CN109194568A - Robot customer service and intelligent customer service system - Google Patents

Robot customer service and intelligent customer service system Download PDF

Info

Publication number
CN109194568A
CN109194568A CN201811072289.2A CN201811072289A CN109194568A CN 109194568 A CN109194568 A CN 109194568A CN 201811072289 A CN201811072289 A CN 201811072289A CN 109194568 A CN109194568 A CN 109194568A
Authority
CN
China
Prior art keywords
customer service
message
session
unit
conversation message
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.)
Granted
Application number
CN201811072289.2A
Other languages
Chinese (zh)
Other versions
CN109194568B (en
Inventor
朱频频
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.)
Shanghai Zhizhen Intelligent Network Technology Co Ltd
Original Assignee
Shanghai Zhizhen Intelligent Network Technology 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 Shanghai Zhizhen Intelligent Network Technology Co Ltd filed Critical Shanghai Zhizhen Intelligent Network Technology Co Ltd
Priority to CN201811072289.2A priority Critical patent/CN109194568B/en
Publication of CN109194568A publication Critical patent/CN109194568A/en
Application granted granted Critical
Publication of CN109194568B publication Critical patent/CN109194568B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L51/00User-to-user messaging in packet-switching networks, transmitted according to store-and-forward or real-time protocols, e.g. e-mail
    • H04L51/02User-to-user messaging in packet-switching networks, transmitted according to store-and-forward or real-time protocols, e.g. e-mail using automatic reactions or user delegation, e.g. automatic replies or chatbot-generated messages

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Manipulator (AREA)
  • Information Transfer Between Computers (AREA)

Abstract

A kind of robot customer service, comprising: first message interactive unit suitable for receiving the conversation message from consumer user end, and returns to response message;Recognition unit, suitable for identifying the content of conversation message received by first message interactive unit;Ability judging unit, being adapted to determine whether can be from main process task conversation message;Storage unit is suitable for storage service resource;First response message generation unit, suitable for being extracted from the Service Source that storage unit stores and the matched response message of the content of conversation message institute;Status control unit is session suspended state suitable for controlling state of the robot customer service in the subsequent session business with consumer user end;Second response message generation unit, suitable for extracting the pacifying property response message with the content matching of conversation message from the Service Source that storage unit stores;Second message interactive unit, suitable for the conversation message from customer's client is sent to customer service client.

