WO2018121759A1 - 智能问答方法及*** - Google Patents

智能问答方法及*** Download PDF

Info

Publication number
WO2018121759A1
WO2018121759A1 PCT/CN2017/120041 CN2017120041W WO2018121759A1 WO 2018121759 A1 WO2018121759 A1 WO 2018121759A1 CN 2017120041 W CN2017120041 W CN 2017120041W WO 2018121759 A1 WO2018121759 A1 WO 2018121759A1
Authority
WO
WIPO (PCT)
Prior art keywords
answer
question
server
local server
message
Prior art date
Application number
PCT/CN2017/120041
Other languages
English (en)
French (fr)
Inventor
熊友军
胡贵
粟德森
Original Assignee
深圳市优必选科技有限公司
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 深圳市优必选科技有限公司 filed Critical 深圳市优必选科技有限公司
Publication of WO2018121759A1 publication Critical patent/WO2018121759A1/zh

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
    • G06F16/90—Details of database functions independent of the retrieved data types
    • G06F16/903—Querying
    • G06F16/9032—Query formulation
    • G06F16/90332—Natural language query formulation or dialogue systems
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F40/00—Handling natural language data
    • G06F40/20—Natural language analysis
    • G06F40/279—Recognition of textual entities
    • G06F40/284—Lexical analysis, e.g. tokenisation or collocates

