建立在线客服后，对接收到的会话消息进行语义分析，得到第一语义描述；After establishing the online customer service, performing semantic analysis on the received session message to obtain a first semantic description;
    根据所述第一语义描述选择预先设置的会话模板；Selecting a preset session template according to the first semantic description;
    按照所述会话模板响应后续接收到的会话消息并向用户提示所述会话模板中设置的问题。Responding to the subsequently received session message according to the session template and prompting the user for the problem set in the session template.
  2. 根据权利要求1所述的方法，其中，所述根据所述第一语义描述选择预先设置的会话模板，包括：The method of claim 1, wherein the selecting a pre-set session template according to the first semantic description comprises:
    在预先设置的会话模板中查找与所述第一语义描述关联性最大的话题对应的会话模板。Searching for a session template corresponding to the topic having the greatest relevance to the first semantic description in a preset session template.
  3. 根据权利要求1所述的方法，其中，所述方法还包括：The method of claim 1 wherein the method further comprises:
    在对接收到的会话消息进行语义分析之前，为每种业务包含的各话题分别建立对应的会话模板，所述会话模板中包括与话题相关的问题以及数据信息。Before the semantic analysis of the received session message, a corresponding session template is established for each topic included in each service, and the session template includes a topic-related question and data information.
  4. 根据权利要求3所述的方法，其中，所述按照所述会话模板响应后续接收到的会话消息并向用户提示所述会话模板中设置的问题，包括：The method according to claim 3, wherein the responding to the subsequently received session message according to the session template and prompting the user for the problem set in the session template comprises:
    在对后续接收到的会话消息进行语义分析得到第二语义描述之后，在所述会话模板中查找与所述第二语义描述对应的话题的数据信息，向用户提示与所述第二语义描述对应的话题数据，按照所述会话模板中定义的问题的提示顺序向用户提示问题。After performing semantic analysis on the subsequently received session message to obtain a second semantic description, searching for the data information of the topic corresponding to the second semantic description in the session template, and prompting the user to correspond to the second semantic description The topic data prompts the user with the prompt in the order of the questions defined in the session template.
  5. 根据权利要求1至4任意一项所述的方法，其中，所述按照所述会话模板响应后续接收到的会话消息并向用户提示所述会话模板中设置的问题，包括：The method according to any one of claims 1 to 4, wherein the responding to the subsequently received session message according to the session template and prompting the user for the problem set in the session template comprises:
    对后续接收到的会话消息进行语义分析，得到第二语义描述；Semantic analysis of the subsequently received session message to obtain a second semantic description;
    若所述第一语义描述与所述第二语义描述表明后续接收到的会话消息与之前接收到的会话消息所属的话题不同，则提示是否结束当前会话流程，若相同，则继续执行按照所述会话模板响应后续接收到的会话消息并向用户提示所述会话模板中设置的问题的步骤；If the first semantic description and the second semantic description indicate that the subsequently received session message is different from the topic to which the previously received session message belongs, prompt whether to end the current session flow, and if they are the same, continue to perform according to the a step of the session template responding to the subsequently received session message and prompting the user for the problem set in the session template;
    在接收到结束会话流程的会话消息的情况下，结束会话流程，返回到根据所述第一语义描述选择预先设置的会话模板的步骤，否则继续执行按照所述会话模板响应后续接收到的会话消息并向用户提示所述会话模板中设置的问题的步骤。After receiving the session message ending the session flow, ending the session flow, returning to the step of selecting a preset session template according to the first semantic description, otherwise continuing to perform the subsequent received session message according to the session template response And prompt the user for the problem set in the session template.
  6. 一种提供在线客服的装置，包括：A device for providing online customer service, including:
    第一语义分析模块，设置为在建立在线客服后，对接收到的会话消息进行语义分析，得到第一语义描述； The first semantic analysis module is configured to perform semantic analysis on the received session message after the online customer service is established, to obtain a first semantic description;
    选择模块，设置为根据所述第一语义描述选择预先设置的会话模板；Selecting a module, configured to select a preset session template according to the first semantic description;
    响应模块，设置为按照所述会话模板响应后续接收到的会话消息并向用户提示所述会话模板中设置的问题。The response module is configured to respond to the subsequently received session message according to the session template and prompt the user for the question set in the session template.
  7. 根据权利要求6所述的装置，其中，所述选择模块具体设置为：The apparatus according to claim 6, wherein the selection module is specifically configured to:
    在预先设置的会话模板中查找与所述第一语义描述关联性最大的话题对应的会话模板。Searching for a session template corresponding to the topic having the greatest relevance to the first semantic description in a preset session template.
  8. 根据权利要求6所述的装置，其中，所述装置还包括：The apparatus of claim 6 wherein said apparatus further comprises:
    建立模块，设置为在对接收到的会话消息进行语义分析之前，为每种业务包含的各话题分别建立对应的会话模板，所述会话模板中包括与话题相关的问题以及数据信息。The establishing module is configured to establish a corresponding session template for each topic included in each service before performing semantic analysis on the received session message, where the session template includes a topic related question and data information.
  9. 根据权利要求8所述的装置，其中，所述响应模块，包括：The apparatus of claim 8, wherein the response module comprises:
    提示模块，设置为在对后续接收到的会话消息进行语义分析，得到第二语义描述之后，在所述会话模板中查找与所述第二语义描述对应的话题的数据信息，向用户提示与所述第二语义描述对应的话题数据，按照所述会话模板中定义的问题的提示顺序向用户提示问题。a prompting module, configured to perform semantic analysis on the subsequently received session message, obtain a second semantic description, and search for data information of the topic corresponding to the second semantic description in the session template, and prompt the user to The second semantic description corresponds to the topic data, and prompts the user according to the prompt sequence of the questions defined in the session template.
  10. 根据权利要求6至9任意一项所述的装置，其中，所述响应模块包括：The apparatus according to any one of claims 6 to 9, wherein the response module comprises:
    第二语义分析模块，设置为对后续接收到的会话消息进行语义分析，得到第二语义描述；a second semantic analysis module is configured to perform semantic analysis on the subsequently received session message to obtain a second semantic description;
    第二提示模块，设置为若所述第一语义描述与所述第二语义描述表明后续接收到的会话消息与之前接收到的会话消息所属的话题不同，则提示是否结束当前会话流程，若相同，则继续执行按照所述会话模板响应后续接收到的会话消息并向用户提示所述会话模板中设置的问题的步骤；The second prompting module is configured to: if the first semantic description and the second semantic description indicate that the subsequently received session message is different from the topic to which the previously received session message belongs, prompting whether to end the current session process, if the same And continuing to perform the step of responding to the subsequently received session message according to the session template and prompting the user for the problem set in the session template;
    执行模块，设置为在接收到结束会话流程的会话消息的情况下，结束会话流程，返回根据所述第一语义描述选择预先设置的会话模板的步骤，否则继续执行按照所述会话模板响应后续接收到的会话消息并向用户提示所述会话模板中设置的问题的步骤。 The execution module is configured to, in the case that the session message of the end session flow is received, end the session flow, and return a step of selecting a preset session template according to the first semantic description, otherwise continue to perform subsequent response according to the session template. The session message to the user and prompts the user for the problem set in the session template.
PCT/CN2015/090187 2015-06-17 2015-09-21 Method and apparatus for providing online customer service WO2016201815A1 (en)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201510336328.5 2015-06-17
CN201510336328.5A CN106326227A (en) 2015-06-17 2015-06-17 Method and device for providing online customer service

