CN108984777A - Customer service method, apparatus and computer readable storage medium - Google Patents
Customer service method, apparatus and computer readable storage medium Download PDFInfo
- Publication number
- CN108984777A CN108984777A CN201810824060.3A CN201810824060A CN108984777A CN 108984777 A CN108984777 A CN 108984777A CN 201810824060 A CN201810824060 A CN 201810824060A CN 108984777 A CN108984777 A CN 108984777A
- Authority
- CN
- China
- Prior art keywords
- user
- participle
- customer service
- character
- crucial
- 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
Links
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Character Discrimination (AREA)
Abstract
The present disclosure discloses a kind of customer service method, apparatus and computer readable storage mediums, are related to field of computer technology.Disclosed method includes: the character in the picture that identification user sends in customer service processes；Character is matched with object database, the corresponding object of identification character, object database includes the object information in e-commerce platform；The description information of object is supplied to user, so that user confirms that the object to be seeked advice from is object.Disclosed method user send picture in the case where, can intelligent recognition user be intended to, determine the object to be seeked advice from of user, make corresponding answer.Disclosed method allows user to send picture instead of being manually entered, and facilitates the consulting of user, improves customer service quality, promotes user experience.
Description
Technical field
This disclosure relates to field of computer technology, in particular to a kind of customer service method, apparatus and computer-readable deposit
Storage media.
Background technique
With the rapid development of artificial intelligence technology, the high speed development of customer service robot service.Promote customer service
Service quality become e-commerce platform management in important goal.
Currently, customer service robot is substantially the intention for understanding user by the text of identification user's input, identification
User wants the object of consulting.
Summary of the invention
Inventor's discovery: in actual application, user can send commodity screenshot to customer service robot for convenience
Or the pictures such as order screenshot are seeked advice from, and current customer service robot can not make these pictures expected from user instead
It answers, causes user experience poor.
A disclosure technical problem to be solved is: how to improve customer service quality, promotes user experience.
According to some embodiments of the present disclosure, a kind of customer service method for providing, comprising: user is in customer service for identification
The character in picture sent in the process；Character is matched with object database, the corresponding object of identification character, number of objects
According to the object information that library includes in e-commerce platform；The description information of object is supplied to user, so that user's confirmation is intended to consult
The object of inquiry is object.
In some embodiments, character is matched with object database, the corresponding object of identification character includes: by word
Symbol is segmented, and is filtered stop words and is corrected error character, obtains crucial participle；By crucial participle and pair in object database
As being matched, the corresponding object of crucial participle is identified.
In some embodiments, object database includes product name in e-commerce platform, the commodity in user's order
At least one of product name in title and user service list；By crucial participle and the object progress in object database
Match, identifying that key segments corresponding object includes: in the case where key participle is text, by crucial participle and object database
Middle product name is matched, and identifies at least one of the corresponding commodity of crucial participle, order, service list；Alternatively, object data
Library includes goods number in e-commerce platform, the single number of service or O/No.；It will be in key participle and object database
Object is matched, and identifies that the corresponding object of crucial participle includes: in the case where key participle is number, according to coding rule
Key participle is matched with goods number, the single number of service or O/No. in object database, determines crucial participle pair
The commodity answered, order, service are at least one of single.
In some embodiments, correcting error character includes: entangling the participle input training in advance after filtering stop words
Mismatch type obtains the correct probability of each participle, and the participle using correct probability lower than threshold value is segmented as mistake, utilizes shape phase
Close word replacement mistake participle, re-enters error correcting model and determines the correct probability of each participle, until choosing correct probability
Participle in the case of highest is as crucial participle；Alternatively, the participle after filtering stop words is matched with erroneous words dictionary, really
The participle of fixed mistake is simultaneously corrected.
In some embodiments, the description information of object is supplied to user includes: that object and user are seeked advice from context
In object matched；In the case where being matched to same object, the description information of object is supplied to user；It is not matching
In the case where same object, requry the users whether the object to be seeked advice from is object, confirms that the object to be seeked advice from is in user
In the case where object, the description information of object is supplied to user.
In some embodiments, the description information of object is supplied to user includes: to incite somebody to action in the case where object is multiple
Each object is supplied to user in the form of a list；The object to be seeked advice from that user selects from list is obtained, consulting is intended to
The description information of object is supplied to user.
In some embodiments, the text in picture is identified according to optical character recognition algorithms；Alternatively, object
Description information include that the title of object, picture, corresponding goods number, corresponding O/No., corresponding service are singly numbered
At least one of information.
According to other embodiments of the disclosure, a kind of customer service device for providing, comprising: character recognition module is used
Character in the picture that identification user sends in customer service processes；Object Identification Module is used for character and number of objects
It is matched according to library, the corresponding object of identification character, object database includes the object information in e-commerce platform；Information hair
Module is sent, for the description information of object to be supplied to user, so that user confirms that the object to be seeked advice from is object.
In some embodiments, Object Identification Module by character for segmenting, filtering stop words and correcting erroneous words
Symbol, obtains crucial participle；Key participle is matched with the object in object database, identifies that crucial participle is corresponding right
As.