Definitions

  • the invention relates to the field of intelligent voice response, in particular to a smart question answering method and system.
  • the smart device voice service platform system usually responds manually.
  • the number of questions asked by users increases, the number of manuals also increases accordingly, and the manual customer service needs to be repeatedly answered for the same questions raised by different customers, thus wasting labor and reducing efficiency. .
  • Intelligent question and answer methods including:
  • the local server receives the problem sent by the message server and the keyword corresponding to the problem
  • the local server searches whether an answer corresponding to the question is stored, and if so, sends a corresponding answer to the message server;
  • the message server sends the problem and the keyword corresponding to the problem to the database server;
  • the database server looks up whether there is an answer corresponding to the question, and if so, sends the answer to the message server and synchronizes the question and answer to the local server.
  • the method further includes:
  • the message server sends a work answer command, and the question and the person answer answer are stored in the database server.
  • the local server receives the problem sent by the message server and the keyword before:
  • the message server receives the user question and sends the question to the word segmentation tool
  • the word segmentation tool extracts the keywords and sends the keywords to the message server.
  • the local server looks for whether the answer corresponding to the question is stored before:
  • the local server matches the received questions and keywords with the problems in the local server. If the matching is successful, the answer corresponding to the question is invoked.
  • the local server is a SOLR server.
  • the invention also relates to a smart question answering system, comprising: a message server, a local server and a database server;
  • the local server is configured to receive a problem sent by the message server and a keyword corresponding to the problem;
  • the local server is further configured to search whether an answer corresponding to the question is stored, and if yes, send a corresponding answer to the message server;
  • the message server is configured to send a question and a keyword corresponding to the problem to the database server if the answer corresponding to the problem is not stored in the local server;
  • the database server is configured to find whether an answer corresponding to the question is stored, and if yes, send the answer to the message server and synchronize the question and the answer to the local server.
  • the message server is further configured to: if the answer corresponding to the problem is not stored in the database server, send a work answer instruction, and store the answer and the answer of the person work in the database server.
  • the message server further includes a word segmentation tool
  • the message server is further configured to receive a user question and send the question to the word segmentation tool;
  • the word segmentation tool is used to extract keywords and send keywords to the message server.
  • the local server is further configured to match the received question and keyword with the problem in the local server, and if the matching is successful, the answer corresponding to the question is invoked.
  • the local server is a SOLR server.
  • the present invention has the beneficial effects of automatically answering user questions, storing problems and answers by the local server, and using the database server as a backup, having a problem update function, and reducing the pressure on the database server load.
  • FIG. 1 is a flowchart of a method for intelligent question answering according to an embodiment of the present invention
  • FIG. 2 is a flowchart of a method for intelligent question answering according to another embodiment of the present invention.
  • FIG. 3 is a flowchart of a smart question answering method according to another embodiment of the present invention.
  • FIG. 4 is a flowchart of a smart question answering method according to another embodiment of the present invention.
  • FIG. 5 is a flowchart of a smart question answering system according to an embodiment of the present invention.
  • an intelligent question and answer method provided by an embodiment of the present invention includes:
  • Step S101 The local server receives the question sent by the message server and the keyword corresponding to the question.
  • the message server receives the user's question, and sends the extracted keyword corresponding to the problem to the local server, where the common problem and the corresponding answer are pre-stored.
  • Step S102 The local server searches whether an answer corresponding to the question is stored, and if so, sends a corresponding answer to the message server.
  • the local server searches whether the answer corresponding to the question or the keyword is stored according to the question and the keyword sent by the message server, and if there is a corresponding question and answer, sends the corresponding answer to the message server, and the message service Then the answer is fed back to the user. If there is no corresponding question and answer, the local server performs the next step.
  • Step S103 If the answer corresponding to the question is not stored in the local server, the message server sends the question and the keyword corresponding to the question to the database server.
  • the local server feeds back to the message server, and the message server sends the question and the keyword corresponding to the problem to the database server, and searches for an answer from the database server.
  • Step S104 The database server searches for whether an answer corresponding to the question is stored, and if so, sends the answer to the message server and synchronizes the question and answer to the local server.
  • the database server receives the question and answer sent by the message server, searches whether the answer corresponding to the question is stored, and if so, sends the answer to the message server, and synchronizes the question and answer to the local server for use next time. If the next question is asked by the user, the answer is directly sent by the local server.
  • the local server pre-stores the questions and answers, and at the same time, the database server acts as a backup, shunting the questions asked by the users, reducing the database load pressure, and constantly updating the local server; searching through questions and keywords to automatically answer questions from users.
  • the local server is a SOLR server.
  • Join the SOLR service to improve the system response speed, while using the database service as a backup, most of the questions asked by users are answered in the SOLR service, and very few access to the database server, thereby reducing the pressure on the database load.
  • SOLR storage and database are easy to expand, support zero-work migration and expansion, thus reducing the workload when adding storage devices when the need to expand the business.
  • the SOLR memory and the database are incrementally synchronized using the asynchronous TASK JOB method, occupying lower server resources and optimizing the synchronization strategy.
  • FIG. 2 is a schematic diagram of a smart question and answer method according to another embodiment of the present invention, including:
  • Step S201 The local server receives the problem sent by the message server and the keyword corresponding to the problem;
  • Step S202 The local server searches whether an answer corresponding to the question is stored, and if yes, sends a corresponding answer to the message server;
  • Step S203 If the answer corresponding to the problem is not stored in the local server, the message server sends the question and the keyword corresponding to the question to the database server;
