CN116595966A - User complaint processing method, device, equipment and storage medium - Google Patents
User complaint processing method, device, equipment and storage medium Download PDFInfo
- Publication number
- CN116595966A CN116595966A CN202310597190.9A CN202310597190A CN116595966A CN 116595966 A CN116595966 A CN 116595966A CN 202310597190 A CN202310597190 A CN 202310597190A CN 116595966 A CN116595966 A CN 116595966A
- Authority
- CN
- China
- Prior art keywords
- complaint
- user
- work order
- standard
- keyword
- 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.)
- Pending
Links
- 238000003672 processing method Methods 0.000 title abstract description 10
- 238000000034 method Methods 0.000 claims abstract description 56
- 238000012545 processing Methods 0.000 claims abstract description 50
- 230000008451 emotion Effects 0.000 claims description 23
- 238000013145 classification model Methods 0.000 claims description 16
- 230000015654 memory Effects 0.000 claims description 16
- 230000008569 process Effects 0.000 claims description 9
- 238000012790 confirmation Methods 0.000 claims description 4
- 238000006243 chemical reaction Methods 0.000 claims description 3
- 238000012163 sequencing technique Methods 0.000 claims description 3
- 230000006854 communication Effects 0.000 description 9
- 238000004891 communication Methods 0.000 description 8
- 238000013527 convolutional neural network Methods 0.000 description 5
- 238000010586 diagram Methods 0.000 description 4
- 230000006870 function Effects 0.000 description 4
- 239000013598 vector Substances 0.000 description 4
- 238000013528 artificial neural network Methods 0.000 description 3
- 238000004590 computer program Methods 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 239000012634 fragment Substances 0.000 description 3
- 125000004122 cyclic group Chemical group 0.000 description 2
- 239000000284 extract Substances 0.000 description 2
- 238000000605 extraction Methods 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 230000005856 abnormality Effects 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004140 cleaning Methods 0.000 description 1
- 238000012937 correction Methods 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000005281 excited state Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 230000036651 mood Effects 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 230000000306 recurrent effect Effects 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 230000006403 short-term memory Effects 0.000 description 1
- 238000010183 spectrum analysis Methods 0.000 description 1
- 238000007619 statistical method Methods 0.000 description 1
- 230000002123 temporal effect Effects 0.000 description 1
- 238000012549 training Methods 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
- 238000013519 translation Methods 0.000 description 1
- 230000001960 triggered effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/10—Text processing
- G06F40/166—Editing, e.g. inserting or deleting
- G06F40/186—Templates
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/33—Querying
- G06F16/3331—Query processing
- G06F16/334—Query execution
- G06F16/3343—Query execution using phonetics
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/33—Querying
- G06F16/3331—Query processing
- G06F16/334—Query execution
- G06F16/3344—Query execution using natural language analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/35—Clustering; Classification
- G06F16/353—Clustering; Classification into predefined classes
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/01—Customer relationship services
- G06Q30/015—Providing customer assistance, e.g. assisting a customer within a business location or via helpdesk
Abstract
The application discloses a user complaint processing method, a device, equipment and a storage medium, belonging to the technical field of data processing, wherein the method comprises the following steps: manually filling in a complaint work order and user complaint voice of the structured complaint template and the target customer service account number are obtained; wherein, the manual filling complaint worksheets correspond to the complaint voices of the users; converting the user complaint voice into user complaint text information; and filling in the structural complaint template based on the user complaint text information and the manually filled-in complaint worksheets to obtain standard complaint worksheets. According to the application, the structural complaint template is filled by using the user complaint text information and manually filling the complaint work order, so that the situation that key information is omitted or random errors exist when customer service personnel manually record the complaint content can be avoided, and the generated complaint work order is matched with the actual complaint requirement of the user.
Description
Technical Field
The present application relates to the field of data processing technologies, and in particular, to a method, an apparatus, a device, and a storage medium for user complaint processing.
Background
With the improvement of consumer rights and interests protection consciousness, the number of complaints of banking users on business is increasing. The current flow of the bank for handling the user complaints is as follows: the user dials a special line for complaint, and the work order is recorded manually by an operator; the work order and the recording and transferring text flow are transferred to a difficult team, the contents of the complaints (services, businesses, products and the like) are manually analyzed and confirmed, and then the flow is transferred to a responsibility unit for follow-up processing.
However, in the current complaint processing flow, when customer service personnel fill in a complaint work order manually, the situation that key information is missed easily occurs, or the situation that random errors exist, such as input errors of the customer service personnel, so that the generated complaint work order is not matched with the actual complaint demands of users.
Disclosure of Invention
The application mainly aims to provide a user complaint processing method, device and equipment and a storage medium, and aims to solve the technical problem that a generated complaint work order is not matched with the actual complaint demand of a user.
In order to achieve the above object, the present application provides a method for handling customer complaints, comprising:
manually filling in a complaint work order and user complaint voice of the structured complaint template and the target customer service account number are obtained; wherein, the manual filling complaint worksheets correspond to the complaint voices of the users;
Converting the user complaint voice into user complaint text information;
and filling in the structural complaint template based on the user complaint text information and the manually filled-in complaint worksheets to obtain standard complaint worksheets.
