CN114418327A - Automatic order recording and intelligent order dispatching method for customer service system - Google Patents
Automatic order recording and intelligent order dispatching method for customer service system Download PDFInfo
- Publication number
- CN114418327A CN114418327A CN202111609645.1A CN202111609645A CN114418327A CN 114418327 A CN114418327 A CN 114418327A CN 202111609645 A CN202111609645 A CN 202111609645A CN 114418327 A CN114418327 A CN 114418327A
- Authority
- CN
- China
- Prior art keywords
- order
- intelligent
- model
- work order
- data
- 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
Images
Classifications
-
- 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
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0631—Resource planning, allocation, distributing or scheduling for enterprises or organisations
- G06Q10/06311—Scheduling, planning or task assignment for a person or group
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/237—Lexical tools
- G06F40/242—Dictionaries
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/279—Recognition of textual entities
- G06F40/289—Phrasal analysis, e.g. finite state techniques or chunking
-
- 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
- G06Q30/06—Buying, selling or leasing transactions
- G06Q30/0601—Electronic shopping [e-shopping]
- G06Q30/0613—Third-party assisted
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Business, Economics & Management (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Artificial Intelligence (AREA)
- General Engineering & Computer Science (AREA)
- Human Resources & Organizations (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Economics (AREA)
- Strategic Management (AREA)
- Computational Linguistics (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Finance (AREA)
- Accounting & Taxation (AREA)
- General Business, Economics & Management (AREA)
- General Health & Medical Sciences (AREA)
- Entrepreneurship & Innovation (AREA)
- Development Economics (AREA)
- Marketing (AREA)
- Health & Medical Sciences (AREA)
- Tourism & Hospitality (AREA)
- Quality & Reliability (AREA)
- Operations Research (AREA)
- Educational Administration (AREA)
- Game Theory and Decision Science (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention discloses an automatic order recording and intelligent order dispatching method for a customer service system, which comprises the following steps: step 1, obtaining dialogue interaction text data; step 2, extracting and synthesizing the record list associated data through an intelligent record list associated algorithm model; step 3, automatically filling in work order content items; step 4, checking the work order content and optimizing and supplementing adjustment; step 5, submitting and generating a work order; step 6, inputting the work order into an intelligent order dispatching identification algorithm model, and calculating and outputting work order processing category data; and 7, inputting the work order processing type data into an intelligent order dispatching model to automatically dispatch the work order. The method is based on a multi-algorithm model, and forms a stable and well-used data model through the training of a problem classification recognition model, thereby effectively realizing the quick and automatic dispatch of the customer service work order.
Description
Technical Field
The invention belongs to the technical field of text mining analysis of a customer service system, and particularly relates to an automatic order recording and intelligent order dispatching method for the customer service system.
Background
When the customer service agent representative is processing customer consultation, if the solution can not be given at the first time, the customer service agent needs to input work orders and then upgrade the work orders to second-line personnel, the work orders are classified and processed, and the work orders are sent to corresponding responsibility departments and cooperative departments.
In actual work, the following problems exist: (1) the problem types are judged by depending on manpower and keywords, so that the accuracy of automatic identification and distribution is low, and the difficulty of manual order distribution is high; (2) the content of the work order complained by the customer depends on manual work to record the work order item by item, the efficiency of recording the work order is low, and the content of the work order is not standardized; (3) the prior art has the defects that keyword grabbing in voice transcription data to judge a dispatch list causes misdispatch when transcription is inaccurate, keywords are fuzzy and the like.
The Chinese patent CN202110919603.1 discloses a method for automatically classifying and allocating call center customer service work orders based on text big data, which comprises the following steps: step 1, a work order system is connected, a work order text is input and then connected in an RESTful mode, and the work order text is obtained; step 2, preprocessing preparation work of the work order text; step 3, constructing a call center customer service work order classification model based on a rapid text classification algorithm; step 4, performing word segmentation on the text by the Jieba word segmentation, removing some common stop words, and submitting the result to a customer service work order classification model after processing; and 5, analyzing the customer service work order classification model to obtain work order classification recommendation and a recommendation result of a flow distribution department. But the redundancy and the error capability are not enough, and the continuous iterative optimization can not be carried out under the condition that the existing system does not conflict.
