CN117668222A - Method for judging electricity utilization clients with complaint tendency - Google Patents
Method for judging electricity utilization clients with complaint tendency Download PDFInfo
- Publication number
- CN117668222A CN117668222A CN202311430279.2A CN202311430279A CN117668222A CN 117668222 A CN117668222 A CN 117668222A CN 202311430279 A CN202311430279 A CN 202311430279A CN 117668222 A CN117668222 A CN 117668222A
- Authority
- CN
- China
- Prior art keywords
- complaint
- data
- users
- tendency
- emotion
- 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
- 238000000034 method Methods 0.000 title claims abstract description 43
- 230000005611 electricity Effects 0.000 title claims abstract description 33
- 230000008451 emotion Effects 0.000 claims abstract description 35
- 230000011218 segmentation Effects 0.000 claims abstract description 18
- 238000012545 processing Methods 0.000 claims abstract description 17
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 13
- 238000012549 training Methods 0.000 claims abstract description 8
- 238000005457 optimization Methods 0.000 claims abstract description 4
- 238000013527 convolutional neural network Methods 0.000 claims description 10
- 230000008909 emotion recognition Effects 0.000 claims description 10
- 230000015654 memory Effects 0.000 claims description 9
- 230000007246 mechanism Effects 0.000 claims description 8
- 238000004458 analytical method Methods 0.000 claims description 6
- 238000013528 artificial neural network Methods 0.000 claims description 6
- 238000009826 distribution Methods 0.000 claims description 5
- 230000006870 function Effects 0.000 claims description 5
- 230000004913 activation Effects 0.000 claims description 4
- 238000004140 cleaning Methods 0.000 claims description 3
- 238000010276 construction Methods 0.000 claims description 3
- 238000002372 labelling Methods 0.000 claims description 3
- 238000001228 spectrum Methods 0.000 claims description 3
- 238000007781 pre-processing Methods 0.000 abstract description 3
- 230000006399 behavior Effects 0.000 description 4
- 230000003993 interaction Effects 0.000 description 3
- 238000003860 storage Methods 0.000 description 3
- 238000013135 deep learning Methods 0.000 description 2
- 238000012217 deletion Methods 0.000 description 2
- 230000037430 deletion Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000005236 sound signal Effects 0.000 description 2
- 238000012706 support-vector machine Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000010224 classification analysis Methods 0.000 description 1
- 238000013480 data collection Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 230000007787 long-term memory Effects 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 239000011159 matrix material Substances 0.000 description 1
- 238000005065 mining Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 238000003062 neural network model Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000003909 pattern recognition Methods 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 230000000306 recurrent effect Effects 0.000 description 1
- 238000000611 regression analysis Methods 0.000 description 1
- 230000008439 repair process Effects 0.000 description 1
- 230000006403 short-term memory Effects 0.000 description 1
- 238000013179 statistical model Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Landscapes
- Telephonic Communication Services (AREA)
Abstract
The invention discloses a method for judging an electricity consumer with complaint tendency, which relates to the technical field of data processing and comprises the following steps: and (3) constructing a judgment algorithm model by combining the emotion characteristics of the electricity customers and the keyword word stock with complaint tendency through data acquisition, data preprocessing, model training and model optimization, and judging whether the incoming call users are users with complaint tendency and the complaint probability of the users or not through the emotion characteristics shown by the incoming call voice of the users, the word segmentation result of the incoming call single text of the users and the comparison result of the keyword word stock with complaint tendency. According to the invention, a judgment algorithm model is constructed, so that users with high complaint tendency probability and the complaints thereof are predicted in advance, and each basic-level power supply service personnel is reminded to develop related work in a targeted manner, thereby achieving the aims of efficiently solving the complaints of the users and improving the customer satisfaction rate, and the complaint tendency recognition method is more accurate than the recognition of the complaint tendency of the users by using a text word segmentation mode.