Optionally, the filling the structured complaint template based on the user complaint text information and the manually filled complaint worksheet to obtain a standard complaint worksheet includes:
extracting key information from the user complaint text information according to a preset complaint vocabulary table to obtain first key information, and extracting key information from the manually filled complaint work order to obtain second key information;
identifying a distinct keyword from the first keyword information that is different from the second keyword information, and a common keyword that is the same as the second keyword information;
if the total number of the first key words in the first key information is consistent with the total number of the second key words in the second key information, filling in the structured complaint template based on the distinguishing key words and the common key words to obtain the standard complaint work order;
if the total number of the first key words in the first key information is inconsistent with the total number of the second key words in the second key information, filling in the structured complaint template based on the distinguishing key words and the second key information, and obtaining the standard complaint work order.
Optionally, the identifying, from the first keyword, a distinct keyword different from the second keyword, and a common keyword identical to the second keyword, includes:
calculating the characteristic distance between the first key word of each item and the second key word of each item;
and identifying a distinguishing keyword which is different from the second keyword from the first keyword and a common keyword which is the same as the second keyword according to the characteristic distance.
Optionally, after the converting the user complaint speech into the user complaint text information, the method further includes:
extracting emotion keywords from the user complaint text information;
and sequencing the standard complaint worksheets based on the emotion keywords to obtain the arrangement sequence of the standard complaint worksheets, so that the staff can process the standard complaint worksheets based on the arrangement sequence.
Optionally, after the structured complaint template is filled in based on the user complaint text information and the manually filled-in complaint worksheet, the method further includes:
Judging whether the standard complaint work orders are complaint work orders or not by using a complaint work order judging model, wherein the complaint work order judging model is an NLP language model;
if the standard complaint work order is a complaint work order, inputting the standard complaint work order into a complaint work order classification model for classification, and obtaining a classification result output by the complaint work order classification model.
Optionally, after the determining whether the standard complaint work order is a complaint work order, the method further includes:
if the standard complaint work order is not the complaint work order, outputting the standard complaint work order to a person for confirming the complaint work order;
and if the manually input complaint work order confirmation information is received, executing the step of inputting the standard complaint work order into the complaint work order classification model to classify, and obtaining a classification result output by the complaint work order classification model.
Optionally, after the structured complaint template is filled in based on the user complaint text information and the manually filled-in complaint worksheet, the method further includes:
and generating a complaint system report based on the type of the standard complaint work order, the processing time of the standard complaint work order and the complaint index of the complaint standard work order.
In a second aspect, the present application provides a user complaint handling device comprising:
the acquisition module is used for acquiring a structured complaint template, a manual filling complaint work order of a target customer service account and a user complaint voice; wherein, the manual filling complaint worksheets correspond to the complaint voices of the users;
the conversion module is used for converting the user complaint voice into user complaint text information;
and the filling module is used for filling the structured complaint template based on the user complaint text information and the manual filling complaint work order to obtain a standard complaint work order.
In a third aspect, the present application provides a user complaint handling device comprising a memory, a processor and a user complaint handling program stored on the processor and executable on the processor, the user complaint handling program being configured to implement the steps of the user complaint handling method as described above.
In a fourth aspect, the present application provides a computer-readable storage medium having stored thereon a computer program which, when processed, performs the user complaint handling method of any embodiment of the present application.
Compared with the prior art, key information in a user complaint telephone is manually input by customer service personnel, the method provided by the embodiment of the application has the advantages that after the user complaint voice is converted into the user complaint text information, the structured complaint template is filled in by combining the user complaint text information and the manual filling complaint work order together, so that the standard complaint work order is obtained, namely, the information which is missed by customer service in the manual filling complaint work order is perfected by using the user complaint text information, the condition that the key information is missed when the customer service personnel records the complaint content manually is avoided, or random errors exist, and the like, so that the matching degree of the generated complaint work order and the actual complaint requirement of the user is higher.
Drawings
FIG. 1 is a schematic diagram of a hardware configuration of a customer complaint handling device according to an embodiment of the present application;
FIG. 2 is a flowchart illustrating a first embodiment of a method for handling customer complaints according to the present application;
FIG. 3 is a flowchart illustrating a second embodiment of a method for handling customer complaints according to the present application;
FIG. 4 is a flowchart illustrating a third embodiment of a method for handling customer complaints according to the present application;
FIG. 5 is a flowchart illustrating a fourth embodiment of a method for handling customer complaints according to the present application;
FIG. 6 is a functional block diagram of a customer complaint handling and recognition device according to the present application.
The achievement of the objects, functional features and advantages of the present application will be further described with reference to the accompanying drawings, in conjunction with the embodiments.
Detailed Description
It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the application.
Because of the prior art, in the current complaint processing flow, when customer service personnel fill in a complaint work order manually, the situation of key information omission is easy to occur, or the situation of random errors exists, such as the input errors of the customer service personnel, so that the generated complaint work order is not matched with the actual complaint demands of users.
Compared with the prior art, key information in a user complaint telephone is manually input by customer service personnel, the method and the system have the advantages that after the user complaint voice is converted into the user complaint text information, the structured complaint template is filled in by combining the user complaint text information and the manual filling complaint work order together, so that a standard complaint work order is obtained, namely, the customer service missing information in the manual filling complaint work order is perfected by utilizing the user complaint text information, the situation that the key information is missing when the customer service personnel records the complaint content manually is avoided, or random errors exist, and the like, and the generated complaint work order and the actual complaint requirement of the user are higher in matching degree.