The chinese invention patent CN201811608156.2 discloses a method for realizing fast generation and dispatch of workflow based on special software and algorithm, which obtains the original data of the content of the voice-to-text of both parties of the call through the special software and algorithm; the original data need to be grouped according to intents and then classified according to the service scenes of clients; and acquiring keywords of the client incoming call feedback appeal according to the service scene and the intention, calculating according to a weight formula, synthesizing time, place and appeal content work order contents, and automatically supplementing to a work order acceptance interface according to an algorithm. However, automatic dispatch relies on keyword crawling to determine the traffic class, and when the translation is inaccurate, the keywords are fuzzy, the configuration is incomplete or repeated, a misdispatch or a failed dispatch card may occur.
The Chinese invention patent CN201410377569.X discloses a work order automatic dispatch system and a method; the system comprises a system operation configuration module for providing system operation parameters, wherein the system operation parameters comprise switch configuration, work order issuing resource positioning, ring tone reminding file configuration and whether to start a ring tone reminding switch; the task service configuration module is used for providing the functions of timing task service configuration, including work number information configuration, time interval configuration, work order address resource information acquisition configuration and supporting the configuration of time interval to automatically acquire national network work order information; the operation management module is used for managing logs and reminding the dispatching failure of the work order condition automatically processed by the system; and the business rule configuration module is used for configuring the conditions of automatic dispatching of various work orders of fault repair, business consultation, complaint, report, suggestion and opinion. However, misdispatch occurs when the manual classification judgment is wrong, and the requirement on experience of customer service staff is high.
Disclosure of Invention
Aiming at the defects of the prior art, the invention provides an automatic order recording and intelligent order dispatching method for a customer service system.
In order to achieve the purpose, the invention adopts the following technical scheme:
an automatic order recording and intelligent order dispatching method for a customer service system comprises the following steps:
step 1, obtaining dialogue interaction text data;
the method specifically comprises the following steps: the system acquires interactive text data of a user and a customer service seat in conversation of a work order to be built, accesses the user by voice, acquires real-time voice transcribed text data, and acquires the conversation text data by an online customer service;
step 2, extracting and synthesizing the record list associated data through an intelligent record list associated algorithm model;
the method specifically comprises the following steps: inputting the dialogue text data obtained in the step 1 into an intelligent recording sheet association algorithm model, and extracting and synthesizing recording sheet association data; associated data extraction includes, but is not limited to: information such as customer name, contact telephone, province, card writing channel number, access mode, product, order number and the like, and summary content of work order problem content is spliced by adopting a system summary algorithm;
step 3, automatically filling in work order content items;
the method specifically comprises the following steps: the system automatically selects and fills the list page according to the associated data and the abstract content in the step 2;
step 4, checking the work order content and optimizing and supplementing adjustment;
the method specifically comprises the following steps: the customer service seat checks whether the work order page filling data is accurate or not, and corresponding optimization and supplementary adjustment are carried out;
step 5, after checking, submitting and generating a work order;
step 6, inputting the work order into an intelligent order dispatching identification algorithm model, and calculating and outputting work order processing category data;
the method specifically comprises the following steps: the system acquires work order content information, inputs the work order content information into an intelligent order dispatching identification algorithm model, calculates the identification algorithm model and outputs work order processing category data;
step 7, inputting the work order processing category data into an intelligent order dispatching model to carry out automatic order dispatching;
the method specifically comprises the following steps: the work order processing category data is used as a parameter to be input into an order dispatching engine to automatically dispatch the order to a undertaking department for circulation processing.
The invention further discloses that the intelligent order recording association algorithm model, the intelligent order dispatching identification algorithm model and the intelligent order dispatching model are independently and locally deployed, and are separately and independently operated with the existing customer service system, interface calling is provided as required, and training and operation are not interfered with each other.