  • Step S204 The database server searches for whether an answer corresponding to the question is stored, and if so, sends the answer to the message server and synchronizes the question and the answer to the local server.
  • Step S205 If the answer corresponding to the question is not stored in the database server, the message server sends a work answer command, and stores the question and the person answer answer in the database server.
  • the database server searches for the answer corresponding to the question according to the question sent by the message server and the keyword corresponding to the question. If the answer corresponding to the question is not stored, the database server feeds back to the message server that the person needs to work. After the message server receives the answer, the answer is fed back to the user. On the other hand, the question and the person's work answer are updated to the database server.
  • Steps S201-S204 in this embodiment are the same as steps S101-S104 in the previous embodiment, and details are not described herein again.
  • the local server and the database server are shunted to answer questions from the user, and the questions and answers answered by the person are stored in the database server, and the local server and the database server are continuously updated to ensure the response rate of the entire system.
  • FIG. 3 is a schematic diagram of a smart question and answer method according to another embodiment of the present invention, including:
  • Step S301 The message server receives the user question and sends the question to the word segmentation tool
  • the word segmentation tool extracts the keywords and sends the keywords to the message server.
  • the message server receives the user question and sends the question to the word segmentation tool, and the word segmentation tool advances the keyword according to the grammar and logic, and sends the keyword to the message server.
  • Step S302 The local server receives the problem sent by the message server and the keyword corresponding to the problem;
  • Step S303 The local server searches whether an answer corresponding to the question is stored, and if yes, sends a corresponding answer to the message server;
  • Step S304 If the answer corresponding to the problem is not stored in the local server, the message server sends the question and the keyword corresponding to the question to the database server;
  • Step S305 The database server searches whether the answer corresponding to the question is stored, and if so, sends the answer to the message server and synchronizes the question and the answer to the local server.
  • Step S306 If the answer corresponding to the question is not stored in the database server, the message server sends a work answer command, and stores the question and the person answer answer to the database server.
  • Steps S302-S206 in this embodiment are the same as steps S201-S205 in the previous embodiment, and details are not described herein again.
  • FIG. 4 is a flowchart of a method for intelligent question answering according to another embodiment of the present invention, including:
  • Step S401 The message server receives the user question and sends the question to the word segmentation tool
  • the word segmentation tool extracts the keywords and sends the keywords to the message server.
  • Step S402 The local server receives the problem sent by the message server and the keyword corresponding to the problem;
  • Step S403 The local server matches the received question and keyword with the problem in the local server, and if the matching is successful, the answer corresponding to the question is invoked.
  • the local server matches the received problem and the keyword with the problem in the local server, and if the matching is successful, if the matching is successful, the local server is called to match the successful matching problem.
  • the answer is passed to the user through the message server, and the accuracy of matching the search answer is enhanced by double matching.
  • Step S404 The local server searches whether an answer corresponding to the question is stored, and if yes, sends a corresponding answer to the message server;
  • Step S405 If the answer corresponding to the problem is not stored in the local server, the message server sends the question and the keyword corresponding to the question to the database server;
  • Step S406 The database server searches whether the answer corresponding to the question is stored, and if so, sends the answer to the message server and synchronizes the question and the answer to the local server.
  • Step S407 If the answer corresponding to the question is not stored in the database server, the message server sends a work answer command, and stores the question and the person answer answer to the database server.
  • Steps S401-S402 and steps S404-S407 in this embodiment are the same as steps S301-S306 in the previous embodiment, and details are not described herein again.
  • the similarity matching algorithm and the keyword matching algorithm are used to ensure the correctness and response rate of the data, and have advantages in various scenarios.
  • an intelligent question answering system provided by an embodiment of the present invention includes: a message server 11, a local server 12, and a database server 13;
  • the local server 12 is configured to receive a problem sent by the message server 11 and a keyword corresponding to the problem;
  • the local server 12 is further configured to search whether an answer corresponding to the question is stored, and if yes, send a corresponding answer to the message server 11;
  • the message server 11 is configured to send a question and a keyword corresponding to the problem to the database server 13 if the answer corresponding to the question is not stored in the local server 12;
  • the database server 13 is configured to find whether an answer corresponding to the question is stored, and if so, send the answer to the message server 11 and synchronize the question and answer to the local server 12.
  • the message server 11 is further configured to: if the answer corresponding to the question is not stored in the database server 13, the sender's work answer command is sent, and the question and the person's work answer are stored in the database server 13.
  • the message server 11 further includes a word segmentation tool
  • the message server 11 is further configured to receive a user question and send the question to the word segmentation tool;
  • the word segmentation tool is used to extract keywords and send keywords to the message server 11.
  • the local server 12 is further configured to match the received questions and keywords with the questions in the local server 12. If the matching is successful, the answer corresponding to the question is invoked.
  • the local server is a SOLR server.
  • the present invention can be implemented by means of software plus a necessary general hardware platform. Based on such understanding, the technical solution of the present invention, which is essential or contributes to the prior art, may be embodied in the form of a software product, which may be stored in a storage medium such as a ROM/RAM or a disk. , an optical disk, etc., includes instructions for causing a computer device (which may be a personal computer, server, or network device, etc.) to perform the methods described in various embodiments of the present invention or portions of the embodiments.
  • a computer device which may be a personal computer, server, or network device, etc.
  • the intelligent question answering method and system can automatically answer user questions, the local server stores questions and answers, and uses the database server as a backup, has a problem update function, and reduces the pressure on the database server load.

