WO2008116403A2 - A method, system and equipment for client service - Google Patents

A method, system and equipment for client service Download PDF

Info

Publication number
WO2008116403A2
WO2008116403A2 PCT/CN2008/070289 CN2008070289W WO2008116403A2 WO 2008116403 A2 WO2008116403 A2 WO 2008116403A2 CN 2008070289 W CN2008070289 W CN 2008070289W WO 2008116403 A2 WO2008116403 A2 WO 2008116403A2
Authority
WO
WIPO (PCT)
Prior art keywords
answer
customer service
sentence
natural language
terminal
Prior art date
Legal status (The legal status 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 status listed.)
Ceased
Application number
PCT/CN2008/070289
Other languages
English (en)
French (fr)
Other versions
WO2008116403A3 (zh
Inventor
Rongling Yu
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Tencent Technology Shenzhen Co Ltd
Original Assignee
Tencent Technology Shenzhen Co Ltd
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 Tencent Technology Shenzhen Co Ltd filed Critical Tencent Technology Shenzhen Co Ltd
Publication of WO2008116403A2 publication Critical patent/WO2008116403A2/zh
Anticipated expiration legal-status Critical
Publication of WO2008116403A3 publication Critical patent/WO2008116403A3/zh
Ceased legal-status Critical Current

Links

Classifications

    • GPHYSICS
    • G06COMPUTING OR CALCULATING; COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/30Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
    • G06F16/33Querying
    • G06F16/332Query formulation
    • G06F16/3329Natural language query formulation