Description
Technical Field
The invention relates to the technical field of data processing, in particular to a method for judging an electricity consumer with complaint tendency.
Background
When a customer call is received by a power grid customer service person, the customer call is recorded according to the problem of the customer, and is converted into a customer service work order, and then the work order is forwarded to a specific power supply bureau for processing. The types of worksheets are divided into: several major categories of electricity service, fault repair, consultation inquiry, opinion suggestion, complaints, etc.
After receiving the customer call, the customer service personnel records and forwards the work order, which is either a complaint type work order or a work order which is not a complaint type. In practice, there are some users who have a tendency to complain, and these users should also be of particular interest.
In the industry, a technology of word segmentation and modeling is adopted in many cases, and users with complaint tendency are judged by judging emotion words, but the method depends on the completeness of customer service personnel record, and description is sufficiently detailed and not omitted.
Disclosure of Invention
Aiming at the problems in the prior art, the invention provides a method for judging the electric clients with complaint tendency, which can identify the complaint tendency of the users, effectively solve the user complaints, reduce the user complaints and improve the client satisfaction rate. The specific technical scheme is as follows:
a judging method for electric clients with complaint tendency comprises the following steps:
data acquisition, namely acquiring historical complaint data, electricity consumption data and customer service record related data of an electricity consumption customer;
data processing, namely cleaning, de-duplicating and labeling the acquired data;
model training, namely introducing a model for identifying the voice emotion by utilizing the preprocessed data, identifying acoustic characteristics through a training module, and learning rules and characteristics of complaint tendency of electric clients;
model optimization, namely searching a non-complaint work order of a complaint user for the last time before complaint by using a user complaint work order of the last three years, performing word segmentation, identifying keywords and word frequency, performing comparison analysis with the word segmentation result of a customer complaint work order of the user who does not have complaint event, and forming a keyword word stock with complaint tendency through a construction module;
and judging the complaint tendency, constructing a judging algorithm model by combining the emotion characteristics of the electricity customers and the keyword word stock with the complaint tendency, and judging whether the incoming call users are users with the complaint tendency and the complaint probability of the users by jointly judging the emotion characteristics shown by the incoming call voices of the users, the word segmentation results of the incoming call single texts of the users and the comparison results of the keyword word stock with the complaint tendency.
Preferably, the historical complaint data includes complaint content, complaint time, and complaint person information.
Preferably, the electricity consumption data includes electricity consumption, electricity load and electricity fee information.
Preferably, the customer service record includes customer service content, service time, and service personnel information.
Preferably, the data cleansing specifically includes correcting errors in the data, deleting duplicate information, and processing missing values, filling the missing values by means, median or mode, for outliers, automatic deletion may be performed.
Preferably, the data deduplication is specifically deleting duplicated data by combining a plurality of field-specific rules.
Preferably, the data tagging is specifically to convert data into a format that is easy to understand and use, and form tags of keyword lexicons.
Preferably, the model for recognizing the voice emotion adopts a method of supporting a vector machine to model and recognize emotion signals, and a convolutional neural network combining a data balance and attention mechanism and a voice emotion recognition method of a long-time memory unit are used for recognizing four emotions of happiness, anger, calm and sadness through a recognition module, and grading anger emotion, including anger, intense dissatisfaction and dissatisfaction.
Preferably, the voice emotion recognition method comprises the following steps:
firstly, extracting a logarithmic Mel spectrogram from voice samples in a voice emotion data set, and carrying out segmentation processing according to sample distribution characteristics so as to realize data balance processing.
The method is used for learning high-level segment voice features by fine-tuning a pre-trained convolutional neural network model in a segmented mel spectrum dataset.
The method comprises the steps of inputting the learned segmented convolutional neural network characteristics into a long-short-time memory unit with an attention mechanism for learning the distinguishing characteristics by considering the difference of emotion recognition actions of different segment areas in voice, and realizing voice emotion classification by combining the long-short-time memory unit and a neural network activation function layer.