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Mathematical Physics (AREA)
  • Computational Linguistics (AREA)
  • Databases & Information Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Artificial Intelligence (AREA)
  • Data Mining & Analysis (AREA)
  • Health & Medical Sciences (AREA)
  • Audiology, Speech & Language Pathology (AREA)
  • General Health & Medical Sciences (AREA)
  • Electrically Operated Instructional Devices (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)

Abstract

本发明公开了智能问答方法，包括：本地服务器接收消息服务端发送的问题以及与问题对应的关键词；本地服务器查找是否存储有与所述问题对应的答案，若有，则发送对应的答案到消息服务端；若本地服务器中没有存储与所述问题对应的答案，则消息服务端将问题及与问题对应的关键词发送到数据库服务器；数据库服务器查找是否存储有与问题对应的答案，若有，则发送答案到消息服务端同时将问题和答案同步到本地服务器中。本发明实施例提供的智能问答方法及***可以自动回答用户提问，本地服务器存储问题和答案，并用数据库服务器作为后备，具有问题更新功能，减少数据库服务器负载的压力。

Description

智能问答方法及*** 技术领域
本发明涉及智能语音应答领域，尤其涉及智能问答方法及***。
背景技术
目前，智能设备语音服务平台***通常由人工应答，当用户提问的问题增多时，人工数也要相应增多，且对于不同客户提出的相同问题人工客服需要反复回答，如此既浪费人力、又降低效率。
发明内容
为了克服现有技术的不足，本发明的目的在于提供智能问答方法及***，其能自动回答用户提问且具有问题更新功能。
本发明的目的采用以下技术方案实现：
智能问答方法，包括：
本地服务器接收消息服务端发送的问题以及与问题对应的关键词；
本地服务器查找是否存储有与所述问题对应的答案，若有，则发送对应的答案到消息服务端；
若本地服务器中没有存储与所述问题对应的答案，则消息服务端将问题及与问题对应的关键词发送到数据库服务器；
数据库服务器查找是否存储有与问题对应的答案，若有，则发送答案到消息服务端同时将问题和答案同步到本地服务器中。
优选的，还包括：
若数据库服务器中没有存储与所述问题对应的答案，则消息服务 端发送人工作答指令，并将问题及人工作答答案存储到数据库服务器中。
优选的，所述本地服务器接收消息服务端发送的问题以及关键词前还包括：
消息服务端接收用户问题并将问题发送到分词工具；
分词工具提取出关键词并将关键词发送到消息服务端。
优选的，所述本地服务器查找是否存储有与所述问题对应的答案前包括：
本地服务器将接收到的问题和关键词与本地服务器中的问题进行匹配，若匹配成功，则调用与所述问题对应的答案。
优选的，所述本地服务器为SOLR服务器。
本发明还涉及智能问答***，包括：消息服务端、本地服务器和数据库服务器；
所述本地服务器，用于接收消息服务端发送的问题以及与问题对应的关键词；
所述本地服务器，还用于查找是否存储有与所述问题对应的答案，若有，则发送对应的答案到消息服务端；
所述消息服务端，用于若本地服务器中没有存储与所述问题对应的答案，则将问题及与问题对应的关键词发送到数据库服务器；
所述数据库服务器，用于查找是否存储有与问题对应的答案，若有，则发送答案到消息服务端同时将问题和答案同步到本地服务器中。
优选的，所述消息服务端，还用于若数据库服务器中没有存储与所述问题对应的答案，则发送人工作答指令，并将问题及人工作答答案存储到数据库服务器中。
优选的，所述消息服务端还包括分词工具；
所述消息服务端还用于，接收用户问题并将问题发送到分词工具；
所述分词工具用于提取出关键词并将关键词发送到消息服务端。
优选的，所述本地服务器，还用于将接收到的问题和关键词与本地服务器中的问题进行匹配，若匹配成功，则调用与所述问题对应的答案。
优选的，所述本地服务器为SOLR服务器。