Definitions

  • the present invention relates to the field of instant messaging, and more particularly to a method, system and apparatus for implementing customer services. Background of the invention
  • customer service (known as customer service) is well known, and the quality of customer service has become one of the key factors for companies and even individuals to win in the fierce competition.
  • the method is implemented by publishing the instant messaging account of the customer service personnel (such as MSN, OICQ, etc.) in the network.
  • the network user When the network user establishes a connection with the network user, the user can provide consulting services through instant communication.
  • the specific implementation process is as follows: (1) The user establishes a friend connection between his own instant messaging account and the instant messaging account of the customer service personnel; (2) the user sends a consulting sentence through the client where the user is located, that is, the user's consulting information, including Statements, phrases, phrases, and other forms are collectively referred to as consulting sentences in this patent; (3) The server in the network receives the consulting sentence sent by the user and forwards it to the client where the customer service personnel are located; (4) Customer service The person responds to the consultation sentence and feeds the answer back to the server; (5) The server receives the answer and forwards it to the user.
  • an instant communication account of multiple customer service personnel can be provided, and Each instant messaging account can communicate with a large number of users at the same time, so it has greater flexibility and scalability than the traditional customer service.
  • the server is only used for information forwarding, and the specific reply still fails to get rid of the manual service mode, and the customer service personnel are required to communicate with the user in real time online, and thus there are certain defects.
  • Statistics show that a large proportion of the users' consultation questions are similar, which makes the customer service staff spend a lot of time answering the same questions, and the efficiency is reduced.
  • the customer service staff is easy. If there is an error in the busy, the accuracy of the response is reduced; in addition, once the customer service personnel leave, they cannot provide timely and effective answers to the user's consultation.
  • the present invention provides a method, device and system for implementing customer service.
  • the main technical solutions are as follows:
  • the present invention provides a method for implementing customer service, comprising the steps of: receiving a sentence from a user terminal;
  • the present invention provides a device for implementing customer service, including the following units:
  • a first unit configured to receive a sentence sent by a user terminal
  • the second unit is configured to perform natural language parsing on the sentence received by the first unit
  • the third unit is configured to search for an answer that matches the parsing result sent by the second unit.
  • the present invention provides a system for implementing customer service, including the following devices: a first terminal, configured to send a sentence, and receive an answer of the sentence;
  • the server is configured to perform natural language parsing on the sentence from the first terminal, find an answer that matches the result of the natural language parsing, and send the answer to the first terminal.
  • the server uses the server to perform natural language parsing on the first terminal, that is, the query sentence sent by the user client, and feeds back the found answer matching the parsing result to the user client, because the method is The consultation can be automatically replied independently of the manual mode, thus reducing the workload of the customer service staff and improving the efficiency, accuracy and timeliness of the customer service.
  • FIG. 1 is a structural diagram of a system for implementing customer service in the present invention
  • FIG. 2 is a structural diagram of a system for implementing a customer service in an embodiment of the present invention
  • FIG. 3 is an internal structural diagram of a database of a server in an embodiment of the present invention
  • FIG. 4 is a database of a server in another embodiment of the present invention
  • Figure 5 is a flow chart of a method for implementing customer service in the present invention
  • FIG. 6 is a flow chart of a method of implementing customer service in one embodiment of the present invention. Mode for carrying out the invention
  • the server after receiving the consultation sentence sent by the user client, the server performs natural language analysis on the stored language experience data, and searches for the answer matching the analysis result from the customer service resource data, and then replies to the user client.
  • the language data used for natural language analysis may be language experience data formed according to the experience of language use. (eg: dictionary, word frequency, grammar rules, semantic rules, etc.);
  • the so-called customer service data contains resource information for providing customer service, which contains a large number of answers that can be provided to the user during the process of providing customer service. Only when the server does not find the answer, the consulting sentence sent by the user client is sent to the terminal of the customer service staff, and the customer service staff answers. Since the method can automatically reply to the consultation independently of the manual mode, the workload of the customer service personnel is reduced, and the efficiency, accuracy and timeliness of the customer service are improved.
  • FIG. 1 shows a system structure for implementing a customer service in the present invention, the system including a server
  • client 200 client 300 client
  • connections between the devices in all of the illustrated embodiments of the present invention are intended to clarify the need for their information interaction and control processes and should therefore be considered as logical connections and should not be limited to physical connections.
  • the server 100 is not only used to forward information, but more importantly, the natural language analysis is performed on the consultation sentence sent by the user, and the corresponding answer is searched according to the analysis result and returned to the user, unless the appropriate answer is found, the sentence will be consulted. Forwarded to the customer service staff for answers, so it can largely replace manual operations.
  • the server 100 is typically an instant messaging server, and the scope of the invention should not be limited to a particular type of server.
  • the client includes a user client and a client client, which are typically various terminal devices capable of instant communication, such as a personal computer (Personal Computer, PC), Personal Digital Assistant (PDA), Mobile Phone (MP), etc., so the scope of protection of the present invention should not be limited to a particular type of client.
  • PC Personal Computer
  • PDA Personal Digital Assistant
  • MP Mobile Phone
  • FIG. 2 shows a system structure for implementing customer service in an embodiment of the present invention.
  • the structure includes a server 100, a first client 200 where the user is located, and a second client 300 where the agent is located.
  • the server 100 includes a transceiver unit 101, a language parsing unit 102, a retrieval determination unit 103, and a database 104, where:
  • the transceiver unit 101 is connected to the language parsing unit 102 for performing information interaction with the first client 200 and the second client 300, and specifically includes: receiving a consultation sentence sent by the user through the first client 200, and It is sent to the language parsing unit 102; the answer found by the retrieval determining unit 103 is transmitted to the first client 200, or the consulting sentence is forwarded to the second client 300 according to the instruction of the retrieval determining unit 103.
  • the language parsing unit 102 is connected to the search judging unit 103 and the database 104, and is configured to perform natural language parsing on the consultation sentence sent by the first client 200 according to the language experience data stored in the database 104, and send the parsing result to the search judgment.
  • unit 103 The language parsing unit 102 is connected to the search judging unit 103 and the database 104, and is configured to perform natural language parsing on the consultation sentence sent by the first client 200 according to the language experience data stored in the database 104, and send the parsing result to the search judgment.
  • the process of the natural language parsing by the language parsing unit 102 includes: after receiving the consulting sentence sent by the transceiving unit 101, first performing word segmentation processing; and then segmenting the result according to the language experience data stored in the database 104.
  • the keyword is further determined, and the finally determined keyword is sent to the retrieval determination unit 103 as an analysis result.
  • the search determining unit 103 is connected to the language parsing unit 102 and the database 104, and receives the parsing result sent by the language parsing unit 102, and searches for the corresponding answer from the customer service resource data stored in the database 104, and determines the processing scheme: If the answer is found, the answer is directly fed back to the first client 200; if no answer is found, the query sentence is forwarded to the second client 300 through the transceiver unit 101 together with the context, and the customer service staff answers.
  • the retrieval decision unit 103 also instructs the transceiving unit 101 to send the consultation sentence and its context together to the second client.
  • the first client's consulting content is very rich, and the content of the consulting sentence sent to the second client may express the semantic environment of the current conversation, the topic (topic), and the previous consulting sentence, that is, the context has Very close relationship, the information given by the consultation sentence itself is not enough to make the customer service The staff gives the correct answer.
  • the current consultation sentence and its context need to be sent to the second client, so that the customer service staff can answer more accurately.
  • the search determining unit 103 is a search engine. If the parsing result sent by the language parsing unit 102 is a keyword, the corresponding search result is searched from the customer service resource data stored in the database 104 according to the keyword. answer.
  • the database 104 is used to store data required for parsing and retrieval by the present invention, including language experience data and customer service resource data, which are respectively extracted by the language parsing unit 102 and the retrieval judging unit 103.
  • the database 104 further includes a language database 1041 and a resource database 1042 for storing the language experience data and the customer service resource data described above, respectively.
  • the language experience data stored by the language database 1041 includes: a dictionary, a word frequency, a syntax rule, a semantic rule, etc.
  • the customer service resource data stored by the resource database 1042 includes various information information that the user may desire to know.
  • the customer service resource data is related information of various commodities, including the model, price, brand, quantity, and the like of the commodity.