For further explanation of the present invention, the process of the intelligent recording list association algorithm model in step 2 includes:
21) performing data cleaning on historical work order data, and selecting a sufficient number of standard samples to form a historical sample set;
22) performing work order data by using a natural language processing technology, performing word segmentation on the unstructured text, calculating word frequency and forming word vectors;
23) for the work order sample which forms the word vector and is formatted, carrying out classification training on information such as products, purchase channels, problem description, user appeal and the like by using a machine learning algorithm, and finally obtaining a standardized entry classification template through multiple training and optimization;
24) when a new client consults, voice data is converted into text data, user information and consultation contents are used as parameters and are transmitted into a trained classification model, and key information required by work order entry is extracted;
25) and the customer service checks whether the data generated by the intelligent record list is accurate or not, and performs corresponding optimization and supplementary adjustment.
Further explaining the present invention, the natural language processing technique in step 22) is specifically: the method comprises the steps of judging words in text segments by dictionary generation, word segmentation and text feature extraction methods, carrying out quantitative statistics on the occurrence frequency, the degree of condensation and the degree of freedom to generate a thematic corpus dictionary, carrying out text data word segmentation by using a dictionary and a general dictionary newly generated by a specific corpus, and vectorizing a Chinese text.
For further explanation of the present invention, the process of the intelligent dispatch identification algorithm model in step 6 includes:
61) after the customer service submits the work order, the content of the work order is automatically transmitted and input into an intelligent order delivery identification algorithm model, and the assignment identification of the work order problem is matched with the order delivery path flow;
62) the system extracts and submits work order text content as intelligent order delivery recognition algorithm model input;
63) measuring, calculating and classifying the theme and the distance through the trained online intelligent order delivery recognition algorithm model;
64) the intelligent order dispatching recognition algorithm model outputs the service problem category result matched by the work order;
65) triggering an intelligent dispatching model, and dispatching according to the configured work order acceptance flow.
For further explanation of the present invention, the training process of the intelligent dispatch model in step 7 includes:
71) inputting a large amount of historical work order content texts, extracting text data, performing word segmentation processing on the texts, and the like;
72) performing similarity characteristic analysis on the processed participles, vectorizing an analysis object, realizing synonym recommendation by using methods of 'tfidf', 'counts' and 'hashing', generating a thematic dictionary, and performing dictionary learning and updating supplementation;
73) clustering texts by using a clustering algorithm to generate a feature dictionary, and performing clustering learning on text features by using a plurality of algorithms such as an LDA theme algorithm, a k mean algorithm, distance calculation and the like to serve as a corpus training basis of a business problem analysis model;
74) analyzing the category of the service problems after the algorithm clustering is manually checked, and renaming related categories;
75) continuously training the linguistic data and the configuration rule of the intelligent order identification model, quickly supplementing the content of a newly input work order text, updating and training the key linguistic data in the identification model, and improving the identification redundancy and accuracy of the system model;
76) and configuring the intelligent dispatching model, setting the corresponding relation between the service problem category and the undertaking processing flow, and realizing the automatic dispatching of the work order.
The invention further explains that the construction of the intelligent order-recording correlation algorithm model, the intelligent order-dispatching identification algorithm model and the intelligent order-dispatching model comprises the following steps: the method comprises the steps of classifying a standardized data sample set through data cleaning, establishing a classification model after characteristic selection, enabling a model main body to comprise LightGBM, xgboost and GBDT models, and performing aggregation learning through various algorithms such as an LDA theme algorithm, a k-means algorithm and distance calculation.
The invention further explains that the data model identification matching realized by the intelligent order-recording correlation algorithm model, the intelligent order-dispatching identification algorithm model and the intelligent order-dispatching model utilizes various integrated algorithm models to judge the threshold value of the sample and optimize the parameters, and utilizes methods such as decision trees, combination characteristic analysis and the like to select the model with higher matching degree.
The invention has the following beneficial effects:
1. the invention adopts a combined algorithm to train and recognize the corpus and the rule of the model: the learning training of a large amount of historical precipitated work order data is implemented by utilizing the combined application of various algorithms such as an LDA theme algorithm, a k-means algorithm, distance calculation and the like, an own industry service dictionary, various classified linguistic data and rule data are formed, the classification and prediction of work order service problems submitted by one-line customer service are realized, and compared with the same industry technology, the method has stronger redundancy and error capacity, and misassignment and card order caused by fuzzy keywords and inaccurate judgment due to insufficient manual experience are avoided.