相比现有技术，本发明的有益效果在于：自动回答用户提问，本地服务器存储问题和答案，并用数据库服务器作为后备，具有问题更新功能，减少数据库服务器负载的压力。
附图说明
图1为本发明一实施例提供的智能问答方法流程图；
图2为本发明另一实施例提供的智能问答方法流程图；
图3为本发明又一实施例提供的智能问答方法流程图；
图4为本发明又一实施例提供的智能问答方法流程图；
图5为本发明实施例提供的智能问答***流程图。
具体实施方式
为了使本发明的目的、技术方案及优点更加清楚明白，以下结合 附图及实施例，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，并不用于限定本发明。
如图1所示，本发明一实施例提供的智能问答方法，包括：
步骤S101：本地服务器接收消息服务端发送的问题以及与问题对应的关键词。
具体的，消息服务端接收用户的问题，将问题以及提取出的与问题对应的关键词发送到本地服务器，本地服务器中预先存储有常见的问题及相应的答案。
步骤S102：本地服务器查找是否存储有与所述问题对应的答案，若有，则发送对应的答案到消息服务端。
具体的，本地服务器根据消息服务端发送的问题和关键词查找是否存储有与所述问题或者关键词对应的答案，若有相应的问题和答案，则发送对应的答案到消息服务端，消息服务端再将答案反馈到用户。若没有相应的问题和答案，本地服务器执行下一步骤。
步骤S103：若本地服务器中没有存储与所述问题对应的答案，则消息服务端将问题及与问题对应的关键词发送到数据库服务器。
具体的，若本地服务器中没有存储与所述问题对应的答案，则本地服务器反馈到消息服务端，消息服务端将问题及与问题对应的关键词发送到数据库服务器，从数据库服务器中查找答案。
步骤S104：数据库服务器查找是否存储有与问题对应的答案，若有，则发送答案到消息服务端同时将问题和答案同步到本地服务器中。
具体的，数据库服务器接收消息服务端发送的问题和答案，查找是否存储有与问题对应的答案，若有，则发送答案到消息服务端，同时将问题和答案同步到本地服务器以待下次使用，若下次用户提问相同的问题，则直接由本地服务器调出答案。
本地服务器中预先存储问题及答案，同时有数据库服务器作为后备，对用户提问的问题进行分流，减小数据库负载压力，不断更新本地服务器；通过问题和关键词进行查找，自动回答用户提问的问题。
优选的，所述本地服务器为SOLR服务器。加入SOLR服务，提高***响应速度，同时用数据库服务做后备，用户提问的问题大部分在SOLR服务就得到解答，极少部分访问数据库服务器，从而减少数据库负载的压力。同时SOLR储存器以及数据库，都具有易扩展的特点，支持零工作迁移及扩展，从而减少后续需要扩张业务时，增加存储设备时的工作量。SOLR存储器与数据库是采用异步TASK JOB方式进行增量同步，占用服务器资源较低，并且优化了同步策略。
图2为本发明另一实施例提供的智能问答方法，包括：
步骤S201：本地服务器接收消息服务端发送的问题以及与问题对应的关键词；
步骤S202：本地服务器查找是否存储有与所述问题对应的答案，若有，则发送对应的答案到消息服务端；
步骤S203：若本地服务器中没有存储与所述问题对应的答案，则消息服务端将问题及与问题对应的关键词发送到数据库服务器；
步骤S204：数据库服务器查找是否存储有与问题对应的答案， 若有，则发送答案到消息服务端同时将问题和答案同步到本地服务器中。
步骤S205：若数据库服务器中没有存储与所述问题对应的答案，则消息服务端发送人工作答指令，并将问题及人工作答答案存储到数据库服务器中。
具体的，数据库服务器根据消息服务端发送的问题及与问题对应的关键词查找是否存储有与问题对应的答案，若没有存储与问题对应的答案，数据库服务器向消息服务端反馈需要人工作答的指令，消息服务端接收人工作答答案后，将答案反馈给用户，另一方面，将该问题及人工作答答案更新存储到数据库服务器。
本实施例中的步骤S201-S204与上一实施例中的步骤S101-S104相同，在此不再赘述。
本地服务器和数据库服务器分流作答用户提问的问题，同时将人工作答的问题和答案存储到数据库服务器，不断更新本地服务器和数据库服务器，保证了整个***的响应率。
图3为本发明又一实施例提供的智能问答方法，包括：
步骤S301：消息服务端接收用户问题并将问题发送到分词工具；
分词工具提取出关键词并将关键词发送到消息服务端。
具体的，消息服务端接收用户问题并将问题发送到分词工具，分词工具根据语法和逻辑提前出关键词，并将关键词发送到消息服务端。
步骤S302：本地服务器接收消息服务端发送的问题以及与问题 对应的关键词；