Referring to fig. 1, fig. 1 is a schematic structural diagram of a user complaint handling device in a hardware running environment according to an embodiment of the present application.
As shown in fig. 1, the user complaint handling device may include: a processor 1001, such as a central processing unit (Central Processing Unit, CPU), a communication bus 1002, a user interface 1003, a network interface 1004, a memory 1005. Wherein the communication bus 1002 is used to enable connected communication between these components. The user interface 1003 may include a Display, an input unit such as a Keyboard (Keyboard), and the optional user interface 1003 may further include a standard wired interface, a wireless interface. The network interface 1004 may optionally include a standard wired interface, a WIreless interface (e.g., a WIreless-FIdelity (WI-FI) interface). The Memory 1005 may be a high-speed random access Memory (Random Access Memory, RAM) Memory or a stable nonvolatile Memory (NVM), such as a disk Memory. The memory 1005 may also optionally be a storage device separate from the processor 1001 described above.
Those skilled in the art will appreciate that the structure shown in FIG. 1 is not limiting of the user complaint handling device and may include more or fewer components than shown, or certain components in combination, or a different arrangement of components.
As shown in fig. 1, an operating system, a data storage module, a network communication module, a user interface module, and a user complaint handling program may be included in the memory 1005 as one type of storage medium.
In the customer complaint handling device shown in FIG. 1, network interface 1004 is primarily used for data communication with a network server; the user interface 1003 is mainly used for data interaction with a user; the processor 1001 and the memory 1005 in the user complaint processing device of the present application may be disposed in the user complaint processing device, and the user complaint processing device invokes the user complaint processing program stored in the memory 1005 through the processor 1001, and executes the user complaint processing method provided by the embodiment of the present application.
Based on the hardware structure of the user complaint processing device but not limited to the hardware structure, the present application provides a first embodiment of a user complaint processing method. Referring to fig. 2, fig. 2 shows a flow chart of a first embodiment of a method for applying for customer complaint treatment.
It should be noted that although a logical order is depicted in the flowchart, in some cases the steps depicted or described may be performed in a different order than presented herein.
In this embodiment, the user complaint processing method includes:
and S10, manually filling in a complaint work order and user complaint voice of the structured complaint template and the target customer service account.
Wherein, the manual filling complaint worksheets correspond to the complaint voices of the users.
The user complaint processing method is realized when the user complaint processing device executes the user complaint processing program.
It can be appreciated that when a user uses a service provided by a service provider, it is inevitable that the user feels unpleasant, resulting in user complaints. In general, a user may establish a call connection with a customer service person of a service provider by dialing a complaint hotline, and then the user states what happens to the customer service person (i.e., the problem that occurs), and what the user needs the service provider to solve and/or make compensation (i.e., the actual complaint needs of the user). In the communication process, customer service personnel can remind the user whether to agree to record the complaint content. After the user agrees, the call content can be recorded, so that the complaint voice of the user is obtained. After the call is ended, the complaint voice of the user can be stored in the local or cloud.
And then, customer service personnel can log in a customer service account number of the customer service personnel on an internal system of a service provider, namely a target customer service account number, call a corresponding complaint work order template, fill in and submit the complaint work order template, and thus, a manual filling-in complaint work order is obtained.
The structured complaint template is a reusable document which can be preconfigured by a practitioner, so that the time for filling the complaint document information of the user and manually filling the complaint work order is saved, and the format of the standard complaint work order is unified.
The structured complaint template can be a complaint work order template composed of five fields of user information, complaint text information, department processing opinion text information, processing result text information and return visit result text information. Of course, the structured complaint template may include more fields, such as complaint type information, complaint form information, etc., for example, the complaint type information may be service class or business class, and the complaint form may be phone complaint or on-the-surface complaint.
In this embodiment, the structured complaint template may be a local or cloud pre-stored in the user complaint processing device, and when step S10 is executed, the structured complaint template and the user complaint voice are triggered to be directly called from the local or cloud by the user complaint processing device when the user complaint processing device receives the manual filling complaint work order submitted by the target customer service account.
Or, the method can also be that when the user complaint processing equipment is timed or meets other triggering conditions of configuration by a manager, a complaint work order is manually filled in by calling the target customer service account, and a structured complaint template and user complaint voice are called from a local or cloud.
And step S20, converting the complaint voice of the user into the complaint text information of the user.
In this embodiment, ASR (Automatic Speech Recognition ) techniques may be employed to convert the user complaint speech into user complaint text information.
Specifically, when the ASR technology is adopted to convert the complaint voice of the user into the complaint Text information of the user, an STT (Speech-to-Text) algorithm is mainly adopted. Specifically, a CNN (Convolutional Neural Networks, convolutional neural network) model and an RNN (Recurrent Neural Network, cyclic neural network) model may be used to construct a speech-to-text network for dividing the structure of speech words by using CTCs (Connectionist Temporal Classification, time-series class classification based on neural network) as a loss function. The CNN model is used for processing the voice input spectrogram and outputting a voice characteristic graph. The RNN model is a cyclic network of two-way LSTM (long short-term memory network) that is used to further process the speech feature map into character sequences of different time or frames. The CTC penalty function is used to calculate the character probabilities and derive the correct character sequence. And finally, generating the probability of generating characters by each step length by using a Softmax normalization layer, aligning the probability with a character sequence, and outputting a final translation result of the voice-to-text network to obtain the user complaint text information.