In some embodiments, object database includes product name in e-commerce platform, the commodity in user's order
At least one of product name in title and user service list；Object Identification Module is used for the case where key participle is text
Under, key participle is matched with product name in object database, identifies the corresponding commodity of crucial participle, order, service
It is at least one of single；Alternatively, object database includes goods number in e-commerce platform, the single number of service or O/No.；
Object Identification Module is used in the case where crucial participle is number, will be in key participle and object database according to coding rule
Goods number, the single number of service or O/No. are matched, and determine that key segments corresponding commodity, order, service Dan Zhongzhi
One item missing.
In some embodiments, Object Identification Module is used to filter the error correction for segmenting input and training in advance after stop words
Model obtains the correct probability of each participle, and the participle using correct probability lower than threshold value is segmented as mistake, close using shape
Word replacement mistake participle, re-enter error correcting model and determine the correct probability of each participle, until choosing correct probability most
Participle in high situation is as crucial participle；Alternatively, Object Identification Module is for the participle and erroneous words after filtering stop words
Dictionary is matched, and is determined the participle of mistake and is corrected.
In some embodiments, information sending module is used to seeking advice from object and user into the object progress in context
Match；In the case where being matched to same object, the description information of object is supplied to user；In the feelings for not being matched to same object
Under condition, requry the users whether the object to be seeked advice from is object, it, will in the case where user confirms that the object to be seeked advice from is object
The description information of object is supplied to user.
In some embodiments, information sending module is used in the case where object is multiple, by each object with list
Form be supplied to user；The object to be seeked advice from that user selects from list is obtained, the description information of the object of consulting is intended to
It is supplied to user.
In some embodiments, the text in picture is identified according to optical character recognition algorithms；Alternatively, object
Description information include that the title of object, picture, corresponding goods number, corresponding O/No., corresponding service are singly numbered
At least one of information.
According to the other embodiment of the disclosure, a kind of customer service device for providing, comprising: memory；And coupling
To the processor of memory, processor is configured as based on the instruction being stored in memory devices, and execution such as realizes aforementioned
The customer service method for embodiment of anticipating.
According to the still other embodiments of the disclosure, a kind of computer readable storage medium provided is stored thereon with calculating
Machine program, wherein the program realizes the customer service method of aforementioned any embodiment when being executed by processor.
After receiving the picture that user sends in customer service processes in the disclosure, identifies the character in picture, pass through
The matching of character and object database, the corresponding object of identification character, and then the description information of object can be provided a user, with
Toilet states user and confirms that the object to be seeked advice from is the object.Disclosed method, can in the case where user sends picture
Intelligent recognition user is intended to, and determines the object to be seeked advice from of user, makes corresponding answer.Disclosed method makes user can be with
It sends picture to replace being manually entered, facilitates the consulting of user, improve customer service quality, promote user experience.
By the detailed description referring to the drawings to the exemplary embodiment of the disclosure, the other feature of the disclosure and its
Advantage will become apparent.
Detailed description of the invention
In order to illustrate more clearly of the embodiment of the present disclosure or technical solution in the prior art, to embodiment or will show below
There is attached drawing needed in technical description to be briefly described, it should be apparent that, the accompanying drawings in the following description is only this
Disclosed some embodiments for those of ordinary skill in the art without creative efforts, can be with
It obtains other drawings based on these drawings.
Fig. 1 shows the flow diagram of the customer service method of some embodiments of the present disclosure.
Fig. 2 shows the flow diagrams of the customer service method of other embodiments of the disclosure.
Fig. 3 shows the structural schematic diagram of the customer service device of some embodiments of the present disclosure.
Fig. 4 shows the structural schematic diagram of the customer service device of other embodiments of the disclosure.
Fig. 5 shows the structural schematic diagram of the customer service device of the other embodiment of the disclosure.
Specific embodiment
Below in conjunction with the attached drawing in the embodiment of the present disclosure, the technical solution in the embodiment of the present disclosure is carried out clear, complete
Site preparation description, it is clear that described embodiment is only disclosure a part of the embodiment, instead of all the embodiments.Below
Description only actually at least one exemplary embodiment be it is illustrative, never as to the disclosure and its application or making
Any restrictions.Based on the embodiment in the disclosure, those of ordinary skill in the art are not making creative work premise
Under every other embodiment obtained, belong to the disclosure protection range.
The disclosure proposes a kind of customer service method, which can use customer service device (customer service clothes
Be engaged in robot) it realizes, which belongs to automatic customer service method, automatic customer service, alternatively referred to as automatic visitor
Clothes, self-service customer service, electronic customer care, intelligent customer service, machine customer service, online customer service or automatic-answering back device customer service etc., are retouched below with reference to Fig. 1
State some embodiments.
Fig. 1 is the flow chart of some embodiments of disclosure customer service method.As shown in Figure 1, the method packet of the embodiment
It includes: step S102~S106.
In step s 102, the character in picture that identification user sends in customer service processes.
User when e-commerce platform is browsed, is done shopping, can to commodity, order or service it is single (for example,
Pre-sales or counseling services list etc. after sale) it is seeked advice from.User can choose unartificial customer service and seek advice from, and service
Picture can be sent to customer service device in journey, for example, commodity screenshot, order screenshot or the single screenshot of service etc..Electronics visitor
After family services client service unit receives the picture of user's transmission, OCR (Optical Character can use
Recognition, optical character identification) algorithm identification picture in character, the pixel in picture is converted to can with computer
The character indicated with the data of identification.