步骤S303：本地服务器查找是否存储有与所述问题对应的答案，若有，则发送对应的答案到消息服务端；
步骤S304：若本地服务器中没有存储与所述问题对应的答案，则消息服务端将问题及与问题对应的关键词发送到数据库服务器；
步骤S305：数据库服务器查找是否存储有与问题对应的答案，若有，则发送答案到消息服务端同时将问题和答案同步到本地服务器中。
步骤S306：若数据库服务器中没有存储与所述问题对应的答案，则消息服务端发送人工作答指令，并将问题及人工作答答案存储到数据库服务器中。
本实施例中的步骤S302-S206与上一实施例中的步骤S201-S205相同，在此不再赘述。
通过问题和关键词同时查找问题，***响应速度更快、更智能。
图4为本发明又一实施例提供的智能问答方法流程图，包括：
步骤S401：消息服务端接收用户问题并将问题发送到分词工具；
分词工具提取出关键词并将关键词发送到消息服务端。
步骤S402：本地服务器接收消息服务端发送的问题以及与问题对应的关键词；
步骤S403：本地服务器将接收到的问题和关键词与本地服务器中的问题进行匹配，若匹配成功，则调用与所述问题对应的答案。
具体的，本地服务器将接收到的问题和关键词与本地服务器中的 问题进行匹配，通过问题匹配和关键词匹配，若都匹配成功，则从本地服务器中调用与该匹配成功的问题相匹配的答案，并通过消息服务端传递给用户，通过双重匹配增强了匹配查找答***性。
步骤S404：本地服务器查找是否存储有与所述问题对应的答案，若有，则发送对应的答案到消息服务端；
步骤S405：若本地服务器中没有存储与所述问题对应的答案，则消息服务端将问题及与问题对应的关键词发送到数据库服务器；
步骤S406：数据库服务器查找是否存储有与问题对应的答案，若有，则发送答案到消息服务端同时将问题和答案同步到本地服务器中。
步骤S407：若数据库服务器中没有存储与所述问题对应的答案，则消息服务端发送人工作答指令，并将问题及人工作答答案存储到数据库服务器中。
本实施例中的步骤S401-S402以及步骤S404-S407与上一实施例中的步骤S301-S306相同，在此不再赘述。
采用相似度匹配算法和关键字匹配算法，极大的保证了数据的正确性以及响应率，在多种场景下具有优势。
进一步的，本发明还提供智能问答***，如图5所示，本发明一实施例提供的智能问答***，包括：消息服务端11、本地服务器12和数据库服务器13；
所述本地服务器12，用于接收消息服务端11发送的问题以及与问题对应的关键词；
所述本地服务器12，还用于查找是否存储有与所述问题对应的答案，若有，则发送对应的答案到消息服务端11；
所述消息服务端11，用于若本地服务器12中没有存储与所述问题对应的答案，则将问题及与问题对应的关键词发送到数据库服务器13；
所述数据库服务器13，用于查找是否存储有与问题对应的答案，若有，则发送答案到消息服务端11同时将问题和答案同步到本地服务器12中。
优选的，所述消息服务端11，还用于若数据库服务器13中没有存储与所述问题对应的答案，则发送人工作答指令，并将问题及人工作答答案存储到数据库服务器13中。
优选的，所述消息服务端11还包括分词工具；
所述消息服务端11还用于，接收用户问题并将问题发送到分词工具；
所述分词工具用于提取出关键词并将关键词发送到消息服务端11。
优选的，所述本地服务器12，还用于将接收到的问题和关键词与本地服务器12中的问题进行匹配，若匹配成功，则调用与所述问题对应的答案。
优选的，所述本地服务器为SOLR服务器。
本实施例中的***与前述实施例中的方法是基于同一发明构思下的两个方面，在前面已经对方法实施过程作了详细的描述，所以本 领域技术人员可根据前述描述清楚地了解本实施例中的***的结构及实施过程，为了说明书的简洁，在此就不再赘述。
通过以上的实施方式的描述可知，本领域的技术人员可以清楚地了解到本发明可借助软件加必需的通用硬件平台的方式来实现。基于这样的理解，本发明的技术方案本质上或者说对现有技术做出贡献的部分可以以软件产品的形式体现出来，该计算机软件产品可以存储在存储介质中，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)执行本发明各个实施例或者实施例的某些部分所述的方法。
本发明实施例提供的智能问答方法及***可以自动回答用户提问，本地服务器存储问题和答案，并用数据库服务器作为后备，具有问题更新功能，减少数据库服务器负载的压力。
对本领域的技术人员来说，可根据以上描述的技术方案以及构思，做出其它各种相应的改变以及形变，而所有的这些改变以及形变都应该属于本发明权利要求的保护范围之内。