Description

Robot customer service and intelligent customer service system
It is on December 30th, 2015 that the application, which is the applying date, and application No. is 201511024389.4, invention and created name is The divisional application of " robot customer service and its client service method, intelligent customer service system ".
Technical field
The present invention relates to instant messaging field more particularly to a kind of robot customer service and intelligent customer service systems.
Background technique
With the rapid development of Internet, the raising of people's service awareness, network customer service has spread to all trades and professions, deep Enter the links to everyday commerce service.
Current customer service system is usually made of machine customer service and artificial customer service, and wherein machine customer service is generally based on net The immediate communication tool of page, artificial customer service be usually the immediate communication tool being embedded in webpage, each artificial customer service client There is manual maintenance at end.In customer service processes, when receiving the conversation message from customer, first carried out by machine customer service Service.When customer thinks that machine customer service can not solve the problems, such as its proposition, then click artificial customer service manually, to artificial customer service into Row consulting.
Above-mentioned customer service system, can quick response client to a certain extent demand.However, in actual application, Machine customer service identifies the content of the conversation message of customer's input, searches and the conversation message from the resource database stored Content matching data, and the obtained data of matching are sent to user client and carry out response, for same problem, usually Repeated answer can be mechanically carried out, is easy to enrage customer, influences Customer Experience.And artificial customer service then can not Always Online", and Multiple customers are usually faced simultaneously, thus there is the case where cannot replying in time.Thus current customer service system is difficult to effectively The demand for services for meeting customer, influences user experience.
Summary of the invention
Present invention solves the technical problem that being how to enhance the validity of customer service, user experience is improved.
In order to solve the above technical problems, the embodiment of the present invention provides a kind of robot customer service, comprising:
First message interactive unit suitable for receiving the conversation message from consumer user end, and returns and disappears with the session The matched response message of content institute of breath, the response message include the robot customer service with behind the consumer user end Pacifying property response message when state in extended meeting call business is session suspended state;
Recognition unit, suitable for identifying the content of conversation message received by the first message interactive unit;
Ability judging unit, suitable for the content of the conversation message identified according to the recognition unit, determination is It is no can be from conversation message described in main process task;
Storage unit is suitable for storage service resource;
First response message generation unit, suitable for being determined to handle the conversation message when the ability judging unit When, it is extracted from the Service Source that the storage unit stores and the matched response message of the content of conversation message institute;
Status control unit is suitable for when the ability judging unit determination can not can not from main process task or the recognition unit When identifying the conversation message, control the robot customer service with the state in the subsequent session business at the consumer user end For session suspended state;
Second response message generation unit, suitable for being extracted and the session from the Service Source that the storage unit stores The pacifying property response message of the content matching of message;
Second message interactive unit, suitable in the robot customer service in the subsequent session business with the consumer user end In state be session suspended state when, the conversation message from consumer user end is sent to customer service client;
The status control unit is further adapted in the session service at the robot customer service with the consumer user end It, will be described in session status, and when the second message interactive unit is received from the conversation message of the customer service client Robot customer service is being switched to session suspended state with the state in the subsequent session business at the consumer user end, and controls institute It states second message interactive unit and executes message exchange operations.
Optionally, the Service Source that the storage unit is stored includes: language empirical data and customer service resource data;Institute State ability judging unit, comprising:
First judgment sub-unit is adapted to determine that whether the robot customer service can be independently according to the language empirical data It answers the content of the conversation message;
Second judgment sub-unit, suitable for determining that the robot customer service can not autonomous basis when first judgment sub-unit When the language empirical data is answered, scene type belonging to the content of the conversation message is determined.
Optionally, the first response message generation unit, comprising:
First generates subelement, is suitable for independently being made according to the language empirical data when the determining robot customer service When answering, extracted from the language empirical data that the storage unit is stored and the matched response of the content of conversation message institute Message;
Second generates subelement, is suitable for independently being made according to the language empirical data when the determining robot customer service When answering, from the content institute extracted in the customer service resource data that the storage unit is stored under affiliated scene with the conversation message Matched response message.
Optionally, the robot customer service further include: state detection unit, suitable for being connect when the first message interactive unit When receiving from the conversation message at consumer user end, the robot customer service is detected in the session service with the consumer user end In state, and when detect the robot customer service with the state meeting of being not in the session service at the consumer user end When talking about suspended state, the operation that the recognition unit executes the content of the identification conversation message is triggered.
Optionally, the recognition unit is further adapted for when objective with robot described in the session service at the consumer user end Clothes are in session suspended state, and the second message interactive unit does not receive the conversation message from the customer service client When, identify the conversation message from consumer user end that the first message interactive unit receives;
The second response message generation unit, is further adapted in the conversation message identified according to the recognition unit Hold, the pacifying property response with the content matching of the conversation message is extracted from the Service Source that the storage unit stores and is disappeared Breath.
Optionally, the robot customer service further include: conversation end detection unit, be suitable for judge the customer service client and Whether the session between the consumer user end terminates;
The status control unit is further adapted for when the session between the determining customer service client and the consumer user end At the end of, restoring state of the robot customer service between the consumer user end in subsequent session business is session shape State.
Optionally, the conversation end detection unit comprises at least one of the following:
First detection sub-unit, is adapted to detect for whether the second message interactive unit free time reaches preset duration, and is reaching When preset duration, the conversation end between the customer service client and the consumer user end is determined;
Second detection sub-unit is adapted to detect for the session from the customer service client and completes signal, and comes detecting When completing signal from the session of the customer service client, the session between the customer service client and the consumer user end is determined Terminate.
Optionally, the second message interactive unit is further adapted for not disappearing described from main process task or unrecognized session Corresponding the pacifying property response message of breath with it is described can not be sent to from together with main process task or unrecognized conversation message it is described Customer service client.
In order to solve the above technical problems, the embodiment of the present invention also provides a kind of intelligent customer service system, comprising: consumer user End, customer service client and above-mentioned robot customer service suitable for conversating with the consumer user end, the customer service client It is conversated with the consumer user end by the robot customer service.
Optionally, the robot customer service is suitable for when the consumer user end and customer service client that are in session status being more When a, when determination can not be from main process task or when can not identify the conversation message, according to the customer service client for being in session status Task amount, it is minimum that the conversation message from consumer user end and corresponding pacifying property response message are sent to task amount Customer service client;Alternatively, recorded according to conversation message, can not be handled described or unrecognized conversation message is sent to it The customer service client of the excessively described consumer user end conversation message of pre-treatment.
Compared with prior art, the technical solution of the embodiment of the present invention has the advantages that
Robot customer service is when receiving from the conversation message at consumer user end, by identifying in the conversation message Hold and determines whether to be capable of handling the conversation message.When being determined to from conversation message described in main process task, from what is stored The response message with the content matching of the conversation message is extracted in Service Source, and is sent to the consumer user end, so that Customer can timely and effectively be replied.And when determination can not be handled or can not identify the conversation message, on the one hand, will The robot customer service is being set as session suspended state with the state in the subsequent session business at the consumer user end, can be with Save the process resource of robot customer service;On the other hand, it is pacified by sending in time with the content matching of the conversation message Property response message, can cannot get reasonable response to avoid customer generates impatience, and the robot itself can not be known It can not be sent to not or from the conversation message of main process task customer service client, be manually selected without customer, in the robot customer service When can not be effectively treated, the timely intervention of artificial customer service may be implemented and effectively solve the problems, such as the customer.In short, above-mentioned side Case can timely and effectively solve the problems, such as customer, and customer by the robot customer service collaborative work with artificial customer service automatically It is the difference for being not felt by robot customer service Yu artificial customer service, the mood in waiting answering is available effectively to pacify, Obtain better Interactive Experience.
Further, by carrying out big data training to the robot customer service, the robot customer service is known Whether scene type belonging to other conversation message is capable of handling the conversation message and carries out subsequent so as to quickly determine Processing, can be improved customer service response speed, further enhances the Interactive Experience of customer.
Further, when robot customer service is determined to independently be answered according to language empirical data, from the language stored The matched response message of content institute of extraction and the conversation message in speech empirical data, and when robot customer service determination can not be certainly When main root is answered according to language empirical data, the matched response message of content institute under affiliated scene with the conversation message is chosen, To can not only be answered according to language empirical data, and can be answered according to customer service resource data, and due to matched money Source data is corresponding with the scene of customer service, thus the range for the response message that robot customer service can answer can be improved and answer The accuracy answered.
Further, when receiving from the conversation message at consumer user end, robot customer service itself detects certainly first Whether body is in session suspended state, and determine itself be not in session suspended state when, just held by the robot customer service Row identifies the operation of the content of the conversation message, and when determining that itself is in session suspended state, then led to by artificial customer service It crosses customer service client and customer conversates interaction, artificial preferably cooperate with robot customer service to realize, on the one hand, When robot customer service is in session suspended state, it is only necessary to the transmission for the message that conversates and any autonomous place need not be carried out Reason, can save the Service Source of robot customer service, and on the other hand, robot customer service only itself is being not in session suspended state Shi Caineng work, the priority of old friend's work customer service is higher, can seize session to avoid robot customer service midway, enables customer Consistent formula service experience is obtained, the user experience of customer is improved.
Further, when robot customer service is in session suspended state, if not receiving the meeting from customer service client Words message identifies the content of conversation message when that is, artificial customer service does not intervene processing also, can recognize that conversation message Content when, corresponding pacifying property response message is extracted from the Service Source stored, and interact with customer in artificial customer service In the process, the robot customer service does not make any identification, from main process task or response to conversation message, but only carries out customer service client The transmission at end and the both sides' session of consumer user end, can be further improved the collaborative e-commerce of robot customer service Yu artificial customer service, into one Step alleviates the impatience in customer's waiting process.
Further, when being in session shape with robot customer service described in the session service at the consumer user end itself State, and when receiving from the conversation message of the customer service client, it will be from the session service with the consumer user end In session status be switched to session suspended state, and execute session between the customer service client and the consumer user end and disappear The transmission of breath operates, and in artificial and robot customer service cooperative working process, the priority of artificial customer service is higher, artificial customer service root It according to needs, can actively intervene at any time, therefore better service can be provided for customer.
Further, when robot customer service determines the conversation end between the customer service client and the consumer user end When, restore from the state in the subsequent session business between the consumer user end to be session status, i.e., in artificial visitor When clothes intervention is completed, the state of robot customer service is automatically switched back into, realizes the seamless rank of artificial customer service Yu robot customer service It connects, so as to advanced optimize the service experience of customer.
Further, by sessionless message transmission between the determination customer service client and the consumer user end up in advance If duration, or when detecting that session window is closed or detect in conversation message that the customer service client is sent comprising aobvious When showing the content that session is completed, signal is completed in the session that the customer service client is sent automatically, can automatically identify artificial visitor Clothes intervention is completed, and so as to reduce the operation of artificial customer service, reduces the workload of artificial customer service, is artificial customer service provider Just.