Publications (1)

Publication Number Publication Date
WO2016201815A1 true WO2016201815A1 (en) 2016-12-22

Family

ID=57544835

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2015/090187 WO2016201815A1 (en) 2015-06-17 2015-09-21 Method and apparatus for providing online customer service

Country Status (2)

Country Link
CN (1) CN106326227A (en)
WO (1) WO2016201815A1 (en)

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108415932A (en) * 2018-01-23 2018-08-17 苏州思必驰信息科技有限公司 Interactive method and electronic equipment
WO2018205224A1 (en) * 2017-05-11 2018-11-15 Microsoft Technology Licensing, Llc Assisting psychological cure in automated chatting
CN109120672A (en) * 2018-07-19 2019-01-01 广州市安服星网络科技有限公司 A kind of intelligent customer service system
CN111831875A (en) * 2019-04-11 2020-10-27 阿里巴巴集团控股有限公司 Data processing method, device, equipment and storage medium
CN113743956A (en) * 2021-08-18 2021-12-03 郑州时空隧道信息技术有限公司 System and method for on-line customer service instant chat

Families Citing this family (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
TWI673661B (en) * 2017-11-14 2019-10-01 財團法人資訊工業策進會 Method and system for automatically establishing intelligent assistant
US11190469B2 (en) * 2019-07-26 2021-11-30 Avaya Management L.P. Enhanced digital messaging
CN110516052A (en) * 2019-07-28 2019-11-29 厦门快商通科技股份有限公司 Robot question and answer exchange method, system and robot
CN111212190B (en) * 2019-12-20 2022-05-17 北京淇瑀信息科技有限公司 Conversation management method, device and system based on conversation strategy management
CN111400465B (en) * 2020-02-25 2023-04-18 支付宝(杭州)信息技术有限公司 Generation method and device of customer service robot, electronic equipment and medium
CN112185383A (en) * 2020-09-30 2021-01-05 深圳供电局有限公司 Processing method and system for customer service return visit
CN113705186B (en) * 2021-07-22 2023-12-22 上海原圈网络科技有限公司 Automatic reply method and device under message semantic analysis