OCR algorithm generally includes picture pretreatment (such as binaryzation, except denoising, Slant Rectify etc.), Page Segmentation, word
Symbol segmentation, character feature extract, and character recognition (identifies character for example, by using support vector machines or neural network model), post-processing
Etc. processes.The picture indicated with pixel can be converted into editable character by the above process, character is in computer-internal
Specific binary coding representation can be used, be shown as editable text.OCR algorithm can refer to existing implementation,
Details are not described herein.
In step S104, character is matched with object database, the corresponding object of identification character.
Object database includes the object information in e-commerce platform, and object is, for example, commodity, order, service Dan Zhongzhi
One item missing.Object information can be at least one of product name, goods number, O/No., the single number of service information.Into one
Step, object information can also be commodity title or page layout information etc..User can be for any right of e-commerce platform
As being seeked advice from, what is stored in object database can be the information for being able to reflect characteristics of objects, be not limited to examples cited.
Character after OCR is identified can filter stop words and correct error character, removal repetitor by participle
The processes such as language obtain crucial participle, and then are matched again with object database, can reduce matching work in this way, more increase
Effect is accurately matched to relevant object.Participle can use existing algorithm, and details are not described herein.For example special symbol of stop words
Number, function word etc. for matching object effect very little participle.Deactivated dictionary can be set, carried out by that will segment with deactivated dictionary
It compares, to identify stop words and be filtered to stop words.
It is likely to occur error character by the character that OCR is identified, subsequent progress object matching is easy to appear and can not be matched to
The case where object or matching error.Therefore can be to by segmenting, the participle after filtering stop words carries out correcting error character
Processing.
In some embodiments, it can use error correcting model trained in advance to correct error character.Such as it can be with
The error character in participle after correcting filtering stop words using following methods.(1) participle is inputted into error correcting model, obtained each
The correct probability of participle.(2) participle of the correct probability lower than threshold value is chosen as mistake participle.(3) judging whether there is can be with
The word of replacement mistake participle, if it is present step (4) are executed, it is no to then follow the steps (5).It (4) will using word similar in shape
Mistake participle is replaced, and return step (1) restarts to execute.(5) the correct probability highest situation of each participle is chosen
Under participle as crucial participle.
Further, error correcting model can be the combination of word2vector and LR (logistic regression) model.It can pass through first
Trained word2vector generates the vector of participle, which can reflect the relevance of participle front and back word.By participle
Vector inputs trained LR model, the correct probability of available participle.LR model using the training sample that manually marks into
Row training, training sample may include the sentence and participle of various user's consultings, wherein each participle is labelled with correct or wrong
Accidentally.Vector forecasting correct probability of the neural network model as error correcting model to participle can also be used, examples cited are not limited to.It is right
It is segmented in mistake, can use word similar in shape and be replaced, the correspondence of each similar words or nearly word form can be stored in advance
Relationship.
For example, " rice " word is identified as " too rice ", when participle, may be divided into " too " and " rice " two words, pass through mistake
Correct model, will input model " too ", since word2vector vector can reflect the relevance of word before and after participle, in advance
First trained model can identify that correct probability " too " is lower, i.e., this word appears in the sentence compared with small probability, therefore, can
That " too " nearly word form " big " will be utilized to be replaced, error correcting model is re-entered, higher correct probability is obtained.Using nearly word form
Or after similar words replacement mistake participle, word segmentation processing can be re-started, then input error correcting model, calculate the correct general of participle
Rate.
In further embodiments, erroneous words dictionary can be set, wherein comprising common erroneous words and it is corresponding just
True word corrects error character using erroneous words dictionary.For example, by filter stop words after participle and erroneous words dictionary into
Row matching determines the participle of mistake and is corrected.It can be pre- by participle front and back when participle is matched with erroneous words dictionary
If other characters in range participate in matching together with participle, this is because erroneous words may be divided into different points when participle
Word.
The sensitive word in participle can be identified in yet other embodiments, by that will segment and the word in sensitive word dictionary
Language is compared, it is determined whether is sensitive word.In the case where including sensitive word in character, user can be prompted to be manually entered letter
Breath is seeked advice from.
By the method in previous embodiment, the character that OCR is identified is segmented, and is filtered stop words and is corrected mistake
After character, repetition participle can also be removed, crucial participle is obtained.Key participle is matched with object database, determining pair
The object answered.It is described below and key participle is subjected to matched some embodiments with object database.
In some embodiments, object database includes product name in e-commerce platform, the commodity in user's order
At least one of product name in title and user service list.Key participle be text in the case where, by key participle with
Product name is matched in object database, identifies at least one of the corresponding commodity of crucial participle, order, service list.
By crucial participle and the product name in e-commerce platform carry out it is matched at the same time it can also by key participle with
The product name in product name or user service list in the order of the user.It can identify whether user is intended to seek advice to have bought
Or the commodity seeked advice from or paid close attention to, to the information matches of commodity, related order and service list be arrived, so as to more quasi-
The intention of true identification user.