Further, by by it is described can not be from pacifying property response corresponding to main process task or unrecognized conversation message Message can not be sent to the customer service client with described from together with main process task or unrecognized conversation message, so that artificial visitor Clothes it can be seen that robot customer service to the content of the response message of customer replies, convenient for the continuity and consistency of hold acknowledge, So as to optimize consistent formula service experience, the user experience of customer is improved.
Further, disappeared by the pacifying property response for characterizing the state that customer service is in searching problem answers to consumer feedback Breath, finds for the content of conversation message with customer service client and provides the process of solution and be performed simultaneously, on the one hand can be with Reflect the virtual condition of customer service and be manually entered without customer service, has not only been saved the response operation time of artificial customer service, but also can be right Customer's mood plays the role of pacifying;On the other hand but also artificial customer service can will be put into and solve the problems, such as customer the more time In the middle, so as to more promptly providing the response for more enabling Customer Satisfaction.
Detailed description of the invention
Fig. 1 is a kind of robot client service method flow chart in the embodiment of the present invention;
Fig. 2 is another robot client service method flow chart in the embodiment of the present invention;
Fig. 3 is a kind of consumer user end interactive interface schematic diagram in the embodiment of the present invention;
Fig. 4 is a kind of robot customer service structural schematic diagram in the embodiment of the present invention;
Fig. 5 is another robot customer service structural schematic diagram in the embodiment of the present invention;
Fig. 6 is a kind of intelligent customer service system structure diagram in the embodiment of the present invention.
Specific embodiment
As previously mentioned, current customer service system, in customer service processes, machine customer service and artificial customer service are each independently For customer service, i.e., first serviced by machine customer service.When customer thinks that machine customer service can not solve the problems, such as its proposition, then Artificial customer service is clicked manually, is seeked advice to artificial customer service.Above scheme there are robot customer service mechanicalness repeat answer it is easy Enrage customer, and artificial customer service can not Always Online" and usual customer it is numerous, it is difficult to reply in time, therefore current customer service system It is difficult to effectively meet the demand for services of customer.
In view of the above-mentioned problems, the embodiment of the present invention is cooperated by robot customer service and artificial customer service, by robot visitor Clothes are unified to receive the conversation message from consumer user end, identifies the content of conversation message and determine whether can be from main process task institute Conversation message is stated, when being determined to from main process task, direct response, and from main process task or described disappear can not can not be identified in determination When the content of breath, on the one hand, session extension will be set as from the state in the subsequent session business at the consumer user end On the other hand the state of rising sends the pacifying property response message with the content matching of the conversation message in time, can be to avoid customer It cannot get reasonable response and generate impatience, and manually selected without customer, while the session at the consumer user end being disappeared Breath is sent to customer service client by manually carrying out response, when the robot customer service can not be effectively treated, may be implemented artificial The timely intervention of customer service and effectively solve the problems, such as the customer, and customer is not felt by robot customer service and artificial customer service It distinguishes, the mood in waiting answering is available effectively to pacify, and obtains better Interactive Experience.
It is understandable to enable above-mentioned purpose of the invention, feature and beneficial effect to become apparent, with reference to the accompanying drawing to this The specific embodiment of invention is described in detail.
The embodiment of the invention provides a kind of robot client service methods, carry out in detail referring to Fig. 1, and by specific steps It describes in detail bright.
S11 receives the conversation message from consumer user end.
In specific implementation, the form of the conversation message can there are many.For example, it may be natural language text forms Sentence.
Customer can be described description object by the sentence that the natural language forms.For example, if description object For product, the conversation message then can be the products such as the size to product, shape, color, quality, date of manufacture, shelf-life The advisory message of relevant information is also possible to the advisory message to after services of product.If description object is service, the session Message then can be service content, and service duration, points for attention provide the consulting for servicing related content such as attendant of service Information.
In specific implementation, the conversation message, can also be with voice, image or view other than being presented with written form The form of frequency is presented.For example, customer is conversated by way of audio or instant audio session or video conversation with customer service.
In specific implementation, customer can also be conversated in the form of picture with customer service consumer user end.It is described Picture can be static images with semantic information, dynamic picture or it is combined or customer's session is targeted product Or the photo or screenshotss information of service object.
In specific implementation, customer can also be sent out the link comprising product or service object by way of conversation message It sees off.
It is understood that in specific implementation, the conversation message can also be natural language text composition sentence, Two or more the form such as data link, voice, image or video is implemented in combination with.For example, customer passes through voice on one side Or video and customer service session, corresponding data link is sent to customer service by consumer user end on one side.
S12 identifies the content of the conversation message, when identifying the content of the conversation message, executes step S13, When that can not identify the content of the conversation message, step S15 is executed.
In specific implementation, the content that the conversation message can be identified by robot customer service, disappears according to different sessions Breath form can carry out identifying processing using different recognition methods.For example, can identify nature by participle and semantics recognition The sentence of spoken and written languages composition.For another example, participle and semanteme can be further carried out by the way that the voice received is switched to text Identification.If in the conversation message include image, can in image text or pattern row extract and make further Identification wait.
S13, according to the content of the conversation message, it is determined whether can be from conversation message described in main process task, when determining energy It is enough from conversation message described in main process task when, execute step S14;When determination from main process task or can not can not identify the conversation message When, execute step S15.
In specific implementation, some conversation messages possibly can not identify, for example, the information content such as video or image it is very big or The form of message of person's complexity may exceed the recognition capability of robot customer service.Alternatively, the content robot visitor of some conversation messages Although clothes can identify, discovery has exceeded itself autonomous processing capacity after identification.For can from the conversation message of main process task, Step S13 can be then executed, by robot customer service from main process task, for that can not disappear from main process task or unrecognized session Breath, then can execute step S15.
S14, the matched response message of content institute of extraction and the conversation message from the Service Source stored, and to The consumer user end is sent.
In specific implementation, the Service Source may include language empirical data and customer service resource data.Robot visitor Clothes, can be according to the content of the conversation message, from the language experience stored after the content for identifying the conversation message The matched response message of institute is extracted in data, or the matched response message of institute is extracted from the customer service resource data stored. For example, when the content for identifying the conversation message is " hello ", by being extracted in the language empirical data stored To matched response message are as follows: " you are good,.There is anything that can service for you? ".For another example, described in identify The content of conversation message are as follows: " one-piece dress that may I ask that part blue that I buys is delivered? " when, then it can be according to language experience number According to and customer service resource data, therefrom extract the corresponding order information of the customer, obtain response message are as follows: " you are good!What you bought One-piece dress is shipped ".The logistics information of the order can be also enclosed in the response message.
S15 will be set as session suspended state in the state in the subsequent session business at the consumer user end certainly, It extracts and is sent to the consumer user end with the pacifying property response message of the content matching of the conversation message, and can not by described in Customer service client is sent to from main process task or unrecognized conversation message.
In specific implementation, some message contents, robot customer service possibly from main process task or can not can not identify.For Such case, on the one hand, robot can will be arranged from the state in the subsequent session business at the consumer user end For session suspended state, after being set as session suspended state, in the subsequently received conversation message from the consumer user end When, the robot customer service will not reprocess, and can save the process resource of robot customer service;On the other hand, extract with it is described The pacifying property response message of the content matching of conversation message is simultaneously sent to the consumer user end to pacify customer's mood, simultaneously will It is described to be sent to customer service client from main process task or unrecognized conversation message, it is handled by artificial customer service intervention.In machine Device people is in session suspended state, and in artificial customer service treatment process, the robot customer service will be received from consumer user end It is transferred directly to other side with the conversation message of customer service client, without doing any identification or processing.
Can not be from main process task or when can not identify conversation message in robot, machine customer service is extracted to disappear with the session The pacifying property response message for ceasing content matching refers to the message that the uneasiness of customer and negative feeling can be made to settle out, Ke Yitong Cross the presentation such as nature word language, picture, voice, music.In specific implementation, for a long time in order to avoid the conversation message of customer It cannot get response and generate mood restless, the pacifying property response message is in searching problem suitable for characterization customer service and answers The state of case.For example, when the content for identifying conversation message are as follows: " I wants to buy double Men's Leather Shoes ", robot customer service determination can not When from main process task, on the one hand by from be set as session suspended state in the subsequent session business at the consumer user end; On the other hand, matching obtains pacifying property response message are as follows: " uh, it is good.Waiting, I goes to look to be suitble to your shoes.", it will be described Pacifying property response message is sent to consumer user end, so that customer is replied in time and patiently waited, and by " I wants to buy This conversation message of double Men's Leather Shoes " is sent to customer service customer side, so that artificial customer service is handled.One also can be set in advance Corresponding the pacifying property response message when robot customer service can not identify the content of specific conversation message a bit, such as connecing The picture received, when can not identify, corresponding pacifying property response message can be set to " just a moment,please, I first has a look ".
It, can be timely and effective from above embodiments it is found that by the robot customer service collaborative work with artificial customer service automatically Solve the problems, such as customer, and customer is the difference for being not felt by robot customer service Yu artificial customer service, in waiting answering Mood it is available effectively pacify, obtain better Interactive Experience.
In specific implementation, further extension or optimization can also be made to above-mentioned client service method as needed.Lead to below Specific embodiment is crossed to be described in detail.
Referring to Fig. 2, the embodiment of the invention provides another robot client service methods, carry out below by way of specific steps detailed Thin description.
S21 receives the conversation message from consumer user end.
In specific implementation, as previously mentioned, the form of the conversation message can there are many.For example, it may be natural language Say the knot of any of them forms such as sentence, data link, voice, image or the video of text composition or diversified forms It closes, which is not described herein again.
S22 is determined from the state in session service and is thened follow the steps S27 if session suspended state;Such as non-meeting Suspended state is talked about, S23 is thened follow the steps.
In order to save the process resource of robot customer service, the collaborative work of robot customer service and artificial customer service is realized, it can be with The state of conversation message by robot customer service from main process task from consumer user end is set as " session status ", or " non-hang-up State ", and can not be from main process task by robot customer service or the state that can not identify when the content of conversation message is set as that " session is hung The state of rising ".
If robot customer service determine from the state in the session service at the consumer user end be non-session hang up State thens follow the steps S23.If robot customer service determines the state itself being in the session service at the consumer user end and is Session suspended state then no longer will independently return the subsequent Customer Problems from the consumer user end under current sessions It is multiple, but be sent to customer service client and artificial customer service is transferred to handle, i.e. execution step S27.
S23 identifies the content of the conversation message, if it is possible to identify, then follow the steps S24;It is no to then follow the steps S26。
In specific implementation, the content that the conversation message can be identified by robot customer service, disappears according to different sessions Breath form can carry out identifying processing using different recognition methods.For example, can identify nature by participle and semantics recognition The sentence of spoken and written languages composition.For another example, participle and semanteme can be further carried out by the way that the voice received is switched to text Identification.If in the conversation message include image, can in image text or pattern row extract and make further The operation such as identification.If can identify, S24 is thened follow the steps.
It should be noted that some conversation messages possibly can not identify, for example, the information content such as video or image it is very big or At this moment the form of message of person's complexity may can execute step S26 beyond the recognition capability of robot customer service.