  • the present invention can store customer service resource data in various ways, such as a question and answer form, and the questions and answers are stored in a database one by one, such as a link form, and the questions and answers are stored in different data segments, but are established between the two.
  • a question and answer form such as a question and answer form
  • the questions and answers are stored in a database one by one, such as a link form
  • the questions and answers are stored in different data segments, but are established between the two.
  • the mapping relationship and so on.
  • the agent can log in to the server 100 periodically to perform maintenance and update on the language experience data and the customer service resource data stored in the database 104.
  • the database 104 includes a language database 1041 and a resource database 1042, respectively storing the aforementioned language experience data and customer service resource data, and further includes an information maintenance module 1043 dedicated to the language database 1041 and the resource database 1042.
  • the stored data is maintained and updated.
  • the information maintenance module 1043 has an information maintenance interface (not shown) for logging in and performing maintenance operations of the agent.
  • the customer service system is used for commodity trading, and the customer service resource data is related information of various commodities. When the customer service personnel inputs the updated product information after logging in, the information maintenance module is adopted.
  • the information maintenance module 1043 converts the item information into a form that the resource database 1042 can store and thus saves it to the resource database 1042 for provisioning. It should be noted that the information maintenance module 1043 in the present invention is not necessarily inside the database 104, and may exist independently, and the invention is not limited to the above examples.
  • a user is registered in the first client 200 for sending a consultation sentence to the server 100 and receiving an answer to the feedback.
  • the second client 300 logs in a customer service personnel, and when the server 100 does not find an answer, receives the user's consulting sentence forwarded by the server 100, and feeds the customer service staff's answer to the server 100, and then feeds back to the server 100.
  • the first client 200 Therefore, the presence of the second client 300 in the present invention is only an aid to the server 100, and is substantially different from the prior art in that it relies entirely on manual operations.
  • FIG. 5 shows a method flow for implementing a customer service in the present invention.
  • the method flow is based on the system structure shown in FIG. 2, and the specific process is as follows:
  • the required language experience data and customer service resource data are pre-stored in the server 100.
  • the language experience data includes a dictionary, a word frequency, a syntax rule, a semantic rule, and the like
  • the customer service resource data includes various information information that the user may desire to understand, such as product related information for conducting commodity transactions, for Business process related information for a new business, and so on.
  • the invention can store customer service resource data in a plurality of ways, and the customer service resource data can include multiple questions and their corresponding answers, wherein the question and the answer can be corresponding to each other, and in addition, the question and answer can be stored in a large amount.
  • the question and answer form that is, the question and the answer are stored statically in the database
  • the link form that is, the question and the answer are stored in different data segments, but a dynamic mapping relationship is established between the two
  • Customer service resource data can also be a multi-dimensional data Library, which indexes the answers based on the type of sentence, topic, keyword, etc.
  • the customer service personnel can periodically log in to the server 100 to perform maintenance and update on the language experience data and the customer service resource data stored in the database 104, thereby continuously improving the customer service system.
  • step S501 the server 100 receives the consultation sentence sent by the user through the first client 200, and performs natural language analysis to obtain the analysis result.
  • the server 100 receives the consultation sentence sent by the user by using the transceiver unit 101, and then performs word segmentation processing by the language parsing unit 102, and then determines according to the language experience data stored in the database 104.
  • the keyword in the word segmentation result is sent to the retrieval determination unit 103 as a result of the analysis.
  • the server 100 uses its retrieval determination unit 103 to find an answer matching the analysis result from the customer service resource data of the database 104, or obtain an answer from the agent.
  • the specific process includes: (1) the search determining unit 103 receives the parsing result of the language parsing unit 102. Since the database 104 stores the customer service resource data in advance, it searches the database 104 whether there is an answer corresponding to the parsing result; (2) Determining the search result: if the suitable answer is searched, the answer is sent to the transceiver unit 101 for transmission, and if not, the command is sent to the transceiver unit 101 to send the user's consultation sentence and its context to the second client. 300.
  • the analysis result received by the retrieval determining unit 103 is a keyword
  • the content matching the keyword is searched for by the keyword to the customer service resource data stored in the database 104.
  • the search determining unit 103 searches for a plurality of answers corresponding to the analysis result, an answer with the highest matching degree is selected and sent to the transceiver unit 101 for transmission to the first A client 200.
  • the retrieval determination unit 103 is an answer searched by a keyword, the degree of matching is determined according to the number of occurrences of the keyword in the searched answer, in an answer The highest number of occurrences of the keyword determines that the answer has the highest degree of matching.
  • the server 100 uses the transceiver unit 101 to reply the obtained answer to the first client 200 where the user is located.
  • the server 100 is disposed between the user and the customer service personnel, and directly communicates with the user, whether the answer searched by the server 100 itself or the answer provided by the customer service personnel is finally forwarded by the server 100 to the user.
  • the server 100 does not search for an answer but is provided by the agent, there is a handover process from the server 100 to the agent, but the process is transparent to the user, and the object that is not aware of the answer has already occurred. Variety.
  • the language parsing unit 102 performs word segmentation processing, and can also extract the sentence type and sentence topic of the consultation sentence.
  • the retrieval judging unit 102 performs retrieval in the customer service resource data stored in the database 104 based on the sentence type of the consultation sentence, the sentence theme, and the keyword. As the conditions for the search increase, the corresponding retrieved answers will be more accurate.
  • the language parsing unit 102 may extract more than one keyword.
  • the retrieval determination unit 103 searches for an answer based on the sentence type, the sentence theme, and each keyword to the customer service resource data stored in the database 104, and then sorts the retrieved candidate answers according to the relevance of the sentence type, the sentence topic, and the keyword. The candidate with the highest relevance is taken as the answer sent to the first client 200. Among them, before the sorting, the retrieved candidate answers can be intersection-processed, and some identical candidate answers are removed.
  • Fig. 6 shows a flow of a method of implementing a customer service in an embodiment of the present invention, the method flow being based on the system structure shown in Fig. 2.
  • the application scenario of this embodiment is e-commerce.
  • the user that is, the buyer, logs in to the first client 200, and the customer service personnel, that is, the seller, logs in the second client 300.
  • the specific process is as follows:
  • the required language experience data and customer service resource data are pre-stored in the database 104 of the server 100.
  • the language experience data includes a dictionary, a word frequency, a syntax rule, a semantic rule, etc.
  • the customer service resource data includes For commodity trading Product-related information, such as the model number, price, quantity, etc. of the product, and the product-related information exists in a question and answer form, and the question and the answer are stored in the database 104 in a one-to-one correspondence.
  • the seller can periodically log in to the server 100 to perform maintenance and update on the language experience data and the customer service resource data stored in the database 104, thereby continuously improving the customer service system.
  • step S601 after receiving the consultation sentence sent by the buyer, the server 100 performs natural language analysis using the language parsing unit 102.
  • step S601 includes two steps: (1) The server 100 performs word segmentation processing on the consulting sentence by using the language parsing unit 102, and the specific word segmentation method includes various types, for example, traditional word segmentation-based word segmentation. The method, the word segmentation method based on the understanding, and the like, thereby obtaining the word segmentation result; (2) further determining the keyword from the word segmentation result, since the database 104 stores language experience data including a dictionary, a word frequency, a syntax rule, a semantic rule, and the like, Then, the keyword in the word segmentation result can be determined, and finally the keyword is sent to the retrieval determination unit 103 as the analysis result.
  • step S602 the server 100 uses the retrieval determination unit 103 to search the database 104 for an answer matching the analysis result.
  • the retrieval determination unit 103 searches the customer service resource data (product related information) stored in the database 104 with the keyword as a search condition, and performs a specific search.
  • the process is similar to the search engine in the prior art.
  • step S603 the retrieval determining unit 103 determines whether an answer is searched, and issues different instructions according to different situations: if no answer is found, a forwarding instruction is sent to the transceiver unit 101; if multiple answers are queried, then a selection is made. A most matching answer and a feedback command is sent to the transceiver unit 101.
  • step S604 since the retrieval determination unit 103 does not search for an answer in the foregoing step S603, and the transceiver unit 101 receives a forwarding instruction, the buyer's consultation sentence is sent. Give the second client 300 where the seller is located, and get the answer from the seller.
  • step S605 after searching for the answer in the foregoing step S603, or obtaining an answer from the seller, the server 100 feeds back the answer to the first client 200 where the buyer is located. So far, a consultation reply process has been completed.