Compared with the prior art, the invention has the beneficial effects that:
1. according to the invention, whether the incoming call user is a user with complaint tendency or not and the complaint probability thereof are judged together by comparing the emotion characteristics shown by the incoming call voice of the user, the word segmentation result of the incoming call electric sheet text of the user with the comparison result of the keyword word stock with complaint tendency, and the recognition of the complaint tendency of the user by using the text word segmentation method is more accurate.
2. According to the invention, a judgment algorithm model is constructed, so that users with high complaint tendency probability and the complaints thereof are predicted in advance, and each basic-level power supply service personnel is reminded to develop related work in a targeted manner, thereby achieving the aims of efficiently solving the complaints of the users, reducing the complaints of the users and improving the satisfaction rate of the clients.
3. According to the invention, after the judgment algorithm model is embedded into the telephone traffic system, the voice call of the user can be analyzed in real time, the emotion state of the user and the probability of complaint tendency are fed back to the customer service personnel in real time, and the customer service personnel are reminded of adjusting the corresponding conversation immediately.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below. Like elements or portions are generally identified by like reference numerals throughout the several figures. In the drawings, elements or portions thereof are not necessarily drawn to scale.
FIG. 1 is a flow chart of a method for determining customers who have a tendency to complain according to the present invention.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and fully with reference to the accompanying drawings, in which it is evident that the embodiments described are some, but not all embodiments of the invention. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
It should be understood that the terms "comprises" and "comprising," when used in this specification and the appended claims, specify the presence of stated features, integers, steps, operations, elements, and/or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and/or groups thereof.
It is also to be understood that the terminology used in the description of the invention is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. As used in this specification and the appended claims, the singular forms "a," "an," and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise.
It should be further understood that the term "and/or" as used in the present specification and the appended claims refers to any and all possible combinations of one or more of the associated listed items, and includes such combinations.
Referring to fig. 1, the present invention provides a method for determining customers who have a tendency to complain, the method comprising the following steps:
s1: and (3) data acquisition: and collecting related data such as historical complaint data, electricity consumption data, customer service records and the like of the electricity consumption customers through a data acquisition module.
The historical complaint data comprises complaint content, complaint time and complaint person information, and the acquisition module collects the historical complaint data through channels of the customer service center and the complaint management system. The electricity consumption data comprise electricity consumption, electricity consumption load and electricity fee information, and the acquisition module collects the electricity consumption data through channels of the electric power monitoring system and the electric quantity statistics system.
The method comprises the steps that electricity consumption data of an electricity consumption customer are collected through a data collection module to judge complaint risks when the customer calls, when the customer calls are received, the system can be matched with whether the customer has business interaction behaviors recently, the complaint cause categories corresponding to the business interaction behaviors recently are determined, then the system can count time intervals and number of calls between the customer calls and the business interaction behaviors recently, the time intervals and the number of calls are used as input quantities to input a pre-trained statistical model, and a certainty interval of the call is obtained. And finally, calculating a corresponding complaint risk score according to the certainty interval of the incoming call, and judging whether the incoming call behavior is a complaint incoming call according to a preset risk threshold.
The customer service records comprise customer service content, service time and service personnel information, and the acquisition module collects the customer service records through a marketing system, a customer service voice platform and an Internet channel platform.
S2: data preprocessing: the data processing module is used for carrying out preprocessing work of cleaning, de-duplication and labelling on the collected data so as to prepare the data for algorithm model training.
Data cleansing is to correct errors in data, delete duplicate information, and process missing values, fill in missing values by mean, median, or mode, and for outliers, automatically delete. Data deduplication is the deletion of duplicate data by combining multiple field-specific rules. Data tagging is the conversion of data into a format that is easy to understand and use, forming tags for keyword lexicons.
S3: model training: and introducing a model for identifying the emotion of the voice by utilizing the preprocessed data, identifying acoustic characteristics by a training module, and learning rules and characteristics of complaint trends of the electricity customers so as to realize judgment of the emotion characteristics of the electricity customers.