S24, it is determined whether can be from main process task, if so, thening follow the steps S25;If not, thening follow the steps S26.
Robot customer service can be according to the content of the conversation message identified, it is determined whether can be from main process task.
In specific implementation, robot can be confirmed whether be deposited from main process task according to the Service Source stored The Service Source of storage may include stored language empirical data and customer service resource data.
Specifically, can determine whether the content of the conversation message can be independently according to the language experience number first According to answering, when being determined to independently be answered according to the language empirical data, it is determined whether can be from main process task;When determining nothing When method is independently answered according to the language empirical data, then scene belonging to the content of the conversation message may further be determined Classification, when there are when scene type belonging to the content of the conversation message, determine that the robot customer service can be autonomous for determination Handle the conversation message.
Wherein, it is the ability for improving robot autonomous processing, the training of big data machine can be carried out to robot customer service, with Identify scene type belonging to conversation message.It in specific implementation, can in advance will be between consumer user end according to scene Conversation message carries out classification fractionation and obtains the customer service resource data of different scenes classification.
In an embodiment of the present invention, in scene category classification and machine learning process, the conversation message is not only wrapped Conversation message between robot customer service and consumer user end is included, further includes the conversation message of artificial customer service Yu consumer user end, All conversation messages are subjected to classification fractionation according to scene, do not take artificial customer service as independent individual here, but conduct The element of some specific scene in entirely service link, to entire customer service operating mode be reconstructed, in the feelings of a large amount of customers Under condition, the working efficiency of artificial customer service, optimization customer service experience can effectively improve.Wherein, the element of scene may include Customer and customer service and the two are directed to the conversation message content of different service scenarios.
S25 extracts the matched response message of institute from the Service Source stored and is sent to the consumer user end.
As previously mentioned, it may include stored language empirical data that Service Source, which may include stored Service Source, With customer service resource data.Correspondingly, in specific implementation, when being determined to independently be answered according to the language empirical data, It is extracted from the language empirical data stored and the matched response message of the content of conversation message institute;When determination can not be certainly When main root is answered according to the language empirical data, belonging to extracted in the customer service resource data stored under scene with the session The matched response message of content institute of message.
S26, by from be set as suspended state in the subsequent session business at the consumer user end, and to the Gu Objective client replys pacifying property response message corresponding with the conversation message, and the conversation message and the pacifying property are answered It answers message and is sent to customer service client.
In specific implementation, can by it is described can not be from pacifying property corresponding to main process task or unrecognized conversation message Response message can not be sent to the customer service client with described from together with main process task or unrecognized conversation message, so that people Work customer service can see robot customer service to the content of the response message of customer replies, convenient for the continuity of hold acknowledge and consistent Property, so as to optimize consistent formula service experience, improve the user experience of customer.
The conversation message is transmitted to corresponding customer service client by S27.
When robot customer service is in session suspended state, it is only necessary to the transmission for the message that conversates and need not carry out it is any From main process task, the conversation message from customer's client and customer service client is transmitted to other side respectively, is intervened by artificial customer service Processing, robot customer service only conversate the forwarding of message, without doing any identification, processing or response, so as to save The Service Source of robot customer service.
As can be seen from the above embodiments, when receiving from the conversation message at consumer user end, robot customer service itself is first First detect and itself whether be in session suspended state, and determine itself be not in session suspended state when, just by the machine People's customer service executes the operation for identifying the content of the conversation message, and when determining that itself is in session suspended state, then by people Work customer service is conversated interaction by customer service client and customer, to realize that artificial and robot customer service preferably cooperates with work Make, on the one hand, when robot customer service is in session suspended state, it is only necessary to the transmission for the message that conversates and need not be appointed What can save the Service Source of robot customer service from main process task, and on the other hand, robot customer service only itself is being not in session It could work when suspended state, the priority of old friend's work customer service is higher, session can be seized to avoid robot customer service midway, so that Customer can obtain consistent formula service experience, improve the user experience of customer.
In specific implementation, it in order to further alleviate the impatience in customer's waiting process, is in robot customer service It, i.e., can when artificial customer service does not intervene processing also if not receiving the conversation message from customer service client when session suspended state It is identified with the content to conversation message, in the content that can recognize that conversation message, from the Service Source stored Extract corresponding pacifying property response message;And receiving the conversation message from customer service client, i.e., after manpower intervention processing, In artificial customer service and customer's interactive process, the robot customer service will not make any identification to conversation message, from main process task or Response, but only carry out the transmission of customer service client and the both sides' session of consumer user end.
In specific implementation, when determining the conversation end between the customer service client and the consumer user end, machine Device people customer service can independently restore from the state in the session service between the consumer user end to be session status, i.e., Non-session suspended state.
When artificial customer service intervenes completion, by automatically switching back into the state of robot customer service, artificial customer service is realized With the seamless connection of robot customer service, so as to advanced optimize the service experience of customer.In specific implementation, can have more Kind mode determines the conversation end between the customer service client and the consumer user end.For example, determining the customer service client Whether the transmission of sessionless message reaches preset duration between end and the consumer user end, when reaching preset duration, it is determined that institute State the conversation end between customer service client and the consumer user end.For another example, if detect from the customer service client Session complete signal, when detecting that session from the customer service client completes signal, it is determined that the customer service client With the conversation end between the consumer user end.In an embodiment of the present invention, the customer service client can detect The customer service client issues the session when session window is closed and completes signal.In an alternative embodiment of the invention, work as inspection Measure when the content completed in the conversation message that the customer service client is sent comprising display session that the consumer user end can also Signal is completed to issue the session.
It is understood that in specific implementation, as needed, the method in above-mentioned difference embodiment, which can combine, to be made With.
Preset duration is reached by sessionless message transmission between the determination customer service client and the consumer user end, or Person is when detecting that session window is closed or detects complete comprising display session in conversation message that the customer service client is sent At content when, signal is completed in session that the customer service client is sent automatically, can be automatically identified artificial customer service and intervened At, so as to reduce the operation of artificial customer service, reduce the workload of artificial customer service, for artificial customer service provide conveniently.
In specific implementation, when with itself be in session status in the session service at the consumer user end, and receive When to conversation message from the customer service client, by from the session shape in the session service at the consumer user end State is switched to session suspended state, and executes the transmission behaviour of conversation message between the customer service client and the consumer user end Make, i.e., in artificial and robot customer service cooperative working process, the priority of artificial customer service is higher, and artificial customer service is as needed, It can actively intervene at any time, so as to provide better service for customer.
It is understood that being to receive a conversation message from consumer user end described in above-described embodiment When process flow, each conversation message received can be handled according to the above process.
To more fully understand those skilled in the art and realizing the present invention, below by way of a specific session context into Row is described in detail.
Referring to consumer user end interactive interface schematic diagram shown in Fig. 3.The interactive interface is the meeting of customer A and customer service B Interactive interface is talked about, is shown by the consumer user end of customer A.Wherein, in intelligent customer service system, no matter robot customer service is also It is artificial customer service, the vivid customer service B unified as one shows that active service person may be robot customer service to customer A, Either some subsequent artificial customer service of customer service client.The box A and B respectively indicates the mark letter of customer A and customer service B Breath, in specific implementation, can use the customer A and the information such as the customized head portrait of customer service B and the pet name in the box A and B It is shown.Temporal information " 5:44 in afternoon ", " 5:50 in afternoon " in dotted line frame 38,39 indicate the consumer user end circle of customer A At the time of the session that face is shown occurs.The conversation message content that right side conversation message is issued by customer A, left side conversation message are The response message content of customer service B.Below to this session interaction during, treatment process performed by intelligent customer service system carry out It is discussed in detail:
Customer A sends conversation message " hello " by consumer user end, and robot customer service, which confirms, itself is not in session extension When the state of rising, the conversation message is identified, whether after identification obtains, first confiring that can be according to language empirical data Answer, when determination can answer according to language data, extract the matched response message of institute " you are good,.What has can Think what you serviced? " and it is sent to the consumer user end of customer A.Customer A then passes through consumer user end and sends conversation message " I want buy double leather shoes ", robot customer service confirms itself be not in session suspended state and identified after, confirmation can not be autonomous Processing, then extract with conversation message " I want buy double leather shoes " institute matched pacifying property response message " uh, it is good.Waiting, I goes to see See the shoes for being suitble to you.", meanwhile, the robot customer service can will be from subsequent with the consumer user end of the customer A Session state setting in session service is session suspended state, while the conversation message " I wants to buy double leather shoes " being sent to Customer service client is handled by artificial customer service.Customer A replys conversation message " good, to thanks ".
It should be noted that artificial customer service may not in position or since artificial customer service needs the customer serviced too many, For the moment can not intervention in time processing, it is such crossed a period of time, customer A further through consumer user end issue conversation message " may I ask Get well? ", robot customer service receive the conversation message " please say hello? " afterwards, detect that itself being in session hangs up shape State then identifies the content of the message, extracts and replys corresponding pacifying property response message " waiting, I is also seeing ", concurrently It send to the consumer user end of the customer A.
Later, artificial customer service intervention, sending conversation message by customer service client, " you are good, I feels this following shoes very It is suitble to you.Http//www.tmall.com/ ", the message that robot customer service sends the artificial customer service with no treatment, It is transferred directly to the consumer user end of the customer A, after customer A sees the conversation message, reply conversation message " uh, it is good, I consider look at ", artificial customer service after customer service client receives this conversation message, successively reply response conversation message " uh, it is good 's.You are good can to contact me at any time ", and " thank to your presence, 88 ".
According to above-mentioned process flow, conversation message 31,32 is the message of robot autonomous processing and response, conversation message 33 It is robot customer service pacifying property response message generated with 34, conversation message 35,36 and 37 is the session of artificial customer service response Message.
It is understood that in specific implementation, conversation message 36 and 37 may be what robot customer service automatically generated Response message.If for example, the conversation message robot customer service of customer service response receive conversation message " uh, it is good, I consider see See " after, it does not receive the response message from customer service client in preset duration, can be confirmed that manpower intervention terminates, then may be used To identify the content of the conversation message.It, in an embodiment of the present invention, can be with after the content for identifying the conversation message According to the language empirical data stored, directly answer " uh, it is good.You are good can to contact me at any time " and " thank you Presence, 88 ".In an alternative embodiment of the invention, true according to language empirical data after the content for identifying the conversation message Surely it can not independently answer, then match the corresponding scene type of the conversation message, and the extraction from the Service Source stored Match response message " uh, it is good.You are good can to contact me at any time " and " thank to your presence, 88 ", and be successively sent to The consumer user end of the customer A.
To more fully understand those skilled in the art and realizing the present invention, additionally providing in the embodiment of the present invention can be real The robot customer service of existing above-mentioned client service method.Referring to Fig. 4, a kind of the specific of robot customer service in the embodiment of the present invention is introduced Structure.
As shown in figure 4, robot customer service 40 includes: first message interactive unit 41, recognition unit 42, ability judging unit 43, storage unit 44, the first response message generation unit 45, status control unit 46,47 and of the second response message generation unit Second message interactive unit 48, in which:
First message interactive unit 41 suitable for receiving the conversation message from consumer user end, and returns and the session The matched response message of content institute of message, including the robot customer service in the subsequent session business with the consumer user end In state be session suspended state when pacifying property response message.
Recognition unit 42, suitable for identifying the content of conversation message received by the first message interactive unit 41.
Ability judging unit 43, suitable for the content of the conversation message obtained according to the recognition unit 42 identification, really Whether determine can be from conversation message described in main process task.
Storage unit 44 is suitable for storage service resource.
First response message generation unit 45, suitable for disappearing when the ability judging unit 43 is determined to handle the session When breath, extracts from the Service Source that the storage unit 44 stores and disappear with the matched response of the content of the conversation message Breath.
Status control unit 46 is suitable for when 43 determination of ability judging unit can not be from main process task or the recognition unit 42 when can not identify the conversation message, control the robot customer service in the subsequent session business at the consumer user end State be session suspended state.
Second response message generation unit 47, suitable for the status control unit 46 control the robot customer service with When state in the subsequent session business at the consumer user end is session suspended state, from the clothes of the storage unit 44 storage The pacifying property response message with the content matching of the conversation message is extracted in resource of being engaged in.
Second message interactive unit 48, suitable in the robot customer service in the subsequent session industry with the consumer user end When state in business is session suspended state, the conversation message from consumer user end is sent to customer service client.
It can be timely by the robot customer service collaborative work with artificial customer service automatically using the scheme of above-described embodiment The problem of efficiently solving customer, and customer is the difference for being not felt by robot customer service Yu artificial customer service, is waiting mistake to be answered Mood in journey is available effectively to pacify, and obtains better Interactive Experience.
In specific implementation, the Service Source that the storage unit 44 is stored may include: language empirical data and visitor Take resource data.
In specific implementation, the message for characterizing the state that customer service is in searching problem answers can be answered as pacifying property Answer message.
In specific implementation, the form of the conversation message may include it is following any one or more: natural language text Sentence, data link, voice, image and the video of word composition.
In specific implementation, described image can be diversified forms, such as the static state with semantic information or can move State picture, perhaps for the targeted product of customer's session or service object photo or screenshotss information or be any of the above two The combination of kind or two or more forms is formed by picture group or sequence of pictures.
In specific implementation, as needed, further extension or optimization, reference can also be made to above-mentioned robot customer service Fig. 5, the embodiment of the invention also provides another robot customer services.It is repeated no more with above-described embodiment structure same section.
With the robot customer service in above-described embodiment the difference is that, as shown in figure 5, robot customer service 50 is in machine It on the basis of people's customer service 40, may also include that machine training unit 51, be suitable for carrying out the robot training of big data machine, To identify that scene type belonging to conversation message, the scene type are that the session between consumer user end disappears according to scene Breath carries out classification fractionation and obtains.
In specific implementation, the ability judging unit 43 may include the first judgment sub-unit 431 and the second judgement Unit 432, in which:
First judgment sub-unit 431 is adapted to determine that whether the robot customer service can be independently according to the language Empirical data is answered the content of the conversation message;
Second judgment sub-unit 432, be suitable for when first judgment sub-unit 431 determine the robot customer service without When method is independently answered according to the language empirical data, scene type belonging to the content of the conversation message is determined.
In specific implementation, the first response message generation unit 45 may include the first generation subelement 451 and Two generate subelement 452, in which:
Described first generates subelement 451, is suitable for independently being passed through according to the language when the determining robot customer service When testing data and answering, extracted from the language empirical data that the storage unit is stored and the content of conversation message institute The response message matched;
Second generates subelement 452, is suitable for when the determining robot customer service can not be independently according to the language experience number When according to answering, under the affiliated scene of extraction in the customer service resource data that the storage unit is stored and in the conversation message Hold the matched response message of institute.
In specific implementation, as shown in figure 5, robot customer service 50 may also include that state detection unit 52, it is suitable for when described When first message interactive unit 41 is received from the conversation message at consumer user end, detect the robot customer service with it is described State in the session service at consumer user end, and work as and detect the robot customer service in the session with the consumer user end When state in business is not in session suspended state, triggers the recognition unit and execute the interior of the identification conversation message The operation of appearance.
Using the robot customer service in the embodiment of the present invention, artificial and robot customer service may be implemented and preferably cooperate with work Make, on the one hand, when robot customer service is in session suspended state, it is only necessary to the transmission for the message that conversates and need not be appointed What can save the Service Source of robot customer service from main process task, and on the other hand, robot customer service only itself is being not in session It could work when suspended state, the priority of old friend's work customer service is higher, session can be seized to avoid robot customer service midway, so that Customer can obtain consistent formula service experience, improve the user experience of customer.
In specific implementation, it is the collaborative e-commerce for further increasing robot customer service Yu artificial customer service, alleviates customer and waited Impatience in journey, the recognition unit 42 be further adapted for when with robot described in the session service at the consumer user end Customer service is in session suspended state, and the second message interactive unit does not receive the session from the customer service client and disappears When breath, the conversation message from consumer user end that the first message interactive unit receives is identified;Second response disappears Generation unit 47 is ceased, is further adapted for the content of the conversation message identified according to the recognition unit, is deposited from the storage unit 44 The pacifying property response message with the content matching of the conversation message is extracted in the Service Source of storage.
In specific implementation, the status control unit 46 is further adapted in the robot customer service and the consumer user Session status is in the session service at end, and the second message interactive unit receives the meeting from the customer service client When talking about message, the robot customer service is being switched to session extension with the state in the subsequent session business at the consumer user end The state of rising, and control the second message interactive unit and execute message exchange operations.Pass through the control of the status control unit 46 System can make in artificial and robot customer service cooperative working process, and the priority of artificial customer service is higher, artificial customer service according to It needs, can actively intervene at any time, therefore better service can be provided for customer.
In specific implementation, as shown in figure 5, robot customer service 50 may also include conversation end detection unit 53, suitable for sentencing Whether the session broken between the customer service client and the consumer user end terminates;Correspondingly, the status control unit 52, It is further adapted for restoring the robot customer service when determining the conversation end between the customer service client and the consumer user end State between the consumer user end in subsequent session business is session status.
Cooperated by the conversation end detection unit 53 and the status control unit 52, intervenes and complete in artificial customer service When, the state of robot customer service can be automatically switched back into, realizes the seamless connection of artificial customer service Yu robot customer service, thus The service experience of customer can be advanced optimized.
In specific implementation, the conversation end detection unit 53 may include following at least one detection unit:
Whether first detection sub-unit 531 is adapted to detect for 48 free time of second message interactive unit up to preset duration, and When reaching preset duration, the conversation end between the customer service client and the consumer user end is determined;
Second detection sub-unit 532 is adapted to detect for the session from the customer service client and completes signal, and detecting When signal is completed in session from the customer service client, the meeting between the customer service client and the consumer user end is determined Words terminate.
Preset duration is reached by sessionless message transmission between the determination customer service client and the consumer user end, or Person is when detecting that session window is closed or detects complete comprising display session in conversation message that the customer service client is sent At content when, signal is completed in session that the customer service client is sent automatically, can be automatically identified artificial customer service and intervened At, so as to reduce the operation of artificial customer service, reduce the workload of artificial customer service, for artificial customer service provide conveniently.
In specific implementation, the second message interactive unit 48 is further adapted for from main process task or can not identifying described Conversation message corresponding to pacifying property response message and it is described can not be from one starting of main process task or unrecognized conversation message It send to the customer service client, so that artificial customer service can see robot customer service in the response message of customer replies Hold, improves the user's body of customer convenient for the continuity and consistency of hold acknowledge so as to optimize consistent formula service experience It tests.
To more fully understand those skilled in the art and realizing the present invention, below by way of specific embodiment to can be realized The intelligent customer service system of above-mentioned client service method describes in detail.
Referring to Fig. 6, intelligent service system in the embodiment of the present invention, may include: consumer user end 61, be suitable for it is described The customer service client 62 and robot customer service 63 that consumer user end 61 conversates, the customer service client and the consumer user End is communicated by the robot customer service.
In specific implementation, the consumer user end and customer service client can be presented in a variety of forms.For example, described Consumer user end and customer service client can be built in the program in browser and when detecting that browser page is opened Automatic running.For another example, the consumer user end and customer service client are independent applications client, are built in general-purpose computations The desktop computer of machine function, mobile phone, in car-mounted terminal and can start and run under the operation of customer tablet computer.For another example, institute It states consumer user end and customer service client can also be to be built in product supplier or ISP is special customized in customer With in equipment, the special equipment can be used for exhibiting and selling or service experience during customer and product supplier or ISP into Row consulting and feedback, preferably to meet customer demand.
It is understood that the way of realization at the consumer user end and customer service client conversed can be identical, it can also With difference, as long as compatible identical communication protocol can be realized session.
Wherein consumer user end 61, the specific structure of customer service client 62 and robot customer service 63 and working principle etc. are specific Implementation may refer to the introduction in the various embodiments described above, and which is not described herein again.
In specific implementation, robot customer service can simultaneously with the session of multiple consumer user ends, robot customer service The conversation message received can be forwarded to different customer service clients.When the consumer user end and customer service for being in session status When client is multiple, when robot customer service determination can not from main process task or when can not identify the conversation message, in order to Reply the conversation message of customer as early as possible, can be according to the task amount of the customer service client in session status, it will be described Conversation message and corresponding pacifying property response message from consumer user end are sent to the least customer service client of task amount.Example Such as, 10 artificial customer services are shared from the background, respectively correspond 10 different customer service clients, it can be according to current each customer service client Hold the quantity at consumer user end corresponding to the conversation message of processing, the conversation message that the customer service robot can will receive It evenly distributes to each customer service client.
In specific implementation, in order to make customer obtain preferably consistent formula service, while the clothes of each artificial customer service being improved Business efficiency, when the consumer user end and customer service client that are in session status are multiple, the robot customer service is in determination Can not be recorded according to conversation message from main process task or when can not identify the conversation message, will described in can not handle or nothing The conversation message of method identification processes the customer service client of consumer user end conversation message before being sent to.In the present invention one In embodiment, the robot customer service can also by by it is described can not handle or unrecognized conversation message corresponding to pacify Property response message be sent to together it is described before process the customer service client of consumer user end conversation message, to avoid people Work customer service does not know about the experience that customer is influenced with the exchanging information for customer.
It is understood that as needed, forwarded the messages in customer service client process in the robot customer service, it can It, can be by specific repeating process to consider that a variety of allocation strategies, each allocation strategy can configure different priority simultaneously It is matched, is not detailed herein according to priority orders.
Those of ordinary skill in the art will appreciate that all or part of the steps in the various methods of above-described embodiment is can It is completed with instructing relevant hardware by program, which can be stored in a computer readable storage medium, storage Medium may include: ROM, RAM, disk or CD etc..
Although present disclosure is as above, present invention is not limited to this.Anyone skilled in the art are not departing from this It in the spirit and scope of invention, can make various changes or modifications, therefore protection scope of the present invention should be with claim institute Subject to the range of restriction.