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Mathematical Physics (AREA)
  • Theoretical Computer Science (AREA)
  • Data Mining & Analysis (AREA)
  • Computational Linguistics (AREA)
  • Human Computer Interaction (AREA)
  • Databases & Information Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Artificial Intelligence (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
  • Machine Translation (AREA)
  • Computer And Data Communications (AREA)

Description

一种实现客户服务的方法、 系统及设备 技术领域
本发明涉及即时通信领域, 更具体地说, 涉及一种实现客户服务的 方法、 系统及设备。 发明背景
现在, 客户服务(筒称客服) 已为人们所熟知, 而客服质量的高低 成为了企业甚至个人在激烈竟争中取胜的关键因素之一。
传统的客服方式有多种, 包括邮件和电话等, 而目前应用最广泛的 是通过电话, 例如许多企业都开通有" 24小时客服热线电话"。 但是这种 方式存在一定的局限性, 当同一时刻有多个用户拨打同一个热线电话号 码时, 会因占线而让部分用户等待, 若等待时间过长便 4艮可能会导致用 户的不满, 而失去用户的信任和支持。
随着因特网的迅猛发展, 实现客户服务的方式的兴起则在一定程度 上弥补了传统客服方式的不足。 该方式的实现是通过在网络中公布客服 人员的即时通信帐号 (例如 MSN、 OICQ等) , 当网络用户与其建立连 接后, 则可通过即时通信的方式为用户提供咨询服务。 具体实现过程如 下: (1 )用户将自身的即时通信帐号与客服人员的即时通信帐号之间 建立好友连接; (2 )用户通过其所在的客户端发送咨询句子, 也即用 户的咨询信息, 包括语句、 短语、 词组等多种形式, 在本专利中统称为 咨询句子; (3 ) 网络中的服务器接收到用户发送的咨询句子, 并将其 转发至客服人员所在的客户端; ( 4 )客服人员对该咨询句子进行答复, 并将答案反馈给服务器; ( 5 )服务器接收到答案并转发给用户。 上述 实现客户服务的方式中, 可以提供多个客服人员的即时通信帐号, 而且 每个即时通信帐号可在同一时刻与大量用户进行沟通, 因此与传统的客 服方式相比具备了较强的灵活性和扩展性。
由上可知, 上述方式中服务器仅用于信息转发, 具体的答复仍未能 摆脱人工服务模式, 且要求客服人员必须实时在线的与用户沟通, 因此 也存在一定的缺陷。 统计数据显示, 用户咨询的问题中有很大的比例都 是类似的, 这样就使得客服人员耗费了大量时间在回答同样的问题上, 效率降低; 另外, 当大量用户同时咨询时, 客服人员容易忙中出错, 答 复准确率降低; 再者, 客服人员一旦离开, 则无法对用户的咨询提供及 时有效的答复。
因此需要一种新的实现客户服务的方法, 能够对用户的咨询提供高 效、 准确、 及时的自动答复。 发明内容
本发明提供一种实现客户服务的方法、 设备和系统, 主要技术方 案如下:
本发明提供了一种实现客户服务的方法, 包括以下步骤: 接收来自用户终端的句子;
对该句子进行自然语言解析;
查找与自然语言解析的结果相匹配的答案;
发送该答案给所述用户终端。
本发明提供了一种实现客户服务的设备, 包括以下单元:
第一单元, 用于接收用户终端发送的句子;
第二单元, 用于对第一单元接收的句子进行自然语言解析; 第三单元, 用于查找与第二单元发送的解析结果相匹配的答案。 本发明提供了一种实现客户服务的系统, 包括以下设备: 第一终端, 用于发送句子, 接收该句子的答案;
服务器, 用于对来自第一终端的句子进行自然语言解析, 查找与自 然语言解析的结果相匹配的答案, 并将答案发送给第一终端。
由上述技术方案可见, 本发明中利用服务器对第一终端, 即用户客 户端发送的咨询句子进行自然语言解析, 并将查找到的与解析结果相匹 配的答案反馈给用户客户端, 该方法由于可独立于人工模式对咨询进行 自动答复, 因此降低了客服人员的工作量, 提高了客服的效率、 准确率 和及时性。 附图简要说明
图 1是本发明中实现客户服务的系统结构图;
图 2是本发明的一个实施例中实现客户服务的系统结构图; 图 3是本发明的一个实施例中服务器的数据库的内部结构图; 图 4是本发明的另一实施例中服务器的数据库的内部结构图; 图 5是本发明中实现客户服务的方法流程图;
图 6是本发明的一个实施例中实现客户服务的方法流程图。 实施本发明的方式
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。
本发明中, 服务器接收到用户客户端发送的咨询句子后, 利用所存 储的语言经验数据对其进行自然语言解析, 并从客服资源数据中查找与 解析结果相匹配的答案, 然后回复给用户客户端。 其中, 用于自然语言 解析的语言数据可以为根据语言使用的经验而形成的语言经验数据 (如: 词典、 词频、 语法规则、 语义规则等等); 所谓客服数据包含了 用于提供客服的资源信息, 其内包含了大量在提供客服过程中可提供给 用户的答案。 只有当服务器没有查找到答案时, 才将用户客户端发送的 咨询句子发送给客服人员的终端, 由客服人员解答。 该方法由于可独立 于人工模式对咨询进行自动答复, 因此降低了客服人员的工作量, 提高 了客户服务的效率、 准确率和及时性。
图 1示出了本发明中实现客户服务的系统结构, 该系统包括服务器
100, 以及与其相连的多个客户端 (客户端 200、 客户端 300 客户端
N )。 应当说明的是, 本发明所有图示中各设备之间的连接关系是为了清 楚阐释其信息交互及控制过程的需要, 因此应当视为逻辑上的连接关 系, 而不应仅限于物理连接。
服务器 100在本发明中不仅仅用于转发信息, 更重要的在于对用户 发送的咨询句子进行自然语言解析, 根据解析结果查找相应的答案并回 复给用户, 除非没有找到合适的答案才将咨询句子转发给客服人员进行 解答, 因此在很大程度上能够替代人工操作。 该服务器 100典型的可为 一个即时通信服务器, 而本发明的保护范围并不应限定为某种特定类型 的服务器。