Claims (10)

  1. 智能问答方法，其特征在于，包括：
    本地服务器接收消息服务端发送的问题以及与问题对应的关键词；
    本地服务器查找是否存储有与所述问题对应的答案，若有，则发送对应的答案到消息服务端；
    若本地服务器中没有存储与所述问题对应的答案，则消息服务端将问题及与问题对应的关键词发送到数据库服务器；
    数据库服务器查找是否存储有与问题对应的答案，若有，则发送答案到消息服务端同时将问题和答案同步到本地服务器中。
  2. 根据权利要求1所述的智能问答方法，其特征在于，还包括：
    若数据库服务器中没有存储与所述问题对应的答案，则消息服务端发送人工作答指令，并将问题及人工作答答案存储到数据库服务器中。
  3. 根据权利要求1或2所述的智能问答方法，其特征在于，所述本地服务器接收消息服务端发送的问题以及关键词前还包括：
    消息服务端接收用户问题并将问题发送到分词工具；
    分词工具提取出关键词并将关键词发送到消息服务端。
  4. 根据权利要求1所述的智能问答方法，其特征在于，所述本地服务器查找是否存储有与所述问题对应的答案前包括：
    本地服务器将接收到的问题和关键词与本地服务器中的问题进行匹配，若匹配成功，则调用与所述问题对应的答案。
  5. 根据权利要求1所述的智能问答方法，其特征在于，所述本地 服务器为SOLR服务器。
  6. 智能问答***，其特征在于，包括：消息服务端、本地服务器和数据库服务器；
    所述本地服务器，用于接收消息服务端发送的问题以及与问题对应的关键词；
    所述本地服务器，还用于查找是否存储有与所述问题对应的答案，若有，则发送对应的答案到消息服务端；
    所述消息服务端，用于若本地服务器中没有存储与所述问题对应的答案，则将问题及与问题对应的关键词发送到数据库服务器；
    所述数据库服务器，用于查找是否存储有与问题对应的答案，若有，则发送答案到消息服务端同时将问题和答案同步到本地服务器中。
  7. 根据权利要求6所述的智能问答***，其特征在于，
    所述消息服务端，还用于若数据库服务器中没有存储与所述问题对应的答案，则发送人工作答指令，并将问题及人工作答答案存储到数据库服务器中。
  8. 根据权利要求6或7所述的智能问答***，其特征在于，所述消息服务端还包括分词工具；
    所述消息服务端还用于，接收用户问题并将问题发送到分词工具；
    所述分词工具用于提取出关键词并将关键词发送到消息服务端。
  9. 根据权利要求6所述的智能问答***，其特征在于，
    所述本地服务器，还用于将接收到的问题和关键词与本地服务器中的问题进行匹配，若匹配成功，则调用与所述问题对应的答案。
  10. 根据权利要求6所述的智能问答***，其特征在于，所述本地服务器为SOLR服务器。
PCT/CN2017/120041 2016-12-31 2017-12-29 智能问答方法及*** WO2018121759A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201611266128.8 2016-12-31
CN201611266128.8A CN106844558A (zh) 2016-12-31 2016-12-31 智能问答方法及***

Publications (1)

Publication Number Publication Date
WO2018121759A1 true WO2018121759A1 (zh) 2018-07-05

Family

ID=59116819

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2017/120041 WO2018121759A1 (zh) 2016-12-31 2017-12-29 智能问答方法及***

Country Status (2)

Country Link
CN (1) CN106844558A (zh)
WO (1) WO2018121759A1 (zh)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110321462A (zh) * 2019-05-24 2019-10-11 平安银行股份有限公司 信息动态更新方法、装置、计算机设备及存储介质
CN112925890A (zh) * 2021-03-05 2021-06-08 湖南神通智能股份有限公司 一种智能问答***

Families Citing this family (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106844558A (zh) * 2016-12-31 2017-06-13 深圳市优必选科技有限公司 智能问答方法及***
CN106844686A (zh) * 2017-01-26 2017-06-13 武汉奇米网络科技有限公司 基于solr的智能客服问答机器人及其实现方法
CN107918678A (zh) * 2017-12-28 2018-04-17 北京洪泰同创信息技术有限公司 问答信息处理方法、问答信息处理***及服务器
CN110059171B (zh) * 2019-04-12 2021-01-01 中国工商银行股份有限公司 智能问答性能提升方法及***
CN110225021A (zh) * 2019-06-04 2019-09-10 秒针信息技术有限公司 网络直播方法及装置
CN111064607A (zh) * 2019-12-10 2020-04-24 珠海格力电器股份有限公司 网络运维***的管理方法、装置及存储介质
CN111274360A (zh) * 2020-01-20 2020-06-12 深圳五洲无线股份有限公司 智能语音问答的答案提取方法、录入方法及智能设备