Claims (10)

1. a kind of robot customer service characterized by comprising
First message interactive unit suitable for receiving the conversation message from consumer user end, and returns and the conversation message The matched response message of content institute, the response message includes the robot customer service in the rear extended meeting with the consumer user end Pacifying property response message when state in call business is session suspended state;
Recognition unit, suitable for identifying the content of conversation message received by the first message interactive unit;
Ability judging unit, suitable for the content of the conversation message identified according to the recognition unit, it is determined whether energy Enough from conversation message described in main process task;
Storage unit is suitable for storage service resource;
First response message generation unit, suitable for when the ability judging unit is determined to handle the conversation message, from It is extracted in the Service Source of the storage unit storage and the matched response message of the content of conversation message institute;
Status control unit, suitable for that can not can not be identified from main process task or the recognition unit when the ability judging unit determination When the conversation message, controlling the robot customer service is being meeting with the state in the subsequent session business at the consumer user end Talk about suspended state;
Second response message generation unit, suitable for being extracted and the conversation message from the Service Source that the storage unit stores Content matching pacifying property response message;
Second message interactive unit, suitable for the robot customer service in the subsequent session business at the consumer user end When state is session suspended state, the conversation message from consumer user end is sent to customer service client;
The status control unit is further adapted in the session service at the robot customer service Yu the consumer user end in meeting Speech phase, and when the second message interactive unit is received from the conversation message of the customer service client, by the machine People's customer service is being switched to session suspended state with the state in the subsequent session business at the consumer user end, and controls described Two information interaction units execute message exchange operations.
2. robot customer service according to claim 1, which is characterized in that the Service Source packet that the storage unit is stored It includes: language empirical data and customer service resource data;The ability judging unit, comprising:
First judgment sub-unit, is adapted to determine that whether the robot customer service can independently answer according to the language empirical data The content of the conversation message;
Second judgment sub-unit, suitable for determining that the robot customer service can not be autonomous according to when first judgment sub-unit When language empirical data is answered, scene type belonging to the content of the conversation message is determined.
3. robot customer service according to claim 2, which is characterized in that the first response message generation unit, comprising:
First generates subelement, suitable for can independently be answered according to the language empirical data when the determining robot customer service When, it extracts from the language empirical data that the storage unit is stored and disappears with the matched response of the content of the conversation message Breath;
Second generates subelement, suitable for can not independently be answered according to the language empirical data when the determining robot customer service When, from the content institute extracted in the customer service resource data that the storage unit is stored under affiliated scene with the conversation message The response message matched.
4. robot customer service according to claim 1, which is characterized in that further include: state detection unit is suitable for when described When first message interactive unit is received from the conversation message at consumer user end, detect the robot customer service with the Gu State in the session service of objective client, and work as and detect the robot customer service in the session industry with the consumer user end When state in business is not in session suspended state, the content that the recognition unit executes the identification conversation message is triggered Operation.
5. robot customer service according to claim 4, which is characterized in that the recognition unit be further adapted for when with the Gu Robot customer service described in the session service of objective client is in session suspended state, and the second message interactive unit does not connect When receiving from the conversation message of the customer service client, identify that the first message interactive unit receives objective from customer The conversation message at family end;
The second response message generation unit, is further adapted for the content of the conversation message identified according to the recognition unit, from The pacifying property response message with the content matching of the conversation message is extracted in the Service Source of the storage unit storage.
6. robot customer service according to claim 1, which is characterized in that further include: conversation end detection unit, suitable for sentencing Whether the session broken between the customer service client and the consumer user end terminates;
The status control unit is further adapted for when the conversation end between the determining customer service client and the consumer user end When, restoring state of the robot customer service between the consumer user end in subsequent session business is session status.
7. robot customer service according to claim 6, which is characterized in that the conversation end detection unit include with down toward Few one kind:
First detection sub-unit, is adapted to detect for whether the second message interactive unit free time reaches preset duration, and default reaching When duration, the conversation end between the customer service client and the consumer user end is determined;
Second detection sub-unit is adapted to detect for the session from the customer service client and completes signal, and detecting from institute When stating the session completion signal of customer service client, the session knot between the customer service client and the consumer user end is determined Beam.
8. robot customer service according to claim 1, which is characterized in that the second message interactive unit is further adapted for institute State can not from pacifying property response message corresponding to main process task or unrecognized conversation message and it is described can not from main process task or Unrecognized conversation message is sent to the customer service client together.
9. a kind of intelligent customer service system characterized by comprising consumer user end, suitable for the consumer user end The described in any item robot customer services of customer service client and claim 1-8 of words, the customer service client and customer visitor Family end is conversated by the robot customer service.
10. intelligent customer service system according to claim 9, which is characterized in that the robot customer service is suitable for when in meeting When the consumer user end of speech phase and customer service client are multiple, when determination from main process task or can not can not identify the session When message, according to the task amount of the customer service client in session status, by the conversation message from consumer user end and Corresponding pacifying property response message is sent to the least customer service client of task amount;Alternatively, recorded according to conversation message, it will be described It can not handle or unrecognized conversation message is sent to the customer service client for processing consumer user end conversation message before End.
CN201811072289.2A 2015-12-30 2015-12-30 Robot customer service and intelligent customer service system Active CN109194568B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201811072289.2A CN109194568B (en) 2015-12-30 2015-12-30 Robot customer service and intelligent customer service system