And step S30, filling in the structural complaint template based on the user complaint text information and the manual filling-in complaint work orders to obtain the standard complaint work orders.
After the user complaint text information is obtained and the complaint work order is manually filled, original information recorded in the user complaint text information and recorded customer service processed information in the manually filled complaint work order can be combined together to fill in blank contents in the structured complaint template, so that a standard complaint work order which is not missed in complaint information and can completely reflect actual complaint demands of users is obtained.
Further, as an optional embodiment, step S30 specifically includes:
step S301, extracting key information from user complaint text information according to a preset complaint vocabulary table to obtain first key information, and extracting key information from a manually filled complaint work order to obtain second key information.
In this embodiment, the preset complaint vocabulary includes a plurality of preset complaint vocabularies, and the preset complaint vocabularies can reflect the problems encountered by the user or the complaint information such as compensation that the user wants to obtain. The preset complaint vocabulary can be configured by a practitioner in advance according to own experience, for example, after a customer service practitioner establishes call connection with a customer, a plurality of key words are determined through the condition of customer voice statement, for example, key words such as 'money is not checked out in real time', 'bank account abnormality' and the like are determined. Alternatively, the preset complaint vocabulary may be generated manually or after the user complaint device extracts the key information from the historical user complaint text information.
It will be appreciated that after the user complaint speech is converted into the user complaint text information, the user complaint text information may include useless information such as complaints or spitting grooves of the user in addition to the first key information related to the user complaint information. Therefore, keyword extraction can be performed on the user complaint text information according to the preset complaint vocabulary, so that the first keyword information can be determined from the user complaint text information quickly and accurately.
It should be noted that, there may be a plurality of first keywords in the first keyword extracted from the user complaint text information, but the relationship between the extracted first keywords and the complaint of the user may not be large, so, in order to avoid the influence of other keywords on the first keyword, the correlation between the first keyword and the user complaint text is improved, and further, as an alternative implementation manner, the extraction of the keywords from the user complaint text information may be performed according to a TF-IDF (Term Frequency-inverse Document Frequency) algorithm and a preset complaint vocabulary.
In this embodiment, the customer service staff can process the complaint voice of the user according to the preset complaint vocabulary, that is, the customer service staff can only fill in keywords when filling in the complaint work order template, so that the customer service staff can directly extract the keywords when manually filling in the complaint work order for extracting the key information. Or the step of extracting key information of the user complaint text information may be referred to, and this embodiment will not be described herein.
Step S302, distinguishing key words different from the second key information and common key words identical to the second key information are identified from the first key information.
In this embodiment, the distinguishing keyword and the common keyword may be identified by comparing each first keyword in the first keyword with each second keyword in the second keyword one by one.
Of course, because customer service naturally processes the complaint voice of the user in the process of filling in the manual complaint work order, the first keyword and the second keyword reflecting the same complaint content may not be completely consistent. In order to avoid comparing such first keyword and second keyword to different keywords during the comparison, further, as an optional embodiment, step S302 specifically includes:
and step A10, calculating the characteristic distance between each first keyword and each second keyword.
And step A20, identifying different key words from the first key information and common key words which are the same as the second key information from the first key information according to the feature distance.
In this embodiment, the feature distance may be used to represent the similarity between each item of the first keyword and the second keyword.
Specifically, each first keyword and each second keyword can be input into a DSSM (Deep Structured Semantic Models, deep semantic matching model) model, the feature distance between each first keyword and each second keyword is calculated through the DSSM model, and the first keyword and the second keyword with the feature distance smaller than a first preset value are regarded as the first keyword and the second keyword reflecting the same complaint content, namely the same common keyword. Of course, it is also possible to identify a first keyword whose feature distances from a plurality of second keywords are all greater than a second preset value, that is, a second keyword that does not exist in the first keyword and reflects the same complaint content, thereby obtaining a differential keyword.
Alternatively, other manners may be adopted to calculate the feature distance between each first keyword and each second keyword, which is not limited in this embodiment.
Step S303, if the total number of the first key words in the first key information is consistent with the total number of the second key words in the second key information, filling in the structured complaint template based on the distinguishing key words and the shared key words to obtain a standard complaint work order.
Step S304, if the total number of the first key words in the first key information is inconsistent with the total number of the second key words in the second key information, filling in the structured complaint template based on the distinguishing key words and the second key information to obtain a standard complaint work order.
It should be noted that, the total number of terms of the first keyword in the first keyword information may be identical to or different from the total number of terms of the second keyword in the second keyword information.
Specifically, when the total number of the first keyword in the first keyword is consistent with the total number of the second keyword in the second keyword, it can be determined that no condition of missing the keyword exists when the customer service staff fills out the complaint work order manually, but in order to avoid the situation that the customer service staff inputs errors and the complaint keyword errors occur, the structural complaint template can be filled out according to the distinguishing keyword and the common keyword, and the standard complaint work order is obtained.
When the total number of the first key words in the first key information is inconsistent with the total number of the second key words in the second key information, the condition that the customer service personnel miss the key information when filling the complaint work order manually can be determined. In order to avoid missing complaint key information, the structural complaint templates are filled in according to the distinguishing key words and the second key information, so that complaint requirements of users can be covered in an overall mode.