The model for recognizing the voice emotion adopts a method of supporting a vector machine to model and recognize emotion signals, and combines a convolutional neural network of a data balance and a attention mechanism with a voice emotion recognition method of a long-short-term memory unit, four emotions of happiness, anger, calm and sadness are recognized through a recognition module, anger is identified in an important way in power supply enterprise service, anger emotion types can be classified, and the anger emotion types comprise anger, intense dissatisfaction and dissatisfaction.
Support vector machines are a common classification method, and in the field of machine learning, a supervised learning model is usually used for pattern recognition, classification and regression analysis. The support vector machine method is based on the structural risk minimization theory, an optimal hyperplane is constructed in the feature space, so that the learner is globally optimized, a certain probability is met for the expectation of the whole sample space, the analysis is carried out for the linear separable condition, and for the linear inseparable condition, the nonlinear mapping algorithm is used for converting the sample which is linearly inseparable in the low-dimensional input space into the high-dimensional feature space to make the sample linearly separable, so that the high-dimensional feature space can carry out linear analysis on the nonlinear features of the sample by adopting the linear algorithm.
The voice emotion recognition method comprises the following steps:
s11: firstly, extracting a logarithmic Mel spectrogram from voice samples in a voice emotion data set, and carrying out segmentation processing according to sample distribution characteristics so as to realize data balance processing.
The logarithmic mel-frequency spectrogram is a characteristic representation method for audio signal processing, which can convert an audio signal into a two-dimensional matrix representing time in the horizontal direction and frequency in the vertical direction, and then take the logarithm of the value of each element to better represent the dynamic range of data.
S12: the method is used for learning high-level segment voice features by fine-tuning a pre-trained convolutional neural network model in a segmented mel spectrum dataset.
The convolutional neural network is a feedforward neural network which comprises convolutional calculation and has a deep structure, and is one of representative algorithms of deep learning.
S13: the method comprises the steps of inputting the learned segmented convolutional neural network characteristics into a long-short-time memory unit with an attention mechanism for learning the distinguishing characteristics by considering the difference of emotion recognition actions of different segment areas in voice, and realizing voice emotion classification by combining the long-short-time memory unit and a neural network activation function layer.
The long and short term memory unit with an attention mechanism is an improved recurrent neural network model, and the modeling capability of the model to sequence data is enhanced by introducing the attention mechanism. The Softmax layer is the most common activation function in a neural network, and can convert the output of the neural network into a probability distribution, and in deep learning, softmax is widely applied to classification problems and some tasks requiring the output of the probability distribution.
S4: model optimization: the method comprises the steps of searching a non-complaint work order of a complaint user before complaining by using a user complaint work order of the past three years through a data acquisition module, performing word segmentation through a data processing module, identifying keywords and word frequency, performing comparison analysis with the word segmentation result of a customer complaint work order of a user who does not have a complaint event, and forming a keyword word stock with complaint tendency through a construction module.
The word frequency analysis or the text mining technology of keyword clustering is used for extracting words related to complaints from the complaint work orders, and keywords with complaint tendency can be screened from the extracted words by means of a word vector model.
S5: complaint tendency judgment: and (3) a mode of combining the emotion characteristics of the electricity customers and the keyword word stock with complaint tendency is established, a judgment algorithm model is constructed, and whether the incoming call users are users with complaint tendency or not and the complaint probability of the users are judged by the emotion characteristics shown by the incoming call voice of the users, the word segmentation result of the incoming call single text of the users and the comparison result of the keyword word stock with complaint tendency.