客户端(客户端 200、客户端 300......客户端 N )包括用户客户端以 及客服人员客户端, 其典型的可为各种能够进行即时通信的终端设备, 例如个人计算机( Personal Computer, PC )、个人数字助理( Personal Digital Assistant, PDA ), 移动电话( Mobile Phone, MP )等, 因此本发明的保 护范围不应限定为某种特定类型的客户端。
图 2示出了本发明的一个实施例中实现客户服务的系统结构, 该结 构包括服务器 100、 用户所在的第一客户端 200、 客服人员所在的第二 客户端 300。 服务器 100包括收发单元 101、 语言解析单元 102、 检索判定单元 103和数据库 104, 其中:
( 1 )收发单元 101与语言解析单元 102相连, 用于与第一客户端 200和第二客户端 300之间进行信息交互, 具体包括: 接收用户通过第 一客户端 200发送的咨询句子, 并将其送入语言解析单元 102中; 将检 索判定单元 103所查找到的答案发送给第一客户端 200, 或者根据检索 判定单元 103的指示将咨询句子转发给第二客户端 300。
( 2 )语言解析单元 102与检索判定单元 103以及数据库 104相连, 用于根据数据库 104存储的语言经验数据对第一客户端 200发送的咨询 句子进行自然语言解析, 并将解析结果送入检索判定单元 103中。
在一个示例方案中, 语言解析单元 102进行自然语言解析的过程包 括: 当接收到收发单元 101发送的咨询句子后,首先对其进行分词处理; 然后根据数据库 104所存储的语言经验数据从分词结果中进一步确定关 键词, 最终确定的关键词则作为解析结果送入检索判定单元 103中。
( 3 )检索判定单元 103与语言解析单元 102以及数据库 104相连, 其接收语言解析单元 102发送的解析结果, 并从数据库 104所存储的客 服资源数据中查找对应的答案, 并判定处理方案: 若查找到答案, 则直 接将答案反馈给第一客户端 200; 若没有查找到答案, 则将咨询句子与 其上下文一起通过收发单元 101转发给第二客户端 300, 由客服人员进 行解答。
在一些复杂的情况下, 检索判定单元 103也会指示收发单元 101将 咨询句子及其上下文一起发送到第二客户端。 如在某些情况下, 第一客 户端的咨询内容很丰富, 发送到第二客户端的咨询句子想要表达的内容 可能和当前对话的语义环境、 话题(主题) 以及此前的咨询句子, 即上 下文有非常密切的关系, 单凭咨询句子本身给出的信息不足以使客服人 员给出正确的答案。 此时就需要将当前的咨询句子及其上下文发送到第 二客户端, 使得客服人员能够更加准确的进行解答
在本发明的一个示例方案中,该检索判定单元 103是一个搜索引擎, 若语言解析单元 102发送过来的解析结果是关键词, 则根据关键词从数 据库 104所存储的客服资源数据中查找对应的答案。
( 4 )数据库 104用于存储本发明进行解析以及检索所需要的数据, 包括语言经验数据和客服资源数据, 分别供语言解析单元 102和检索判 定单元 103进行提取。
在本发明的一个示例方案中, 如图 3所示, 数据库 104进一步包括 一个语言数据库 1041和一个资源数据库 1042, 分别存储上述的语言经 验数据和客服资源数据。 在该示例方案中, 语言数据库 1041 存储的语 言经验数据包括: 词典、 词频、 句法规则、 语义规则等等; 资源数据库 1042存储的客服资源数据包括用户可能期望了解的各种资讯信息。在一 个实施例中, 若该客服系统用于进行商品交易, 则客服资源数据则是各 种商品的相关信息, 包括商品的型号、 价格、 品牌、 数量等等。 本发明 可采取多种方式存储客服资源数据, 例如问答形式, 问题与答案一一对 应地存储于数据库中, 例如链接形式, 问题与答案存储于不同的数据段 中, 但是在两者之间建立了映射关系, 等等。
在本发明的另一示例方案中, 客服人员可定期登陆服务器 100, 对 数据库 104中存储的语言经验数据和客服资源数据进行维护更新。 在图 4中, 数据库 104包括一个语言数据库 1041和一个资源数据库 1042, 分别存储前述的语言经验数据和客服资源数据, 此外还包括一个信息维 护模块 1043,专用于对语言数据库 1041和资源数据库 1042所存储的数 据进行维护更新。 该信息维护模块 1043 具有一个信息维护接口 (图中 未示出), 用于客服人员的登陆和执行维护操作。 在一个实施例中, 若 该客服系统用于进行商品交易, 客服资源数据则是各种商品的相关信 息, 客服人员登陆后输入更新后的商品信息时, 则通过该信息维护模块
1043将该商品信息转化为资源数据库 1042能够存储的形式, 从而保存 到资源数据库 1042 中以供应用。 应当说明的是, 本发明中该信息维护 模块 1043不一定在数据库 104内部, 也可独立存在, 并发明并不限定 于以上示例。
第一客户端 200中登录有用户, 用于发送咨询句子给服务器 100, 并接收其反馈的答案。
第二客户端 300中登陆有客服人员, 用于当服务器 100没有找到答 案时, 则接收服务器 100转发过来的用户的咨询句子, 并将客服人员的 解答反馈给服务器 100, 再由服务器 100反馈给第一客户端 200。 因此, 本发明中第二客户端 300的存在仅作为服务器 100的辅助, 与现有技术 完全依赖人工操作相比则具有了本质区别。
图 5示出了本发明中实现客户服务的方法流程, 该方法流程基于图 2所示的系统结构, 具体过程如下:
在执行本发明的所有步骤之前, 服务器 100中预先存储了所需的语 言经验数据和客服资源数据。 在一个示例方案中, 语言经验数据包括词 典、 词频、 句法规则、 语义规则等等, 客服资源数据包括用户可能期望 了解的各种资讯信息, 例如用于进行商品交易的商品相关信息, 用于开 展一项新业务的业务流程相关信息, 等等。 本发明可采取多种方式存储 客服资源数据, 客服资源数据可以包含多个问题及其各自对应的答案, 其中, 问题与答案之间可以——对应, 另外, 问题与答案的存储方式可 以有多种, 例如: 1 ) 问答形式, 即问题与答案静态对应地存储于数据 库中, 2 )链接形式, 即问题与答案存储于不同的数据段中, 但是在两 者之间建立了动态映射关系; 客服资源数据也可以是一个多维度的数据 库, 它根据句子类型、 主题、 关键词等维度对答案建索引等等。 在本发 明中, 客服人员可定期登陆服务器 100, 对数据库 104中存储的语言经 验数据和客服资源数据进行维护更新, 从而不断完善该客服系统。