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20060277165A1 (en) * 2005-06-03 2006-12-07 Fuji Xerox Co., Ltd. Question answering system, data search method, and computer program
CN102789496A (en) * 2012-07-13 2012-11-21 携程计算机技术（上海）有限公司 Method and system for implementing intelligent response
CN102868695A (en) * 2012-09-18 2013-01-09 天格科技（杭州）有限公司 Conversation tree-based intelligent online customer service method and system

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101588311A (en) * 2008-05-19 2009-11-25 北京亿企通信息技术有限公司 Method and system for submitting user questions through online customer service system
US20130211964A1 (en) * 2012-01-31 2013-08-15 Global Village Concerns Systems and methods for generation of an online store

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20060277165A1 (en) * 2005-06-03 2006-12-07 Fuji Xerox Co., Ltd. Question answering system, data search method, and computer program
CN102789496A (en) * 2012-07-13 2012-11-21 携程计算机技术（上海）有限公司 Method and system for implementing intelligent response
CN102868695A (en) * 2012-09-18 2013-01-09 天格科技（杭州）有限公司 Conversation tree-based intelligent online customer service method and system

Cited By (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2018205224A1 (en) * 2017-05-11 2018-11-15 Microsoft Technology Licensing, Llc Assisting psychological cure in automated chatting
US11551804B2 (en) 2017-05-11 2023-01-10 Microsoft Technology Licensing, Llc Assisting psychological cure in automated chatting
CN108415932A (en) * 2018-01-23 2018-08-17 苏州思必驰信息科技有限公司 Interactive method and electronic equipment
CN108415932B (en) * 2018-01-23 2023-12-22 思必驰科技股份有限公司 Man-machine conversation method and electronic equipment
CN109120672A (en) * 2018-07-19 2019-01-01 广州市安服星网络科技有限公司 A kind of intelligent customer service system
CN111831875A (en) * 2019-04-11 2020-10-27 阿里巴巴集团控股有限公司 Data processing method, device, equipment and storage medium
CN111831875B (en) * 2019-04-11 2024-05-31 阿里巴巴集团控股有限公司 Data processing method, device, equipment and storage medium
CN113743956A (en) * 2021-08-18 2021-12-03 郑州时空隧道信息技术有限公司 System and method for on-line customer service instant chat

Also Published As

Publication number Publication date
CN106326227A (en) 2017-01-11

Similar Documents

Publication Publication Date Title
WO2016201815A1 (en) Method and apparatus for providing online customer service
CN105592237B (en) A kind of method, apparatus and intelligent customer service robot of session switching
CN110263144A (en) A kind of answer acquisition methods and device
CN111192060B (en) Full-channel self-service response implementation method based on power IT service
CN101076184B (en) Method and system for realizing automatic reply
CN106202301B (en) A kind of intelligent response system based on deep learning
CN106790004B (en) Customer service auxiliary real-time prompt system based on artificial intelligence
CN110347863B (en) Speaking recommendation method and device and storage medium
WO2018201964A1 (en) Processing method for session information, server, and computer readable storage medium
CN109753560B (en) Information processing method and device of intelligent question-answering system
CN110266899A (en) The recognition methods and customer service system that client is intended to
CN110035187A (en) A method of realizing AI and operator attendance seamless switching in the phone
CN104809197A (en) On-line question and answer method based on intelligent robot
JP2013089241A (en) Multi-modal customer care system
CN107135143A (en) Many chat robots switching systems and its dialogue method
CN112913219A (en) Dynamic termination point communication channel
CN109065052A (en) A kind of speech robot people
CN113474768A (en) Intent driven contact center
Fensel et al. Why we need knowledge graphs: Applications
CN116595148A (en) Method and system for realizing dialogue flow by using large language model
CN110675875B (en) Intelligent voice conversation technology telephone experience method and device
CN113840040A (en) Man-machine cooperation outbound method, device, equipment and storage medium
CN114586323A (en) Dynamic routing of communications to different end nodes
CN117424960A (en) Intelligent voice service method, device, terminal equipment and storage medium
CN113724036A (en) Method and electronic equipment for providing question consultation service

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: 15895388

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 15895388

Country of ref document: EP

Kind code of ref document: A1