In a word, whether the incoming call user is a user with complaint tendency or not and the complaint probability of the user are judged by comparing the emotion characteristics shown by the incoming call voice of the user, the word segmentation result of the incoming call electric sheet text of the user and the keyword word stock with complaint tendency; by constructing a judgment algorithm model, users with high complaint tendency probability and the complaints thereof are predicted in advance, so that power supply service personnel of each base layer are reminded of carrying out related work in a targeted manner, and the aims of efficiently solving the complaints of the users, reducing the complaints of the users and improving the satisfaction rate of the clients are achieved; after the judgment algorithm model is embedded into the telephone traffic system, voice call of the user can be analyzed in real time, the emotion state of the user and the probability of complaint tendency are fed back to customer service personnel in real time, and the customer service personnel are reminded of adjusting the corresponding conversation in real time.
Those of ordinary skill in the art will appreciate that the elements of the examples described in connection with the embodiments disclosed herein may be implemented as electronic hardware, computer software, or combinations of both, and that the elements of the examples have been described generally in terms of functionality in the foregoing description to clearly illustrate this interchangeability of hardware and software. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the solution. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present invention.
In the embodiments provided in the present invention, it should be understood that the division of the units is merely a logic function division, and there may be other division manners in actual implementation, for example, multiple units may be combined into one unit, one unit may be split into multiple units, or some features may be omitted.
In addition, each functional unit in the embodiments of the present invention may be integrated in one processing unit, or each unit may exist alone physically, or two or more units may be integrated in one unit. The integrated units may be implemented in hardware or in software functional units.
The integrated units, if implemented in the form of software functional units and sold or used as stand-alone products, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present invention may be embodied essentially or in part or all of the technical solution or in part in the form of a software product stored in a storage medium, including instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a Read-0nlyMemory (ROM), a random access memory (RAM, randomAccessMemory), a removable hard disk, a magnetic disk, or an optical disk, or the like, which can store program codes.
Finally, it should be noted that: the above embodiments are only for illustrating the technical solution of the present invention, and not for limiting the same; although the invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some or all of the technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit of the invention, and are intended to be included within the scope of the appended claims and description.
Claims (9)
1. A judging method for an electricity customer with complaint tendency is characterized by comprising the following steps:
data acquisition, namely acquiring historical complaint data, electricity consumption data and customer service record related data of an electricity consumption customer;
data processing, namely cleaning, de-duplicating and labeling the acquired data;
model training, namely introducing a model for identifying the voice emotion by utilizing the preprocessed data, identifying acoustic characteristics through a training module, and learning rules and characteristics of complaint tendency of electric clients;
model optimization, namely searching a non-complaint work order of a complaint user for the last time before complaint by using a user complaint work order of the last three years, performing word segmentation, identifying keywords and word frequency, performing comparison analysis with the word segmentation result of a customer complaint work order of the user who does not have complaint event, and forming a keyword word stock with complaint tendency through a construction module;
and judging the complaint tendency, constructing a judging algorithm model by combining the emotion characteristics of the electricity customers and the keyword word stock with the complaint tendency, and judging whether the incoming call users are users with the complaint tendency and the complaint probability of the users by jointly judging the emotion characteristics shown by the incoming call voices of the users, the word segmentation results of the incoming call single texts of the users and the comparison results of the keyword word stock with the complaint tendency.
2. The method of claim 1, wherein the historical complaint data includes complaint content, complaint time, and complaint person information.
3. The method of claim 1, wherein the electricity consumption data includes electricity consumption amount, electricity load, and electricity fee information.
4. The method of claim 1, wherein the customer service record includes customer service content, service time, and service personnel information.
5. The method according to claim 1, wherein the data cleansing specifically comprises correcting errors in the data, deleting duplicate information, and processing missing values, filling the missing values by mean, median or mode, and automatically deleting the missing values.
6. The method of claim 1, wherein the data deduplication is performed by combining a plurality of field-specific rules to delete duplicate data.
7. The method of claim 1, wherein the data tagging is specifically a tag that converts data into a format that is easy to understand and use, forming a keyword lexicon.