Further, as an alternative implementation manner, after obtaining the text information of the complaints of the user, a FastText algorithm may be used in the NLP algorithm model to classify the text of the complaints of the user, so as to obtain a text classification tag. Specifically, the received word vector is transmitted to the CNN network structure, and the word vector label, namely the text classification label, is output. It should be noted that, because the NLP algorithm model cannot directly process the unstructured text, the unstructured text needs to be converted into the structured text, that is, the first key information is extracted from the user complaint text information by using the preset complaint vocabulary, a dedicated keyword lexicon is generated according to the first key information, word vectors are generated by combining the keyword lexicon, and the word vectors are input into the NLP algorithm model to obtain the text classification tag.
In this embodiment, after obtaining the text classification label according to the text information of the user complaints, the standard complaint worksheets can be distributed to the corresponding processing departments according to the text classification label, so that the problem of the user can be solved more quickly and accurately, and the user experience is improved.
Further, as an optional implementation manner, on the basis of extracting first key information from user complaint text information by using a preset complaint vocabulary to generate a structured text, data cleaning and text correction are further required for the first key information, and finally complaint information of text classification results, work order association attributes and other channels is summarized to an information input acquisition module to obtain structured complaint data with uniform formatting, namely a standard complaint work order, and the standard complaint work order is distributed to a corresponding processing department according to the type of the standard complaint work order.
In the embodiment, after the user complaint voice is converted into the user complaint text information, the structural complaint template is filled according to the user complaint text information and the manual filling complaint work order, so that the standard complaint work order is obtained, namely, the information of customer service omission in the manual filling complaint work order is perfected by using the user complaint text information, the situation that key information is omitted when customer service personnel manually record complaint contents is avoided, or random errors exist, and the like, so that the generated complaint work order is matched with the actual complaint demands of the user. Meanwhile, the user complaint text information is classified, and the standard complaint work orders are distributed to corresponding processing departments according to the classification result, so that the problem of users can be solved more quickly and accurately, and the user experience is improved.
Further, a second embodiment of the user complaint handling method of the present application is presented.
Referring to fig. 3, in the present embodiment, after step S30, further includes:
and S31, extracting emotion keywords from the user complaint text information.
And S32, sequencing the standard complaint worksheets based on the emotion keywords to obtain the sequence of the standard complaint worksheets, so that the staff can process the standard complaint worksheets based on the sequence.
It will be appreciated that the user may determine the user's current mood by the user's word when complaining. And extracting words representing emotion, such as 'I'm feel very happy ',' I'm disappointed', and the like, from the user complaint text information. Or, as an alternative implementation manner, to determine the emotion degrees of the user at the time, such as gas generation and very gas generation, words representing degrees in the text information of the complaint of the user can be extracted, for example, when the text of the complaint of the user is "i very gas generation", the extracted emotion keywords can be "very" and "gas generation" so as to represent the gas generation degree of the user.
Specifically, after the emotion keywords are extracted from the user complaint text, the standard complaint work orders can be ordered according to the emotion keywords, so that the order of the standard complaint work orders is obtained, namely, the urgency of the standard complaint work orders is determined according to the emotion keywords, for example, when the emotion keywords are 'angry', 'very angry', the standard complaint work orders corresponding to the emotion keywords 'very angry' are arranged at the first position, the standard complaint work orders corresponding to the emotion keywords 'angry' are arranged at the second position, the order of processing the standard complaint work orders is reasonably formulated according to the urgency degree of the standard complaint work orders, namely, the standard complaint work orders with extremely urgent creep conditions are processed in priority, the slow emergency conditions are general standard complaint work orders, and the standard complaint work orders are processed according to the order of the standard complaint work orders.
Or, further, as an optional implementation manner, after the complaint voice of the user is obtained, emotion spectrum analysis can be performed on the voice information to obtain an emotion analysis result of the user. Specifically, after the complaint voice of the user is obtained, the complaint voice of the user is divided into a plurality of voice fragments according to a preset unit time, the syllable quantity in each voice fragment is detected, the maximum syllable quantity is determined from the syllable quantities, then the average syllable quantity is obtained according to the syllable quantity in the voice fragments and a plurality of preset unit time periods, and the emotion of the user is determined according to the difference value between the average syllable quantity and the maximum syllable quantity. Meanwhile, the volume of the user in each preset unit time is determined, the maximum volume is determined from the volumes, the average volume of the volumes is determined, and the emotion of the user is determined according to the difference value between the average volume and the maximum volume. If the user passes through the voice recognition method, the voice recognition method and the voice recognition system are fast, when the voice of the user is high, the difference value between the average syllable number and the maximum syllable number is close to zero, the difference value between the average voice and the maximum voice is close to zero, and at the moment, the emotion of the user can be determined to be in an excited state when the user complains, so that the event is important for the user and needs to be processed fast.
Further, as an optional implementation manner, after the standard complaint worksheet is obtained, the complaint background and the reason of the user can be divided by combining the user portrait and the transaction information of the user in the bank, so that the complaint requirements of the user can be more accurately determined.
In this embodiment, the emotion keywords are used to determine the emotion of the user during complaints, and the standard complaint worksheets are ordered according to the emotion keywords, and when the order of the standard complaint worksheets is forward, the standard complaint worksheets can be distributed to departments for urgent processing, so that the problem of the user can be solved quickly. Meanwhile, the intelligent strategy recommendation method and the intelligent strategy recommendation system integrate multiple party data sources such as complaint data and customer information by combining an artificial intelligent technology, conduct intelligent strategy recommendation processing on the customer complaint information, conduct 24-hour full-flow real-time supervision, quickly respond to customer complaints with high quality, and improve service quality.