In some embodiments, object database includes goods number in e-commerce platform, the single number of service or order
Number.In the case where key participle is number, key is segmented and goods number, clothes in object database according to coding rule
At least one of business list is numbered or O/No. is matched, and determining key segments corresponding commodity, order, services list.
Different numbering types corresponds to different coding rules, and different regular expressions can be set, and key is segmented
In number matched with regular expression, determine type belonging to the number, and then will be in the number and object database
The number of corresponding types is matched.Determine that the key segments corresponding commodity, order or service list will pass through to number.Logical
Cross number be matched to commodity in the case where, can the similarly matched commodity whether the user order or service list in, in turn
Determine that the corresponding order of crucial participle or service are single.In the case where number accurately can not be matched to corresponding object, can incite somebody to action
Number carries out fuzzy matching, the object conduct for the preset quantity that will match to according to the number in preset rules and object database
Number corresponding object.
In some embodiments, object database includes the title of commodity in e-commerce platform.By key participle with it is right
The title of commodity is matched in image data library, identifies at least one of the corresponding commodity of crucial participle, order, service list.
The title of commodity can be subjected to word segmentation processing, key participle is compared with title participle, determines crucial point
The corresponding object of word.In the case where being matched to commodity by crucial participle, can the similarly matched commodity whether in the user
Order or service list in, and then determine and crucial segment corresponding order or service is single.Correspondence can not be accurately matched in number
Object in the case where, number according to the number in preset rules and object database can be subjected to fuzzy matching, will matched
The object of the preset quantity arrived is as the corresponding object of number.
In some embodiments, object database includes specific word, the word of specific word such as commodity, order, service etc,
It can indicate object type information.In the case where that can not be matched to specific object and crucial participle is text, key is segmented
It is matched with the specific word of expression object type information in object database, determines the classification of object.It is quotient in object type
In the case where product, can be interested using user or the commodity bought or the commodity of fast sale as the corresponding object of crucial participle.
It is more than the commodity of threshold value that the interested commodity of user, which include the commodity of user's concern, user's browsing time, and user's the last time is clear
The commodity etc. look at.Fast-selling commodity include the commodity that sales volume is more than corresponding threshold value in preset time.It is order in object type
In the case of, corresponding object is segmented using the order of the user as key.In the case where object type is service list, by the use
The service nonoculture at family is the corresponding object of crucial participle.
In some embodiments, multiple crucial participles are matched to identical object, such as are compiled by product name and commodity
Number the same commodity are matched to, result can be merged, remove duplicate object, so that more accurate determination user is intended to
The object of consulting.
In step s 106, the description information of object is supplied to user, so that user confirms that the object to be seeked advice from is pair
As.
Title of the description information of object for example including object, picture, corresponding goods number, corresponding O/No.,
At least one of corresponding service list number, state information.The state of object for example, in dispatching, the states such as be completed.It can
The description information of object is provided a user in the form of through BOOL card form pop-up window, and options for user confirmation is provided and is intended to
Whether the object of consulting is the object.
In some embodiments, object the object in context can be seeked advice from user to match；It is being matched to phase
In the case where with object, the description information of object is supplied to user；In the case where not being matched to same object, ask to user
Ask whether the object to be seeked advice from is object, in the case where user confirms that the object to be seeked advice from is object, the description of object is believed
Breath is supplied to user.If had already appeared in user's search record, the probability that user is intended to seek advice from the object is higher, Ke Yizhi
Connect the description information for providing a user object.It can record in real time and update the object in user's consulting context, user's consulting
Object in context can be the object obtained by identifying picture, be also possible to obtain by the text of identification user's input
Object, or the object obtained by the identification user information that (such as link) sends otherwise.
In some embodiments, in the case where object is multiple, each object is supplied to user in the form of a list；
The object to be seeked advice from that user selects from list is obtained, the description information for being intended to the object of consulting is supplied to user.It can lead to
It crosses BOOL card and list object is shown to user, user can select wherein one or more objects as being intended to consult by clicking
The object of inquiry.
In the case where character and product name carry out matched situation, multiple commodity may be matched to, or carry out mould in number
It pastes in matched situation, it is also possible to which specific commodity can not be accurately matched to by being matched to multiple commodity or other, be matched to multiple
In the case where dependent merchandise.The commodity that the interested commodity of user or the commodity bought or fast sale can be chosen, as object
It is supplied to user, selects wherein one or more to seek advice from for user.The interested commodity of user include the quotient of user's concern
Product, user's browsing time are more than the commodity of threshold value, the commodity etc. that user's the last time browses.Fast-selling commodity include preset time
Interior sales volume is more than the commodity of corresponding threshold value.The commodity that items list includes preset quantity can be provided a user.Construct commodity column
Table can choose preset quantity according to descending sequence of the descending sequence of user interest degree or commodity fast sale degree etc.
Commodity.
After the method for above-described embodiment receives the picture that user sends in customer service processes, the word in picture is identified
Symbol identifies the corresponding object of character by the matching of character and object database, and then can provide a user the description of object
Information, so that user confirms that the object to be seeked advice from is object.The method of above-described embodiment is in the case where user sends picture, energy
Enough intelligent recognition users are intended to, and determine the object to be seeked advice from of user, make corresponding answer.The method of above-described embodiment to use
Family can send picture and replace being manually entered, and facilitate the consulting of user, improve customer service quality, promote user experience.