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101076184A (zh) * 2006-07-31 2007-11-21 腾讯科技（深圳）有限公司 一种实现自动应答的***及方法
CN101097573A (zh) * 2006-06-28 2008-01-02 腾讯科技（深圳）有限公司 一种自动问答***及方法
CN102194005A (zh) * 2011-05-26 2011-09-21 卢玉敏 聊天机器人***及自动聊天方法
CN103902652A (zh) * 2014-02-27 2014-07-02 深圳市智搜信息技术有限公司 自动问答***
US20160180242A1 (en) * 2014-12-17 2016-06-23 International Business Machines Corporation Expanding Training Questions through Contextualizing Feature Search
CN106844558A (zh) * 2016-12-31 2017-06-13 深圳市优必选科技有限公司 智能问答方法及***

Family Cites Families (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103177108A (zh) * 2013-03-26 2013-06-26 中山大学 一种基于internet的医疗自动答疑方法及其***
CN103279528A (zh) * 2013-05-31 2013-09-04 俞志晨 一种基于人机结合的问答***及方法
CN104809197A (zh) * 2015-04-24 2015-07-29 同程网络科技股份有限公司 基于智能机器人的在线问答方法
CN105912692B (zh) * 2016-04-22 2019-09-27 华讯方舟科技有限公司 一种智能语音对话的方法和装置

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101097573A (zh) * 2006-06-28 2008-01-02 腾讯科技（深圳）有限公司 一种自动问答***及方法
CN101076184A (zh) * 2006-07-31 2007-11-21 腾讯科技（深圳）有限公司 一种实现自动应答的***及方法
CN102194005A (zh) * 2011-05-26 2011-09-21 卢玉敏 聊天机器人***及自动聊天方法
CN103902652A (zh) * 2014-02-27 2014-07-02 深圳市智搜信息技术有限公司 自动问答***
US20160180242A1 (en) * 2014-12-17 2016-06-23 International Business Machines Corporation Expanding Training Questions through Contextualizing Feature Search
CN106844558A (zh) * 2016-12-31 2017-06-13 深圳市优必选科技有限公司 智能问答方法及***

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110321462A (zh) * 2019-05-24 2019-10-11 平安银行股份有限公司 信息动态更新方法、装置、计算机设备及存储介质
CN112925890A (zh) * 2021-03-05 2021-06-08 湖南神通智能股份有限公司 一种智能问答***

Also Published As

Publication number Publication date
CN106844558A (zh) 2017-06-13

Similar Documents

Publication Publication Date Title
WO2018121759A1 (zh) 智能问答方法及***
CN105678324A (zh) 基于相似度计算的问答知识库的建立方法、装置及***
KR102043365B1 (ko) 음성 인식 전자 디바이스에서 선택적으로 오프라인으로 실행가능한 음성 액션을 위한 데이터의 로컬 유지
US9990591B2 (en) Automated assistant invocation of appropriate agent
CN102194005B (zh) 聊天机器人***及自动聊天方法
US20210081564A1 (en) Hybrid cognitive system for ai/ml data privacy
CN102098342B (zh) 一种基于事务级的数据同步方法、装置及***
CN108984778A (zh) 一种智能交互自动问答***以及自我学习方法
CN102521712B (zh) 一种流程实例数据处理方法和装置
US10679627B2 (en) Processing system for intelligently linking messages using markers based on language data
US10956480B2 (en) System and method for generating dialogue graphs
CN104679910A (zh) 智能应答方法及***
WO2020006827A1 (zh) 一种智能应答的方法和装置
GB2505985A (en) Associating expert speakers with conversation segments
WO2016179939A1 (zh) 一种智能客服处理方法及其装置
CN108874946B (zh) 一种id管理方法及装置
CN105700897B (zh) 一种启动应用程序的方法、装置及终端设备
KR20190138562A (ko) 정보를 생성하기 위한 방법 및 장치
US20160335267A1 (en) Method and apparatus for natural language search for variables
CN107168775A (zh) 一种悬挂事务自动处理的方法和装置
KR101424568B1 (ko) 트랜잭션 재시작 가능한 클라이언트 장치와 데이터베이스 서버 및 방법
CN113051389B (zh) 知识推送方法及装置
CN111367924A (zh) 数据同步方法、装置、电子设备、存储介质
CN113111658A (zh) 校验信息的方法、装置、设备和存储介质
CN106874341B (zh) 一种数据库同步方法

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: 17888175

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: 17888175

Country of ref document: EP

Kind code of ref document: A1