Further, a third embodiment of the user complaint handling method of the present application is presented, referring to fig. 4.
In this embodiment, step S30 further includes:
and step S33, judging whether the standard complaint work order is a complaint work order by using a complaint work order judging model, wherein the complaint work order judging model is an NLP language model.
And step S34, if the standard complaint work order is a complaint work order, inputting the standard complaint work order into a complaint work order classification model for classification, and obtaining a classification result output by the complaint work order classification model.
It can be understood that part of errors exist in the standard complaint worksheets output by the complaint information input acquisition and processing flow, namely, worksheets with wrong-split complaints exist. In order to avoid the situation of wrongly dividing the complaint work orders, the complaint work order judgment model is also required to be utilized to judge whether the standard complaint work order is the complaint work order.
Specifically, when the complaint work order judgment model is an NLP language model, after inputting the standard complaint work order into the NLP language model, extracting third key information in the standard complaint work order according to a preset complaint vocabulary, verifying the third key information by using the second key information, and when all second key words in the second key words are consistent with all third key words in the third key information, namely, the similarity between all second key words and all third key words is higher, determining to input the standard complaint work order into the complaint work order classification model to classify the standard complaint work order, obtaining a classification result output by the complaint work order classification model, and distributing the standard complaint work order to a corresponding processing department according to the classification result.
In this embodiment, the standard complaint work orders are judged, so that the subsequent distribution of the non-complaint work orders to the processing department can be avoided, and the workload of staff is reduced. Meanwhile, when the standard complaint work orders are complaint work orders, the standard complaint work orders are input into a complaint work order classification model to be classified, classification results are obtained, and the standard complaint work orders are distributed to corresponding processing departments according to the classification results, so that the processing departments can be rapidly and accurately determined, the processing period of the standard complaint work orders is shortened, and particularly, before the method is adopted, the number of complaints processed annually is 80000+ pieces, the average processing time of each complaint is 5 days, and the average processing time of each standard complaint work order is 210 seconds; with this method, the number of complaints treated each year is 500000+ pieces, the average duration of each complaint treatment is 3.5 days, and 1.5 seconds are required for each standard complaint work order classification. Meanwhile, the manual intervention error can be reduced, the standard complaint work orders can be efficiently and stably analyzed and processed, and the circulation efficiency is further improved.
Further, a fourth embodiment of the user complaint handling method of the present application is presented, referring to fig. 5.
In this embodiment, after step S33, the method further includes:
And step S35, outputting to a manual work to confirm the complaint work if the standard complaint work is not the complaint work.
And step S36, if the manually input complaint work order confirmation information is received, the standard complaint work order is input into the complaint work order classification model to be classified, and a classification result output by the complaint work order classification model is obtained.
In this embodiment, in order to avoid misjudgment and to avoid the situation that user complaints are not handled timely, when the standard complaint work order is not a complaint work order, the standard complaint work order which is not a complaint work order is sent to the staff, after the staff receives the standard complaint work order, the staff can judge whether the standard complaint work order is a complaint work order according to the preset complaint vocabulary and the content recorded by the standard complaint work order, and operate corresponding controls according to the judgment result to generate complaint work order confirmation information, that is, the standard complaint work order is not a complaint work order, or the complaint work order when the standard complaint work order is the complaint work order.
It can be appreciated that when the standard complaint work order is not a complaint work order, the case setting processing is performed on the standard complaint work order which is not the complaint work order. When the standard complaint work orders are used, the standard complaint work orders can be sent to the processing departments for processing, or the standard complaint work orders can be input into the complaint work order classification model for classification, classification results are obtained, and the standard complaint work orders are distributed to the corresponding processing departments according to the classification results. And meanwhile, inputting the standard complaint worksheet into the complaint worksheet judgment model again for training, so that the complaint worksheet judgment model is further optimized, and the accuracy of the complaint worksheet judgment model is improved.
Further, a sixth embodiment of the user complaint handling method of the present application is presented.
In this embodiment, step S30 further includes:
and S37, generating a complaint system report based on the type of the standard complaint work order, the processing time of the standard complaint work order and the complaint index of the complaint standard work order.
In this embodiment, the type of the standard complaint work order may be determined according to the content of the standard complaint work order, for example, when the content of the standard complaint work order relates to the RMB deposit, the type of the standard complaint work order is the money deposit, and when the content of the standard complaint work order relates to the credit card, the type of the standard complaint work order is the credit card. According to the generation time of the standard complaint work orders and the case settling time of the standard complaint work orders, the processing time of the standard complaint work orders can be determined. The complaint index can be data such as a transaction rate, a feedback rate, satisfaction, a response rate, a transfer rate and the like.
In this embodiment, statistical analysis is performed according to parameters such as the type, processing duration, complaint index and the like of a standard complaint work order, and a complaint system report is generated, so that the complaint flow can be monitored and managed from different dimensions, and defects can be timely processed, and if the processing duration is longer, corresponding measures are taken to shorten the processing duration.
Based on the same application concept, the present application proposes a user complaint handling device, and referring to fig. 6, fig. 6 is a schematic block diagram of a first embodiment of the user complaint handling device according to the present application.