Specific object can not may be all matched to by the matching of above-mentioned character and object database.It can use user's hair
The page layout information of the picture sent determines object.Object database includes page layout information, for indicating different type pair
As the layout of the corresponding page.For example, the page layout of commodity details page, order page and service single page is different.User sends
Screenshot can correspond to different page layout information, OCR technique can restore layout information i.e. page layout information therefore can
To determine object type by matching page layout information.It, can be interested by user in the case where object type is commodity
Or the commodity bought segment corresponding object as crucial.The interested commodity of user include the commodity of user's concern, user
Browsing time is more than the commodity of threshold value, the commodity etc. that user's the last time browses.In the case where object type is order, by this
The order of user segments corresponding object as crucial.In the case where object type is service list, by the service list of the user
Corresponding object is segmented as key.
Other embodiments of disclosure customer service method are described below with reference to Fig. 2.
Fig. 2 is the flow chart of other embodiments of disclosure customer service method.As shown in Fig. 2, the method for the embodiment
It include: step S202~S218.
In step S202, the picture that user sends in customer service processes is received.
In step S204, the character in OCR algorithm identification picture is utilized.
In step S206, character is segmented, filter stop words and corrects error character, obtains crucial participle.
In step S208, identify in crucial participle whether include sensitive word, if so, thening follow the steps S209, otherwise
Execute step S210.
In step S209, sensitive word response is carried out to user.
Such as prompt user's switching live customer service, and live customer service link is provided.]
In step S210, character is matched with object database, the corresponding object of identification character.
In step S212, the quantity of the object of identification is judged.In the case where unidentified object out, step is executed
S213 executes step S214 in the case where identifying multiple objects, in the case where identifying an object, executes step
S216。
In step S213, user is replied according to default answer-mode.
For example, the problem of prompt user's picture can not identify, input verbal description is to be seeked advice from.
In step S214, each object is supplied to user in the form of a list.
The user can be prompted selects wherein one or more objects to be seeked advice from.
In step S215, the information for the object to be seeked advice from that user selects from list is received.Execute step S218.
In step S216, object and user are seeked advice from into the object in context and matched, judge in context whether
There are the objects, if so, S218 is thened follow the steps, it is no to then follow the steps S217.
In step S217, requry the users whether the object to be seeked advice from is the object, if so, thening follow the steps
Otherwise S218 executes step S213.
In step S218, the description information of object is supplied to user.
It can prompt user that can inquire any problem for the object simultaneously.
The disclosure also provides a kind of customer service device, is described below with reference to Fig. 3.
Fig. 3 is the structure chart of some embodiments of disclosure customer service device.As shown in figure 3, the device of the embodiment
30 include: character recognition module 302, Object Identification Module 304, information sending module 306.
Character recognition module 302, the character in picture that user sends in customer service processes for identification.
In some embodiments, the text in picture is identified according to optical character recognition algorithms.
Object Identification Module 304, for matching character with object database, the corresponding object of identification character is right
Image data library includes the object information in e-commerce platform.
In some embodiments, Object Identification Module 304 by character for segmenting, filtering stop words and correcting mistake
Character obtains crucial participle；Key participle is matched with the object in object database, identifies that crucial participle is corresponding right
As.
Further, in some embodiments, participle input of the Object Identification Module 304 for after filtering stop words is preparatory
Trained error correcting model obtains the correct probability of each participle, and the participle using correct probability lower than threshold value is segmented as mistake, benefit
The replacement mistake participle of the word similar in shape, re-enters error correcting model and determines the correct probability of each participle, until choosing
Participle in the case of correct probability highest is as crucial participle.Alternatively, Object Identification Module 304 will be for after filtering stop words
Participle is matched with erroneous words dictionary, is determined the participle of mistake and is corrected.
Further, in some embodiments, object database includes product name in e-commerce platform, in user's order
Product name and user service list at least one of product name.Object Identification Module 304 is used for
In the case where text, key participle is matched with product name in object database, the corresponding commodity of the crucial participle of identification,
At least one of order, service list
In some embodiments, object database includes goods number in e-commerce platform, the single number of service or order
Number；Object Identification Module 304 is used in the case where key participle is number, is segmented and object according to coding rule by crucial
Goods number, the single number of service or O/No. are matched in database, determine that key segments corresponding commodity, order, clothes
At least one of business list.
Information sending module 306, for the description information of object to be supplied to user, so that user confirms pair to be seeked advice from
As for object.
In some embodiments, the object that information sending module 306 is used to seek advice from object and user in context carries out
Matching；In the case where being matched to same object, the description information of object is supplied to user；It is not matched to same object
In the case of, requry the users whether the object to be seeked advice from is object, in the case where user confirms that the object to be seeked advice from is object,
The description information of object is supplied to user.
In some embodiments, information sending module 306 is used in the case where object is multiple, by each object to arrange
The form of table is supplied to user；The object to be seeked advice from that user selects from list is obtained, the description letter of the object of consulting is intended to
Breath is supplied to user.
In some embodiments, the description information of object includes the title of object, picture, corresponding goods number, correspondence
O/No., corresponding service singly number at least one of information.