8. The method for judging customers who have a tendency to complain according to claim 1, wherein the model for speech emotion recognition models and recognizes emotion signals by using a method of supporting a vector machine, and the speech emotion recognition method of a convolutional neural network and a long and short memory unit combining data balance and attention mechanisms recognizes four emotions of happiness, anger, calm and sadness by a recognition module, and classifies anger emotion, including anger, intense dissatisfaction and dissatisfaction.
9. The method for determining a customer who has a tendency to complain according to claim 8, wherein the voice emotion recognition method comprises the steps of:
firstly, extracting a logarithmic Mel spectrogram from voice samples in a voice emotion data set, and carrying out segmentation processing according to sample distribution characteristics so as to realize data balance processing.
The method is used for learning high-level segment voice features by fine-tuning a pre-trained convolutional neural network model in a segmented mel spectrum dataset.
The method comprises the steps of inputting the learned segmented convolutional neural network characteristics into a long-short-time memory unit with an attention mechanism for learning the distinguishing characteristics by considering the difference of emotion recognition actions of different segment areas in voice, and realizing voice emotion classification by combining the long-short-time memory unit and a neural network activation function layer.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311430279.2A CN117668222A (en) | 2023-10-31 | 2023-10-31 | Method for judging electricity utilization clients with complaint tendency |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311430279.2A CN117668222A (en) | 2023-10-31 | 2023-10-31 | Method for judging electricity utilization clients with complaint tendency |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117668222A true CN117668222A (en) | 2024-03-08 |
Family
ID=90063089
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311430279.2A Pending CN117668222A (en) | 2023-10-31 | 2023-10-31 | Method for judging electricity utilization clients with complaint tendency |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117668222A (en) |
-
2023
- 2023-10-31 CN CN202311430279.2A patent/CN117668222A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11004013B2 (en) | Training of chatbots from corpus of human-to-human chats | |
| US7912714B2 (en) | Method for segmenting communication transcripts using unsupervised and semi-supervised techniques | |
| US9910845B2 (en) | Call flow and discourse analysis | |
| US20210224832A1 (en) | Method and apparatus for predicting customer purchase intention, electronic device and medium | |
| US11507850B2 (en) | System and method for call centre management | |
| WO2021068843A1 (en) | Emotion recognition method and apparatus, electronic device, and readable storage medium | |
| US9477752B1 (en) | Ontology administration and application to enhance communication data analytics | |
| US11950020B2 (en) | Methods and apparatus for displaying, compressing and/or indexing information relating to a meeting | |
| CN109710766B (en) | Complaint tendency analysis early warning method and device for work order data | |
| US20150066549A1 (en) | System, Method and Apparatus for Voice Analytics of Recorded Audio | |
| CN112101971B (en) | Sensitive client identification method, system and storage medium | |
| CN111667284A (en) | Customer service switching method and device | |
| CN113468296A (en) | Model self-iteration type intelligent customer service quality inspection system and method capable of configuring business logic | |
| CN116663664A (en) | Customer marketing scene data analysis system and method based on NLP algorithm | |
| CN112235470B (en) | Incoming call client follow-up method, device and equipment based on voice recognition | |
| CN113434670A (en) | Method and device for generating dialogistic text, computer equipment and storage medium | |
| CN117151870A (en) | Portrait behavior analysis method and system based on guest group | |
| CN113297365A (en) | User intention determination method, device, equipment and storage medium | |
| CN116828109A (en) | Intelligent evaluation method and system for telephone customer service quality | |
| CN117668222A (en) | Method for judging electricity utilization clients with complaint tendency | |
| CN115936538A (en) | Intelligent customer service management system for power grid | |
| CN115687788A (en) | Intelligent business opportunity recommendation method and system | |
| CN114529383A (en) | Method and system for realizing tax payment tracking and tax loss early warning | |
| CN115080732A (en) | Complaint work order processing method and device, electronic equipment and storage medium | |
| Severiukhina et al. | Segment-wise Users' Response Prediction based on Activity Traces in Online Social Networks |
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 |