The acquisition module is used for acquiring a structured complaint template, a manual filling complaint work order of a target customer service account and a user complaint voice; manually filling in a complaint work order corresponding to the complaint voice of the user;
the conversion module is used for converting the complaint voice of the user into the complaint text information of the user;
the filling module is used for filling the structural complaint template based on the user complaint text information and the manual filling complaint work orders to obtain standard complaint work orders.
In this embodiment, reference may be made to various implementations of the user complaint processing method in the foregoing embodiments for the implementation of the user complaint processing device and the technical effects achieved by the implementation of the user complaint processing device, which are not described herein again.
According to the technical scheme, through mutual coordination among the functional modules, a complaint work order and a user complaint voice are manually filled in by acquiring a structured complaint template and a target customer service account; manually filling in a complaint work order corresponding to the complaint voice of the user; converting the complaint voice of the user into text information of the complaint of the user; and filling the structured complaint template based on the user complaint text information and the manual filling complaint work order to obtain a standard complaint work order. Compared with the prior art, key information in a user complaint telephone is manually input by a customer service staff, after the user complaint voice is converted into the user complaint text information, the structured complaint template is filled in by combining the user complaint text information and the manual filling complaint work order, so that a standard complaint work order is obtained, namely, the information missing from the customer service in the manual filling complaint work order is perfected by using the user complaint text information, the situation that the key information is missing when the customer service staff records the complaint content manually is avoided, or random errors exist, and the like, so that the matching degree of the generated complaint work order and the actual complaint requirement of the user is higher.
In addition, the embodiment of the application also provides a computer storage medium, wherein a user complaint processing program is stored on the storage medium, and the user complaint processing program is executed by a processor to realize the steps of the user complaint processing method. Therefore, a detailed description will not be given here. In addition, the description of the beneficial effects of the same method is omitted. For technical details not disclosed in the embodiments of the computer-readable storage medium according to the present application, please refer to the description of the method embodiments of the present application. As an example, the program instructions may be deployed to be executed on one computing device or on multiple computing devices at one site or distributed across multiple sites and interconnected by a communication network.
Those skilled in the art will appreciate that implementing all or part of the above-described methods may be accomplished by way of computer programs, which may be stored on a computer-readable storage medium, and which, when executed, may comprise the steps of the embodiments of the methods described above. The storage medium may be a magnetic disk, an optical disk, a Read-Only Memory (ROM), a Random access Memory (Random AccessMemory, RAM), or the like.
It should be further noted that the above-described apparatus embodiments are merely illustrative, where elements described as separate elements may or may not be physically separate, and elements shown as elements may or may not be physical elements, may be located in one place, or may be distributed over a plurality of network elements. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment. In addition, in the drawings of the embodiment of the device provided by the application, the connection relation between the modules represents that the modules have communication connection, and can be specifically implemented as one or more communication buses or signal lines. Those of ordinary skill in the art will understand and implement the present application without undue burden.
From the above description of the embodiments, it will be apparent to those skilled in the art that the present application may be implemented by means of software plus necessary general purpose hardware, or of course by means of special purpose hardware including application specific integrated circuits, special purpose CPUs, special purpose memories, special purpose components, etc. Generally, functions performed by computer programs can be easily implemented by corresponding hardware, and specific hardware structures for implementing the same functions can be varied, such as analog circuits, digital circuits, or dedicated circuits. However, a software program implementation is a preferred embodiment for many more of the cases of the present application. Based on such understanding, the technical solution of the present application may be embodied essentially or in a part contributing to the prior art in the form of a software product stored in a readable storage medium, such as a floppy disk, a usb disk, a removable hard disk, a Read-only memory (ROM), a random-access memory (RAM, randomAccessMemory), a magnetic disk or an optical disk of a computer, etc., including several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute the method of the embodiments of the present application.
The foregoing description is only of the preferred embodiments of the present application, and is not intended to limit the scope of the application, but rather is intended to cover any equivalents of the structures or equivalent processes disclosed herein or in the alternative, which may be employed directly or indirectly in other related arts.
Claims (10)
1. A method of customer complaint treatment, the method comprising:
manually filling in a complaint work order and user complaint voice of the structured complaint template and the target customer service account number are obtained; wherein, the manual filling complaint worksheets correspond to the complaint voices of the users;
converting the user complaint voice into user complaint text information;
and filling in the structural complaint template based on the user complaint text information and the manually filled-in complaint worksheets to obtain standard complaint worksheets.
2. The method for processing the user complaints according to claim 1, wherein the filling in the structured complaint template based on the user complaint text information and the manually filled-in complaint form to obtain a standard complaint form comprises:
extracting key information from the user complaint text information according to a preset complaint vocabulary table to obtain first key information, and extracting key information from the manually filled complaint work order to obtain second key information;
Identifying a distinct keyword from the first keyword information that is different from the second keyword information, and a common keyword that is the same as the second keyword information;
if the total number of the first key words in the first key information is consistent with the total number of the second key words in the second key information, filling in the structured complaint template based on the distinguishing key words and the common key words to obtain the standard complaint work order;
if the total number of the first key words in the first key information is inconsistent with the total number of the second key words in the second key information, filling in the structured complaint template based on the distinguishing key words and the second key information, and obtaining the standard complaint work order.
3. The user complaint handling method of claim 1, wherein the identifying a distinct keyword from the first keyword that is different from the second keyword and a common keyword that is the same as the second keyword includes:
calculating the characteristic distance between the first key word of each item and the second key word of each item;
And identifying a distinguishing keyword which is different from the second keyword from the first keyword and a common keyword which is the same as the second keyword according to the characteristic distance.