Customer service device in embodiment of the disclosure can realize respectively by various calculating equipment or computer system, under
Face combines Fig. 4 and Fig. 5 to be described.
Fig. 4 is the structure chart of some embodiments of disclosure customer service device.As shown in figure 4, the device of the embodiment
40 include: memory 410 and the processor 420 for being coupled to the memory 410, and processor 420 is configured as being based on being stored in
Instruction in memory 410 executes the customer service method in the disclosure in any some embodiments.
Wherein, memory 410 is such as may include system storage, fixed non-volatile memory medium.System storage
Device is for example stored with operating system, application program, Boot loader (Boot Loader), database and other programs etc..
Fig. 5 is the structure chart of other embodiments of disclosure customer service device.As shown in Fig. 2, the dress of the embodiment
Setting 50 includes: memory 510 and processor 520, similar with memory 410 and processor 420 respectively.It can also include defeated
Enter output interface 530, network interface 540, memory interface 550 etc..These interfaces 530,540,550 and memory 510 and place
It can for example be connected by bus 560 between reason device 520.Wherein, input/output interface 530 is display, mouse, keyboard, touching
It touches the input-output equipment such as screen and connecting interface is provided.Network interface 540 provides connecting interface for various networked devices, such as can be with
It is connected to database server or cloud storage server etc..Memory interface 550 is that the external storages such as SD card, USB flash disk mention
For connecting interface.
The disclosure also provides a kind of computer readable storage medium, is stored thereon with computer program, wherein the program quilt
Processor realizes the customer service method of aforementioned any embodiment when executing.
Those skilled in the art should be understood that embodiment of the disclosure can provide as method, system or computer journey
Sequence product.Therefore, complete hardware embodiment, complete software embodiment or combining software and hardware aspects can be used in the disclosure
The form of embodiment.Moreover, it wherein includes the calculating of computer usable program code that the disclosure, which can be used in one or more,
Machine can use the meter implemented in non-transient storage medium (including but not limited to magnetic disk storage, CD-ROM, optical memory etc.)
The form of calculation machine program product.
The disclosure is referring to method, the process of equipment (system) and computer program product according to the embodiment of the present disclosure
Figure and/or block diagram describe.It is interpreted as to be realized by computer program instructions each in flowchart and/or the block diagram
The combination of process and/or box in process and/or box and flowchart and/or the block diagram.It can provide these computer journeys
Sequence instruct to general purpose computer, special purpose computer, Embedded Processor or other programmable data processing devices processor with
A machine is generated, so that the instruction generation executed by computer or the processor of other programmable data processing devices is used for
Realize the dress for the function of specifying in one or more flows of the flowchart and/or one or more blocks of the block diagram
It sets.
These computer program instructions, which may also be stored in, is able to guide computer or other programmable data processing devices with spy
Determine in the computer-readable memory that mode works, so that it includes referring to that instruction stored in the computer readable memory, which generates,
Enable the manufacture of device, the command device realize in one box of one or more flows of the flowchart and/or block diagram or
The function of being specified in multiple boxes.
These computer program instructions also can be loaded onto a computer or other programmable data processing device, so that counting
Series of operation steps are executed on calculation machine or other programmable devices to generate computer implemented processing, thus in computer or
The instruction executed on other programmable devices is provided for realizing in one or more flows of the flowchart and/or block diagram one
The step of function of being specified in a box or multiple boxes.
The foregoing is merely the preferred embodiments of the disclosure, not to limit the disclosure, all spirit in the disclosure and
Within principle, any modification, equivalent replacement, improvement and so on be should be included within the protection scope of the disclosure.
Claims (16)
1. a kind of customer service method, comprising:
The character in picture that identification user sends in customer service processes；
The character is matched with object database, identifies that the corresponding object of the character, the object database include
Object information in e-commerce platform；
The description information of the object is supplied to the user, so that the user confirms that the object to be seeked advice from is described right
As.
2. customer service method according to claim 1, wherein
It is described to match the character with object database, identify that the corresponding object of the character includes:
The character is segmented, stop words is filtered and corrects error character, obtains crucial participle；
The crucial participle is matched with the object in the object database, identifies that the crucial participle is corresponding right
As.
3. customer service method according to claim 2, wherein
The object database includes product name in e-commerce platform, product name and the use in user's order
Family services at least one of the product name in list；
It is described to match the crucial participle with the object in the object database, identify that the crucial participle is corresponding
Object includes:
In the case where the crucial participle is text, the crucial participle is carried out with product name in the object database
It matches, identifies at least one of the corresponding commodity of the crucial participle, order, service list；
Alternatively,
The object database includes goods number in e-commerce platform, the single number of service or O/No.；
It is described to match the crucial participle with the object in the object database, identify that the crucial participle is corresponding
Object includes:
It, will be in the crucial participle and the object database according to coding rule in the case where the key participle is number
Goods number, the single number of service or O/No. are matched, and determine that the corresponding commodity of the crucial participle, order, service are single
At least one of.
4. customer service method according to claim 2, wherein
The correction error character includes:
By the error correcting model trained in advance of the participle input after filtering stop words, the correct probability of each participle is obtained, it will be correct
Probability is segmented lower than the participle of threshold value as mistake, is replaced the mistake using word similar in shape and is segmented, re-enters error correction
Model and the correct probability for determining each participle, until the participle in the case of choosing correct probability highest is as crucial participle；
Alternatively, the participle after filtering stop words is matched with erroneous words dictionary, determine that the participle of mistake is simultaneously corrected.