在步骤 S501 中, 服务器 100接收用户通过第一客户端 200发送的 咨询句子, 并对其进行自然语言解析, 得到解析结果。
在上述步骤的第一个示例方案中, 服务器 100利用其收发单元 101 接收到用户发送的咨询句子, 再利用其语言解析单元 102首先对其进行 分词处理, 然后根据数据库 104存储的语言经验数据确定分词结果中的 关键词, 再将关键词作为解析结果送入检索判定单元 103中。
在步骤 S502中,服务器 100利用其检索判定单元 103从数据库 104 的客服资源数据中查找与解析结果相匹配的答案, 或者从客服人员处获 取答案。具体过程包括: ( 1 )检索判定单元 103接收到语言解析单元 102 的解析结果,由于数据库 104预先存储了客服资源数据,则从数据库 104 中搜索是否存在与解析结果相对应的答案; (2 )对搜索结果进行判定: 若搜索到合适的答案则将答案送入收发单元 101进行发送, 若没有搜索 到则向收发单元 101发出指令将用户的咨询句子及其上下文一并发送给 第二客户端 300。
在上述步骤搜索答案的一个示例方案中, 若检索判定单元 103所接 收到的解析结果是关键词, 则利用关键词到数据库 104存储的客服资源 数据中搜索与该关键词相匹配的内容。
在上述步骤对搜索结果进行判定的一个示例方案中, 若检索判定单 元 103搜索到与解析结果相对应的多个答案时, 从中选择一个匹配程度 最高的答案, 并送入收发单元 101发送给第一客户端 200。 在该示例方 案的一个实施例中, 若检索判定单元 103是通过关键词搜索的答案, 则 根据关键词在所搜索到的答案中出现的次数来确定匹配程度, 在某答案 中该关键词出现的次数最高则判定该答案具有最高的匹配程度。
在步骤 S503中, 服务器 100利用收发单元 101将所获取的答案回 复给用户所在的第一客户端 200。 在本发明中, 服务器 100设置于用户 和客服人员之间, 与用户直接沟通, 不论是服务器 100本身搜索到的答 案还是客服人员提供的答案, 最终均由服务器 100转发至用户。 在服务 器 100没有搜索到答案而由客服人员提供的情形下, 存在一个从服务器 100到客服人员的切换过程, 但是该过程对用户是透明的, 其意识不到 为其提供答案的对象已经发生了变化。
上述的实施例中, 语言解析单元 102进行分词处理的同时, 还可以 提取咨询句子的句子类型和句子主题。 检索判断单元 102根据咨询句子 的句子类型、 句子主题和关键词在数据库 104存储的客服资源数据中进 行检索。 由于检索的条件增加了, 相应的检索出的答案会更加准确。
语言解析单元 102提取出来的关键词可能不止一个。 检索判定单元 103会根据句子类型、 句子主题以及每个关键词到数据库 104存储的客 服资源数据中查找答案, 然后根据与句子类型、 句子主题和关键词的相 关度对检索到的候选答案排序, 取出相关度最高的候选答案作为发送给 第一客户端 200的答案。 其中在进行排序前还可以对检索到的候选答案 进行求交集处理, 去掉一些相同的候选答案。
图 6示出了本发明的一个实施例中实现客户服务的方法流程, 该方 法流程基于图 2所示的系统结构。 该实施例的应用场景是电子商务, 用 户即买家, 登陆于第一客户端 200中, 客服人员即卖家, 登陆于第二客 户端 300中。 具体过程如下:
在执行本发明的所有步骤之前, 服务器 100的数据库 104中预先存 储了所需的语言经验数据和客服资源数据, 语言经验数据包括词典、 词 频、 句法规则、 语义规则等等, 客服资源数据包括用于进行商品交易的 商品相关信息, 例如商品的型号、 价格、 数量等, 且这些商品相关信息 是以问答形式存在, 问题与答案一一对应地存储于数据库 104中。此外, 卖家可定期登陆服务器 100, 对数据库 104中存储的语言经验数据和客 服资源数据进行维护更新, 从而不断完善该客服系统。
在步骤 S601 中, 服务器 100接收到买家发送的咨询句子后, 利用 其语言解析单元 102对其进行自然语言解析。
在该实施例中, 步骤 S601 的具体实现包括两个步骤: (1 )服务器 100利用语言解析单元 102对咨询句子进行分词处理, 具体的分词方法 包括多种, 例如传统的基于字符串匹配的分词方法、 基于理解的分词方 法等, 从而得到分词结果; (2 )进一步从分词结果中确定关键词, 由于 数据库 104中存储了包括词典、 词频、 句法规则、 语义规则等在内的语 言经验数据, 则可利用其确定分词结果中的关键词, 最终将该关键词作 为解析结果送入检索判定单元 103中。
在步骤 S602中, 服务器 100利用检索判定单元 103从数据库 104 中搜索与解析结果相匹配的答案。
在该实施例中, 由于前述步骤 S601 的解析结果是关键词, 则检索 判定单元 103则以关键词作为搜索条件, 在数据库 104存储的客服资源 数据(商品相关信息) 中进行搜索, 具体的搜索过程与现有技术中的搜 索引擎类似。
在步骤 S603 中, 检索判定单元 103判定是否搜索到答案, 并根据 不同的情形发出不同的指令: 若没有查询到答案, 则发出转发指令给收 发单元 101 ; 若查询到多个答案, 则从中选择一个匹配程度最高的答案, 并发出反馈指令给收发单元 101。
在步骤 S604中, 由于前述步骤 S603中检索判定单元 103没有搜索 到答案, 收发单元 101接收到一个转发指令, 则将买家的咨询句子发送 给卖家所在的第二客户端 300, 从卖家处获取解答。
在步骤 S605中, 前述步骤 S603中搜索到答案, 或者从卖家处获得 答案后, 则服务器 100将答案反馈给买家所在的第一客户端 200。 至此, 完成了一次咨询答复过程。
应当说明的是, 本发明典型的应用但不限于电子商务和企业客服, 在其他类似的场景中也可以应用本发明所阐述的方法。
以上所述仅为本发明的较佳实施例而已, 并不用以限制本发明, 凡 在本发明的精神和原则之内所作的任何修改、 等同替换和改进等, 均应 包含在本发明的保护范围之内。