4. The user complaint handling method of claim 1, wherein after converting the user complaint speech into user complaint text information, the method further comprises:
extracting emotion keywords from the user complaint text information;
and sequencing the standard complaint worksheets based on the emotion keywords to obtain the arrangement sequence of the standard complaint worksheets, so that the staff can process the standard complaint worksheets based on the arrangement sequence.
5. The method for processing user complaints according to claim 1, wherein the method further comprises, after filling in the structured complaint template based on the user complaint text information and the manually filled-in complaint form to obtain a standard complaint form:
judging whether the standard complaint work orders are complaint work orders or not by using a complaint work order judging model, wherein the complaint work order judging model is an NLP language model;
if the standard complaint work order is a complaint work order, inputting the standard complaint work order into a complaint work order classification model for classification, and obtaining a classification result output by the complaint work order classification model.
6. The method of claim 5, wherein after the determining whether the standard complaint work order is a complaint work order, the method further comprises:
if the standard complaint work order is not the complaint work order, outputting the standard complaint work order to a person for confirming the complaint work order;
and if the manually input complaint work order confirmation information is received, executing the step of inputting the standard complaint work order into the complaint work order classification model to classify, and obtaining a classification result output by the complaint work order classification model.
7. The method for processing user complaints according to claim 1, wherein the method further comprises, after filling in the structured complaint template based on the user complaint text information and the manually filled-in complaint form to obtain a standard complaint form:
and generating a complaint system report based on the type of the standard complaint work order, the processing time of the standard complaint work order and the complaint index of the complaint standard work order.
8. A user complaint handling device, characterized in that the user complaint handling device comprises:
the acquisition module is used for acquiring a structured complaint template, a manual filling complaint work order of a target customer service account and a user complaint voice; wherein, the manual filling complaint worksheets correspond to the complaint voices of the users;
The conversion module is used for converting the user complaint voice into user complaint text information;
and the filling module is used for filling the structured complaint template based on the user complaint text information and the manual filling complaint work order to obtain a standard complaint work order.
9. A user complaint handling device, comprising: a processor, a memory and a user complaint handling program stored in the memory, which when executed by the processor, implements the steps of the user complaint handling method as claimed in any one of claims 1 to 7.
10. A computer-readable storage medium, wherein a user complaint handling program is stored on the computer-readable storage medium, which when executed by a processor implements the user complaint handling method according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310597190.9A CN116595966A (en) | 2023-05-24 | 2023-05-24 | User complaint processing method, device, equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310597190.9A CN116595966A (en) | 2023-05-24 | 2023-05-24 | User complaint processing method, device, equipment and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116595966A true CN116595966A (en) | 2023-08-15 |
Family
ID=87606000
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310597190.9A Pending CN116595966A (en) | 2023-05-24 | 2023-05-24 | User complaint processing method, device, equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116595966A (en) |
-
2023
- 2023-05-24 CN CN202310597190.9A patent/CN116595966A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11004013B2 (en) | Training of chatbots from corpus of human-to-human chats | |
| US11507850B2 (en) | System and method for call centre management | |
| US20210224832A1 (en) | Method and apparatus for predicting customer purchase intention, electronic device and medium | |
| US20160117624A1 (en) | Intelligent meeting enhancement system | |
| CN111182162B (en) | Telephone quality inspection method, device, equipment and storage medium based on artificial intelligence | |
| CN110493019B (en) | Automatic generation method, device, equipment and storage medium of conference summary | |
| US11798539B2 (en) | Systems and methods relating to bot authoring by mining intents from conversation data via intent seeding | |
| CN113688221B (en) | Model-based conversation recommendation method, device, computer equipment and storage medium | |
| US20190266254A1 (en) | System to process electronic records using a request orchestration platform | |
| US11816609B2 (en) | Intelligent task completion detection at a computing device | |
| US20220138770A1 (en) | Method and apparatus for analyzing sales conversation based on voice recognition | |
| CN115665325A (en) | Intelligent outbound method, device, electronic equipment and storage medium | |
| US20230102179A1 (en) | Computer systems and computer-based methods for automated caller intent prediction | |
| CN112235470B (en) | Incoming call client follow-up method, device and equipment based on voice recognition | |
| CN111597818A (en) | Call quality inspection method, call quality inspection device, computer equipment and computer readable storage medium | |
| US20230090049A1 (en) | Computer systems and computer-based methods for automated callback scheduling utilizing call duration prediction | |
| CN113873088A (en) | Voice call interaction method and device, computer equipment and storage medium | |
| US20230244968A1 (en) | Smart Generation and Display of Conversation Reasons in Dialog Processing | |
| CN116886823A (en) | Seat quality inspection method, device, equipment and medium | |
| CN111145053A (en) | Enterprise law consultant management system and method based on artificial intelligence | |
| WO2023050669A1 (en) | Neural network-based information pushing method and system, device, and medium | |
| CN116595966A (en) | User complaint processing method, device, equipment and storage medium | |
| CN113111157B (en) | Question-answer processing method, device, computer equipment and storage medium | |
| CN114239602A (en) | Session method, apparatus and computer program product | |
| CN114333813A (en) | Implementation method and device for configurable intelligent voice robot 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 |