5. customer service method according to claim 1, wherein
The description information by the object is supplied to user and includes:
The object is seeked advice from the object in context with the user to match；
In the case where being matched to same object, the description information of the object is supplied to user；
In the case where not being matched to same object, inquire whether the object to be seeked advice from is the object to the user, in institute
In the case where the user's object to be seeked advice from of confirmation is stated as the object, the description information of the object is supplied to the user.
6. customer service method according to claim 1, wherein
The description information by the object is supplied to user and includes:
In the case where the object is multiple, each object is supplied to the user in the form of a list；
The object to be seeked advice from that the user selects from list is obtained, the description information of the object to be seeked advice from is supplied to
User.
7. any one of -6 customer service method according to claim 1, wherein
Text in the picture is identified according to optical character recognition algorithms；
Alternatively, the description information of the object includes the title of the object, picture, corresponding goods number, corresponding order
At least one of number, corresponding service list number information.
8. a kind of customer service device, comprising:
Character recognition module, the character in picture that user sends in customer service processes for identification；
Object Identification Module identifies the corresponding object of the character, institute for matching the character with object database
Stating object database includes the object information in e-commerce platform；
Information sending module, for the description information of the object to be supplied to the user, so that user confirmation is intended to consult
The object of inquiry is the object.
9. customer service device according to claim 8, wherein
The Object Identification Module obtains key for segmenting the character, filtering stop words and correcting error character
Participle；The crucial participle is matched with the object in the object database, identifies that the crucial participle is corresponding right
As.
10. customer service device according to claim 9, wherein
The object database includes product name in e-commerce platform, product name and the use in user's order
Family services at least one of the product name in list；The Object Identification Module is used for the case where the crucial participle is text
Under, the crucial participle is matched with product name in the object database, identifies the corresponding quotient of the crucial participle
At least one of product, order, service list；
Alternatively, the object database includes goods number in e-commerce platform, the single number of service or O/No.；It is described right
As identification module is used in the case where the crucial participle is number, according to coding rule by it is described it is crucial segment with it is described right
Goods number, the single number of service or O/No. are matched in image data library, are determined the corresponding commodity of the crucial participle, are ordered
At least one of single, service list.
11. customer service device according to claim 9, wherein
The Object Identification Module is used to filter the error correcting model for segmenting input and training in advance after stop words, obtains each point
The correct probability of word, the participle using correct probability lower than threshold value are segmented as mistake, replace the mistake using word similar in shape
It accidentally segments, re-enter error correcting model and determines the correct probability of each participle, until in the case of choosing correct probability highest
Participle is as crucial participle；
Alternatively, the Object Identification Module is used to match the participle after filtering stop words with erroneous words dictionary, determine wrong
Participle accidentally is simultaneously corrected.
12. customer service device according to claim 8, wherein
The information sending module is used to the object seeking advice from the object in context with the user and match；It is matching
In the case where same object, the description information of the object is supplied to user；In the case where not being matched to same object,
Inquire whether the object to be seeked advice from is the object to the user, confirms that the object to be seeked advice from is the object in the user
In the case where, the description information of the object is supplied to the user.
13. customer service device according to claim 8, wherein
The information sending module is used to be supplied to each object in the form of a list in the case where the object is multiple
The user；The object to be seeked advice from that the user selects from list is obtained, by the description information of the object to be seeked advice from
It is supplied to user.
14. according to the described in any item customer service devices of claim 8-13, wherein
Text in the picture is identified according to optical character recognition algorithms；
Alternatively, the description information of the object includes the title of the object, picture, corresponding goods number, corresponding order
At least one of number, corresponding service list number information.
15. a kind of customer service device, comprising:
Memory；And
It is coupled to the processor of the memory, the processor is configured to based on the finger being stored in the memory devices
It enables, executes such as the described in any item customer service methods of claim 1-7.