Claims

权利要求书
1、 一种实现客户服务的方法, 其特征在于, 包括:
接收来自用户终端的句子;
对所述句子进行自然语言解析;
查找与所述自然语言解析的结果相匹配的答案;
发送所述答案给所述用户终端。
2、 根据权利要求 1所述的方法, 其特征在于, 所述查找与自然语言 解析的结果相匹配的答案包括:
从客服数据中查找与所述自然语言解析的结果相匹配的答案。
3、 根据权利要求 2所述的方法, 其特征在于, 如果在所述客服数据 中没有找到与解析结果相匹配的答案, 进一步包括:
转发所述句子到客服终端;
接收所述客服终端根据所述句子确定的答案;
转发所接收的答案到所述用户终端。
4、 根据权利要求 3所述的方法, 其特征在于, 进一步转发所述句子 的上下文到客服终端;
所接收的答案为: 客服终端根据所述句子及其上下文确定的答案。
5、 根据权利要求 1或 4所述的方法, 其特征在于, 所述对所述句子 进行自然语言解析包括:
提取句子的句子类型和句子主题;
对所述句子进行分词处理;
从所述分词处理的结果中确定关键词;
所述从客服数据中查找与所述自然语言解析的结果相匹配的答案 包括: 从所述客服数据中查找与所述句子类型、 句子主题和关键词中任意 一者或几者的组合相关度最高的答案。
6、 根据权利要求 1至 4任一项所述的方法, 其特征在于, 根据预存 的语言数据对所述句子进行自然语言解析。
7、 根据权利要求 6所述的方法, 其特征在于, 进一步包括: 对所述 客服数据和所述语言数据进行更新维护。
8、 根据权利要求 7所述的方法, 其特征在于, 所述语言数据包括: 词典、 词频、 语法规则和语义规则中的任一者或任几者。
9、 根据权利要求 1至 4任一项所述的方法, 其特征在于, 所述客服 数据包括至少一个问题及其对应的答案;
所述从客服数据中查找与所述自然语言解析的结果相匹配的答案 包括: 查找与所述自然语言解析的结果相匹配的问题, 得到与查找到的 问题对应的答案。
10、 一种实现客户服务的设备, 其特征在于, 包括,
第一单元, 用于接收用户终端发送的句子;
第二单元, 用于对所述第一单元接收的所述句子进行自然语言解 第三单元, 用于查找与所述第二单元发送的解析结果相匹配的答 案。
11、 根据权利要求 10所述的设备, 其特征在于, 进一步包括: 数据库, 用于存储在第二单元中进行自然语言解析时使用的语言数 据以及在第三单元查找答案时使用的客服数据。
12、 根据权利要求 11所述的设备, 其特征在于, 所述数据库包括: 语言数据库, 用于存储在第二单元中进行自然语言解析时使用的语 言数据, 所述语言包括词典、 词频、 语法规则和语义规则中任一者或任 几者;
客服数据库, 用于存储在第三单元查找答案时使用的客服数据, 所 述客服数据包括至少一个问题及其对应的答案。
13、 根据权利要求 11所述的设备, 其特征在于, 所述数据库进一步 包括:
用于对所述语言数据库和所述客服数据库中存储的数据进行维护 更新的模块。
14、 根据权利要求 10所述的设备, 其特征在于, 所述第一单元进一 步用于在所述第三单元没有查找到与所述自然语言解析的结果相匹配 的答案时, 将所述句子发送到客服终端, 并接收所述客服终端根据所述 句子确定的答案, 转发所述答案到所述用户终端。
15、 一种实现客户服务的系统, 其特征在于, 包括:
第一终端, 用于发送句子, 接收所述句子的答案;
服务器, 用于对来自第一终端的所述句子进行自然语言解析, 查找 与所述自然语言解析的结果相匹配的答案, 并将所述答案发送给所述第 一终端。
16、 根据权利要求 15所述的系统, 其特征在于, 进一步包括第二终 端;
服务器, 进一步用于当未查找到与所述自然语言解析的结果相匹配 的答案时, 将所述句子转发到第二终端, 接收所述第二终端发送的答案 并转发给所述第一终端;
第二终端, 用于接收所述句子, 根据所述句子确定答案, 并发送所 述答案到所述服务器。
17、 根据权利要求 15或 16所述的系统, 其特征在于, 所述服务器为 权利要求 10至 14任一项所述的设备。
PCT/CN2008/070289 2007-03-28 2008-02-05 A method, system and equipment for client service Ceased WO2008116403A2 (en)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CNA2007100737164A CN101193071A (zh) 2007-03-28 2007-03-28 一种基于即时通信的客服方法、系统及设备
CN200710073716.4 2007-03-28

Publications (2)

Publication Number Publication Date
WO2008116403A2 true WO2008116403A2 (en) 2008-10-02
WO2008116403A3 WO2008116403A3 (zh) 2011-02-10

Family

ID=39487829

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2008/070289 Ceased WO2008116403A2 (en) 2007-03-28 2008-02-05 A method, system and equipment for client service

Country Status (2)

Country Link
CN (1) CN101193071A (zh)
WO (1) WO2008116403A2 (zh)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106899487A (zh) * 2016-07-06 2017-06-27 阿里巴巴集团控股有限公司 通信处理方法、装置、服务器及设备