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201511024389.4A CN105577529B (en) 2015-12-30 2015-12-30 Robot customer service and its client service method, intelligent customer service system
CN201811072289.2A CN109194568B (en) 2015-12-30 2015-12-30 Robot customer service and intelligent customer service system

Related Parent Applications (1)

Application Number Title Priority Date Filing Date
CN201511024389.4A Division CN105577529B (en) 2015-12-30 2015-12-30 Robot customer service and its client service method, intelligent customer service system

Publications (2)

Publication Number Publication Date
CN109194568A true CN109194568A (en) 2019-01-11
CN109194568B CN109194568B (en) 2020-07-14

Family

ID=55887218

Family Applications (3)

Application Number Title Priority Date Filing Date
CN201511024389.4A Active CN105577529B (en) 2015-12-30 2015-12-30 Robot customer service and its client service method, intelligent customer service system
CN201811072289.2A Active CN109194568B (en) 2015-12-30 2015-12-30 Robot customer service and intelligent customer service system
CN201811073663.0A Active CN109245994B (en) 2015-12-30 2015-12-30 Robot customer service method

Family Applications Before (1)

Application Number Title Priority Date Filing Date
CN201511024389.4A Active CN105577529B (en) 2015-12-30 2015-12-30 Robot customer service and its client service method, intelligent customer service system

Family Applications After (1)

Application Number Title Priority Date Filing Date
CN201811073663.0A Active CN109245994B (en) 2015-12-30 2015-12-30 Robot customer service method

Country Status (1)

Country Link
CN (3) CN105577529B (en)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114610199A (en) * 2022-03-21 2022-06-10 北京明略昭辉科技有限公司 Session message processing method, device, storage medium and electronic equipment
CN115277613A (en) * 2022-07-01 2022-11-01 浪潮通信信息系统有限公司 Robot chat system based on business
CN115473864A (en) * 2022-11-02 2022-12-13 平安银行股份有限公司 Robot chatting method, computer device and storage medium