16. a kind of computer readable storage medium, is stored thereon with computer program, wherein when the program is executed by processor
The step of realizing any one of claim 1-7 the method.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810824060.3A CN108984777B (en) | 2018-07-25 | 2018-07-25 | Customer service method, apparatus and computer-readable storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810824060.3A CN108984777B (en) | 2018-07-25 | 2018-07-25 | Customer service method, apparatus and computer-readable storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108984777A true CN108984777A (en) | 2018-12-11 |
| CN108984777B CN108984777B (en) | 2021-04-30 |
Family
ID=64550992
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810824060.3A Active CN108984777B (en) | 2018-07-25 | 2018-07-25 | Customer service method, apparatus and computer-readable storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108984777B (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110543812A (en) * | 2019-07-19 | 2019-12-06 | 拉扎斯网络科技（上海）有限公司 | information extraction method and device, electronic equipment and storage medium |
| CN111325652A (en) * | 2018-12-17 | 2020-06-23 | 北京京东尚科信息技术有限公司 | Information embedding method, information extracting method, information embedding device, information extracting device, information embedding equipment and computer readable storage medium |
| CN111985945A (en) * | 2019-05-21 | 2020-11-24 | 阿里巴巴集团控股有限公司 | Order determining method and device |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103514547A (en) * | 2013-03-29 | 2014-01-15 | 洛阳中创网络技术服务有限公司 | Online customer service communication method and system for enterprise website |
| CN107609056A (en) * | 2017-08-25 | 2018-01-19 | 百度在线网络技术（北京）有限公司 | A kind of question and answer processing method and equipment based on picture recognition |
| CN107657230A (en) * | 2017-09-27 | 2018-02-02 | 安徽硕威智能科技有限公司 | A kind of bank self-help robot character recognition device |
| CN107807960A (en) * | 2017-09-30 | 2018-03-16 | 平安科技（深圳）有限公司 | Intelligent customer service method, electronic installation and computer-readable recording medium |
| US10275522B1 (en) * | 2015-06-11 | 2019-04-30 | State Farm Mutual Automobile Insurance Company | Speech recognition for providing assistance during customer interaction |
-
2018
- 2018-07-25 CN CN201810824060.3A patent/CN108984777B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103514547A (en) * | 2013-03-29 | 2014-01-15 | 洛阳中创网络技术服务有限公司 | Online customer service communication method and system for enterprise website |
| US10275522B1 (en) * | 2015-06-11 | 2019-04-30 | State Farm Mutual Automobile Insurance Company | Speech recognition for providing assistance during customer interaction |
| CN107609056A (en) * | 2017-08-25 | 2018-01-19 | 百度在线网络技术（北京）有限公司 | A kind of question and answer processing method and equipment based on picture recognition |
| CN107657230A (en) * | 2017-09-27 | 2018-02-02 | 安徽硕威智能科技有限公司 | A kind of bank self-help robot character recognition device |
| CN107807960A (en) * | 2017-09-30 | 2018-03-16 | 平安科技（深圳）有限公司 | Intelligent customer service method, electronic installation and computer-readable recording medium |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111325652A (en) * | 2018-12-17 | 2020-06-23 | 北京京东尚科信息技术有限公司 | Information embedding method, information extracting method, information embedding device, information extracting device, information embedding equipment and computer readable storage medium |
| CN111985945A (en) * | 2019-05-21 | 2020-11-24 | 阿里巴巴集团控股有限公司 | Order determining method and device |
| CN110543812A (en) * | 2019-07-19 | 2019-12-06 | 拉扎斯网络科技（上海）有限公司 | information extraction method and device, electronic equipment and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108984777B (en) | 2021-04-30 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| AU2019261735B2 (en) | System and method for recommending automation solutions for technology infrastructure issues | |
| CN110222272B (en) | Potential customer mining and recommending method | |
| CN107330445B (en) | User attribute prediction method and device | |
| CN104951428B (en) | User's intension recognizing method and device | |
| CN108536852A (en) | Question and answer exchange method and device, computer equipment and computer readable storage medium | |
| US20220172258A1 (en) | Artificial intelligence-based product design | |
| CN111680165B (en) | Information matching method and device, readable storage medium and electronic equipment | |
| CN108984777A (en) | Customer service method, apparatus and computer readable storage medium | |
| WO2020034880A1 (en) | Logistics object information processing method, device and computer system | |
| CN111754278A (en) | Article recommendation method and device, computer storage medium and electronic equipment | |
| CN110516033A (en) | A kind of method and apparatus calculating user preference | |
| CN113781149A (en) | Information recommendation method and device, computer-readable storage medium and electronic equipment | |
| CN115293332A (en) | Method, device and equipment for training graph neural network and storage medium | |
| CN113590678A (en) | Portrait analysis method based on internet finance and big data analysis server | |
| CN111177581A (en) | Multi-platform-based social e-commerce website commodity recommendation method and device | |
| EP4283496A1 (en) | Techniques for automatic filling of an input form to generate a listing | |
| Lettner et al. | Mobile interaction analysis: towards a novel concept for interaction sequence mining | |
| CN116501979A (en) | Information recommendation method, information recommendation device, computer equipment and computer readable storage medium | |
| CN115860872A (en) | Target object determination method and device, electronic equipment and storage medium | |
| CN112328899B (en) | Information processing method, information processing apparatus, storage medium, and electronic device | |
| CN111680499B (en) | Method and system for analyzing article application based on natural language processing | |
| CN113204333A (en) | Software interface design draft front-end element identification method | |
| CN113763084A (en) | Product recommendation processing method, device, equipment and storage medium | |
| CN112445910B (en) | Information classification method and system | |
| CN114841747A (en) | Attribute analysis system, training method, attribute analysis method, and attribute analysis device |
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 | ||
| TR01 | Transfer of patent right | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20210527 Address after: 100176 room 1004, 10th floor, building 1, 18 Kechuang 11th Street, Beijing Economic and Technological Development Zone, Daxing District, Beijing Patentee after: Beijing Huijun Technology Co.,Ltd. Address before: 100195 Beijing Haidian Xingshikou Road 65 West Cedar Creative Garden 4 District 11 Building East 1-4 Floor West 1-4 Floor Patentee before: BEIJING JINGDONG SHANGKE INFORMATION TECHNOLOGY Co.,Ltd. Patentee before: BEIJING JINGDONG CENTURY TRADING Co.,Ltd. |