Families Citing this family (33)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101888452A (zh) * 2009-05-11 2010-11-17 中兴通讯股份有限公司 一种多接入式客服系统及其方法
CN102831482A (zh) * 2012-08-01 2012-12-19 浙江兴旺宝明通网络有限公司 针对泵阀行业的一种基于智能问答的启发式询问系统
CN104469028B (zh) * 2013-09-24 2017-10-24 中国移动通信集团江苏有限公司 一种服务提供方法、会话服务器及客服系统
CN103546881A (zh) * 2013-11-06 2014-01-29 从兴技术有限公司 上行短信的识别方法及装置
CN104881410B (zh) * 2014-02-27 2020-03-17 腾讯科技(深圳)有限公司 客户平台信息交互方法、装置及系统
CN104268667A (zh) * 2014-09-03 2015-01-07 陈飞 一种协助管理人员进行纠纷调解的方法及系统
CN105681609A (zh) * 2014-11-20 2016-06-15 中兴通讯股份有限公司 一种客服中心座席的辅助方法和装置
CN106155532A (zh) * 2015-04-23 2016-11-23 腾讯科技(深圳)有限公司 一种快捷回复的方法、终端及系统
CN105162977A (zh) * 2015-08-26 2015-12-16 上海银天下科技有限公司 说辞推荐方法和装置
CN105631686A (zh) * 2015-12-22 2016-06-01 北京奇虎科技有限公司 客服系统解决问题的方法及装置
CN106936689A (zh) * 2015-12-30 2017-07-07 航天信息软件技术有限公司 基于即时通讯公众号的客户服务方法及设备
CN107135140B (zh) * 2016-02-29 2020-07-14 阿里巴巴集团控股有限公司 一种即时通信方法和装置
WO2017173633A1 (zh) * 2016-04-07 2017-10-12 汤美 教育项目的智能回复方法及系统
CN105939246A (zh) * 2016-04-15 2016-09-14 北京思特奇信息技术股份有限公司 一种基于im的智能客服方法及系统
CN105791104B (zh) * 2016-05-18 2019-06-07 北京奔影网络科技有限公司 客服方法和装置
CN107423307B (zh) * 2016-05-24 2020-11-17 创新先进技术有限公司 一种互联网信息资源的分配方法及装置
CN107528761A (zh) * 2016-06-21 2017-12-29 天脉聚源(北京)科技有限公司 一种微信公众号自动回复信息的方法和系统
CN106340308A (zh) * 2016-08-25 2017-01-18 北京云知声信息技术有限公司 语音答复方法及装置
CN106484916A (zh) * 2016-11-04 2017-03-08 上海智臻智能网络科技股份有限公司 一种招生问答方法及装置
CN108171347A (zh) * 2016-12-07 2018-06-15 北京京东尚科信息技术有限公司 一种预约业务的实现方法和装置
CN107784051B (zh) * 2016-12-15 2020-09-18 平安科技(深圳)有限公司 在线客服应答系统及方法
TWI596491B (zh) * 2017-01-16 2017-08-21 網智服務國際股份有限公司 知識庫之智能化控制系統與其回饋方法
CN107135247B (zh) * 2017-02-16 2019-11-29 江苏南大电子信息技术股份有限公司 一种人与人工智能协同工作的服务系统及方法
CN107153965A (zh) * 2017-04-05 2017-09-12 芜湖恒天易开软件科技股份有限公司 一种多终端的智能客服解决方法
CN107766416A (zh) * 2017-09-08 2018-03-06 阿里巴巴集团控股有限公司 数据分析方法、装置及系统
CN108400933A (zh) * 2018-03-21 2018-08-14 广东您好科技有限公司 一种基于语义理解和知识图谱实现智能通讯的方法及装置
CN108718273A (zh) * 2018-05-11 2018-10-30 平安科技(深圳)有限公司 基于即时通信的会话方法及装置、存储介质、电子设备
CN110543587A (zh) * 2019-07-24 2019-12-06 湖北爱运动体育信息科技有限公司 一种基于物联网的体育健身交流平台
CN110852119A (zh) * 2019-11-11 2020-02-28 广州点动信息科技股份有限公司 一种电子商务自动客服系统
CN112183953A (zh) * 2020-09-08 2021-01-05 北京达佳互联信息技术有限公司 客服资源的分配方法、装置、电子设备和存储介质
CN112559550B (zh) * 2020-10-30 2021-09-07 北京智源人工智能研究院 基于语义规则和多维模型的多数据源nl2sql系统
CN114491051B (zh) * 2022-04-02 2022-07-29 四川省大数据中心 一种建筑用地项目审批系统
CN115292482B (zh) * 2022-10-09 2023-01-17 深圳市华云中盛科技股份有限公司 一种用于公益诉讼的消息处理方法

Family Cites Families (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6658377B1 (en) * 2000-06-13 2003-12-02 Perspectus, Inc. Method and system for text analysis based on the tagging, processing, and/or reformatting of the input text
GB2394098A (en) * 2001-06-20 2004-04-14 Cellesense Technologies Ltd Method and system for providing information services to subscribed users of a cellular network using an interface device
CN1335574A (zh) * 2001-09-05 2002-02-13 罗笑南 智能语义搜索方法
JP2003248559A (ja) * 2002-02-25 2003-09-05 Sony Corp 情報処理装置、端末装置及び情報処理システム
US20050187920A1 (en) * 2004-01-23 2005-08-25 Porto Ranelli, Sa Contextual searching
CN1564159A (zh) * 2004-04-05 2005-01-12 宋春雨 一种自然语言短信息查询服务方法
CN1808427A (zh) * 2005-01-22 2006-07-26 鸿富锦精密工业(深圳)有限公司 专利信息检索的语法转换方法
WO2006081835A1 (en) * 2005-02-06 2006-08-10 Linguit Gmbh Method and apparatus for mobile information access in natural language

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106899487A (zh) * 2016-07-06 2017-06-27 阿里巴巴集团控股有限公司 通信处理方法、装置、服务器及设备

Also Published As

Publication number Publication date
CN101193071A (zh) 2008-06-04
WO2008116403A3 (zh) 2011-02-10

Similar Documents

Publication Publication Date Title
WO2008116403A2 (en) A method, system and equipment for client service
US12405982B2 (en) Tool for providing contextual data for natural language queries
CN111480159B (zh) 用于增强的企业列表的知识搜索引擎平台
CN101076184B (zh) 一种实现自动应答的系统及方法
CN109189898A (zh) 一种智能应答方法和系统
US20100138402A1 (en) Method and system for improving utilization of human searchers
US20110125697A1 (en) Social media contact center dialog system
WO2011053758A1 (en) Method and system of processing a query using human assistants
WO2011005854A1 (en) Method and system of providing search tools
CN108452526B (zh) 游戏故障原因的查询方法和装置、存储介质、电子装置
US11531816B2 (en) Search apparatus based on synonym of words and search method thereof
US11256770B2 (en) Data-driven online business name generator
CN103177039A (zh) 一种数据处理方法及装置
CN111046151A (zh) 一种消息处理方法及装置
CN121303331A (zh) 智能问答方法及计算机程序产品
US20200351241A1 (en) Data-driven online domain name generator
CN109145092B (zh) 一种数据库更新、智能问答管理方法、装置及其设备
CN118093843B (zh) 基于知识库的问答处理系统
US20210195030A1 (en) Electronic communication system for providing expert information and method of using same
WO2010060117A1 (en) Method and system for improving utilization of human searchers
CN106933848A (zh) 一种信息发送方法和装置
CN113325959B (zh) 一种输入语料的推荐方法和装置
GB2497194A (en) Method for managing email
Anityasari et al. Text mining implementation in complaint management: A case study at Surabaya city office for population administration and civil registration (COPACR)
JP2024003862A (ja) チャットシステムおよびチャットプログラム

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: 08706662

Country of ref document: EP

Kind code of ref document: A2

NENP Non-entry into the national phase

Ref country code: DE

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

Free format text: COMMUNICATION UNDER RULE 112(1) EPC, EPO FORM 1205A DATED 10/02/10

122 Ep: pct application non-entry in european phase

Ref document number: 08706662

Country of ref document: EP

Kind code of ref document: A2