Families Citing this family (26)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP6219448B1 (en) * 2016-05-16 2017-10-25 Cocoro Sb株式会社 Customer service control system, customer service system and program
CN106528789A (en) * 2016-11-01 2017-03-22 安徽爱依特科技有限公司 Smart service system based on robot
TWI649668B (en) * 2016-12-30 2019-02-01 金財通商務科技服務股份有限公司 Method of authentication and dialogue robot based on instant messaging program
CN108280521A (en) * 2017-01-06 2018-07-13 南通使爱智能科技有限公司 Intelligent robot service system
WO2019019175A1 (en) * 2017-07-28 2019-01-31 Beijing Yi Zhang Yun Feng Technology Co., Ltd. Automatic outbound instant messages
CN107609058B (en) * 2017-08-25 2021-05-18 维沃移动通信有限公司 Sales method, robot customer service end, artificial customer service end and user end
US10171662B1 (en) 2017-09-22 2019-01-01 International Business Machines Corporation Intervention in conversation between virtual agent and user
CN116541498A (en) 2018-01-04 2023-08-04 微软技术许可有限责任公司 Providing emotion care in a conversation
CN110300234A (en) * 2018-03-21 2019-10-01 上海易谷网络科技股份有限公司 IVR intelligent sound navigation silence seat technology
CN110392168B (en) * 2018-04-16 2021-06-01 华为技术有限公司 Call processing method, device, server, storage medium and system
CN108712583B (en) * 2018-06-05 2021-03-23 上海智蕙林医疗科技有限公司 Robot-based manual service method and system
CN108900726B (en) * 2018-06-28 2021-01-29 北京首汽智行科技有限公司 Artificial customer service switching method based on voice robot
CN108846127A (en) * 2018-06-29 2018-11-20 北京百度网讯科技有限公司 A kind of voice interactive method, device, electronic equipment and storage medium
CN109151008A (en) * 2018-08-09 2019-01-04 北京中关村科金技术有限公司 A kind of human-computer dialogue flow control method and system of unaware
CN109684459A (en) * 2018-12-28 2019-04-26 联想(北京)有限公司 A kind of information processing method and device
CN109787885A (en) * 2019-01-16 2019-05-21 中民乡邻投资控股有限公司 A kind of question and answer method of servicing
CN110019755A (en) * 2019-02-27 2019-07-16 杭州简简科技有限公司 Digital people's livelihood system and construction method based on artificial intelligence big data
CN110191242A (en) * 2019-05-21 2019-08-30 辽宁聆智科技有限公司 The interactive system based on telephone network that artificial intelligence is combined with artificial customer service
CN110149450A (en) * 2019-05-22 2019-08-20 欧冶云商股份有限公司 Intelligent customer service answer method and system
US11190469B2 (en) * 2019-07-26 2021-11-30 Avaya Management L.P. Enhanced digital messaging
CN110519469A (en) * 2019-08-23 2019-11-29 上海科技发展有限公司 Intelligent sound exchange method, system, medium and device applied to telephone service platform
CN111225115B (en) * 2019-11-25 2021-06-25 中国银行股份有限公司 Information providing method and device
CN112367553B (en) * 2020-11-10 2022-07-01 北京达佳互联信息技术有限公司 Message interaction method and device, electronic equipment and storage medium
CN112583896B (en) * 2020-11-30 2023-03-21 北京达佳互联信息技术有限公司 Session management method, session management device, electronic equipment, session management server and storage medium
CN112862508A (en) * 2021-03-19 2021-05-28 泰康保险集团股份有限公司 Man-machine cooperative processing method and device
CN114117157B (en) * 2021-11-19 2024-04-09 招联消费金融股份有限公司 Session processing method, apparatus, computer device and storage medium

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20070263838A1 (en) * 2006-04-07 2007-11-15 Brady Wiseman Method and system for informing customer service agent of details of user's interaction with voice-based knowledge retrieval system
CN103279528A (en) * 2013-05-31 2013-09-04 俞志晨 Question-answering system and question-answering method based on man-machine integration
CN104731895A (en) * 2015-03-18 2015-06-24 北京京东尚科信息技术有限公司 Auto-answer method and device
CN105162688A (en) * 2015-08-13 2015-12-16 北京智齿博创科技有限公司 Method for chatting with online customer service representative

Family Cites Families (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20060074831A1 (en) * 2004-09-20 2006-04-06 Hyder Andrew D Virtual assistant
CN101431573B (en) * 2007-11-08 2013-02-20 上海撼世网络科技有限公司 Method and equipment for implementing automatic customer service through human-machine interaction technology
CN102739557B (en) * 2011-04-08 2016-06-22 腾讯科技(深圳)有限公司 User's shunt method in customer service and device
CN105072173A (en) * 2015-08-03 2015-11-18 谌志群 Customer service method and system for automatically switching between automatic customer service and artificial customer service

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20070263838A1 (en) * 2006-04-07 2007-11-15 Brady Wiseman Method and system for informing customer service agent of details of user's interaction with voice-based knowledge retrieval system
CN103279528A (en) * 2013-05-31 2013-09-04 俞志晨 Question-answering system and question-answering method based on man-machine integration
CN104731895A (en) * 2015-03-18 2015-06-24 北京京东尚科信息技术有限公司 Auto-answer method and device
CN105162688A (en) * 2015-08-13 2015-12-16 北京智齿博创科技有限公司 Method for chatting with online customer service representative

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114610199A (en) * 2022-03-21 2022-06-10 北京明略昭辉科技有限公司 Session message processing method, device, storage medium and electronic equipment
CN115277613A (en) * 2022-07-01 2022-11-01 浪潮通信信息系统有限公司 Robot chat system based on business
CN115473864A (en) * 2022-11-02 2022-12-13 平安银行股份有限公司 Robot chatting method, computer device and storage medium

Also Published As

Publication number Publication date
CN109245994B (en) 2021-04-23
CN109245994A (en) 2019-01-18
CN105577529A (en) 2016-05-11
CN105577529B (en) 2019-03-05
CN109194568B (en) 2020-07-14

Similar Documents

Publication Publication Date Title
CN105577529B (en) Robot customer service and its client service method, intelligent customer service system
CN105553833B (en) Customer service system and its method of servicing, robot customer service
US11032419B2 (en) Intelligent customer service systems, customer service robots, and methods for providing customer service
JP7404467B2 (en) Dynamic communication routing based on consistency weighting and routing rules
US11057520B2 (en) System and method of sentiment modeling and application to determine optimized agent action
JP7259030B2 (en) Automated bot creation based on scripts
CN111460112A (en) Online customer service consultation method, device, medium and electronic equipment
WO2020125775A1 (en) Multi-service-party online customer service information processing method and apparatus
JP7309889B2 (en) Dynamic text message processing for endpoint communication channel selection
CN112306708A (en) Enhanced digital messaging
US8503664B1 (en) Quality review of contacts between customers and customer service agents
CN104769575A (en) Multiple output relaxation machine learning model
CN111131005A (en) Dialogue method, device, equipment and storage medium of customer service system
CN111045510A (en) Man-machine interaction method and system based on augmented reality
CN111405224A (en) Online interaction control method and device, storage medium and electronic equipment
CN110489519B (en) Session method based on session prediction model and related products
WO2020119568A1 (en) Dialogue emotion analysis method and device, storage medium and processor
CN111144906A (en) Data processing method and device and electronic equipment
CN105323155A (en) Method for communication of on-line service system with third party instant messaging system
CN110166349A (en) Sending method, device and the electronic equipment of message
CN112053165A (en) Information interaction method, device, server and storage medium
WO2021109741A1 (en) Serving method, apparatus, system and device, and storage medium
Chen Applying artificial intelligence in CRM: Case studies of intelligent virtual agents and pegasystems
CN112150178A (en) Intelligent glasses and application method thereof, intelligent system and application method thereof
CN114943545A (en) Information processing method, system and storage medium

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant
PE01 Entry into force of the registration of the contract for pledge of patent right
PE01 Entry into force of the registration of the contract for pledge of patent right

Denomination of invention: Robot customer service and intelligent customer service system

Effective date of registration: 20230223

Granted publication date: 20200714

Pledgee: China Construction Bank Corporation Shanghai No.5 Sub-branch

Pledgor: SHANGHAI XIAOI ROBOT TECHNOLOGY Co.,Ltd.

Registration number: Y2023980033272