2. The order delivery recognition model system is independently and locally deployed, and the training and the operation are not interfered mutually: the order identification system and the existing customer service system operate independently and provide interface calling as required, continuous training optimization is carried out on the on-line order identification algorithm model, more accurate business classification corpora and classification rules are formed, compared with the existing same-industry technology, the prediction condition is more sufficient, and more reasonable keyword grabbing judgment is achieved in a single mode. Meanwhile, model corpora and rule training continuously optimizes iteration, and the model corpora and rule training does not conflict with intelligent dispatch identification application in operation.
3. The invention uses the association algorithm to realize the extraction and pre-filling of the association data in the information items of the menu page and the synthesis of the abstract: the contents of the work order problem contents and the reply contents are spliced by extracting the customer service conversation content management information and the abstract algorithm, so that the automatic filling of various data in the work order entry page is realized, the first-line seat entry efficiency is improved, the filling of the work order page entry contents is standardized, the work order filling quality is improved, and convenience is provided for subsequent quality inspection.
Drawings
FIG. 1 is an overall flow chart of intelligent dispatching.
FIG. 2 is a flow chart of intelligent dispatch model training.
FIG. 3 is a flow chart of an intelligent transcript correlation algorithm model.
FIG. 4 is a flow chart of an intelligent dispatch identification algorithm model.
Detailed Description
The invention will be further explained with reference to the drawings.
Example 1:
an automatic order recording and intelligent order dispatching method for a customer service system comprises the following steps:
step 1, obtaining dialogue interaction text data;
step 2, extracting and synthesizing the record list associated data through an intelligent record list associated algorithm model;
step 3, automatically filling in work order content items;
step 4, checking the work order content and optimizing and supplementing adjustment;
step 5, submitting and generating a work order;
step 6, inputting the work order into an intelligent order dispatching identification algorithm model, and calculating and outputting work order processing category data;
and 7, inputting the work order processing type data into an intelligent order dispatching model to automatically dispatch the work order.
Example 2:
an automatic order recording and intelligent order dispatching method for a customer service system comprises the following steps:
step 1, obtaining dialogue interaction text data;
the method specifically comprises the following steps: the system acquires interactive text data of a user and a customer service seat in conversation of a work order to be built, accesses the user by voice, acquires real-time voice transcribed text data, and acquires the conversation text data by an online customer service;
step 2, extracting and synthesizing the record list associated data through an intelligent record list associated algorithm model;
the method specifically comprises the following steps: inputting the dialogue text data obtained in the step 1 into an intelligent recording sheet association algorithm model, and extracting and synthesizing recording sheet association data; associated data extraction includes, but is not limited to: information such as customer name, contact telephone, province, card writing channel number, access mode, product, order number and the like, and summary content of work order problem content is spliced by adopting a system summary algorithm;
step 3, automatically filling in work order content items;
the method specifically comprises the following steps: the system automatically selects and fills the list page according to the associated data and the abstract content in the step 2;
step 4, checking the work order content and optimizing and supplementing adjustment;
the method specifically comprises the following steps: the customer service seat checks whether the work order page filling data is accurate or not, and corresponding optimization and supplementary adjustment are carried out;
step 5, after checking, submitting and generating a work order;
step 6, inputting the work order into an intelligent order dispatching identification algorithm model, and calculating and outputting work order processing category data;
the method specifically comprises the following steps: the system acquires work order content information, inputs the work order content information into an intelligent order dispatching identification algorithm model, calculates the identification algorithm model and outputs work order processing category data;
step 7, inputting the work order processing category data into an intelligent order dispatching model to carry out automatic order dispatching;
the method specifically comprises the following steps: the work order processing category data is used as a parameter to be input into an order dispatching engine to automatically dispatch the order to a undertaking department for circulation processing.
The intelligent order recording association algorithm model, the intelligent order delivery identification algorithm model and the intelligent order delivery model are independently and locally deployed, are separately and independently operated with the conventional customer service system, provide interface calling as required, and have no mutual interference between training and operation
Example 3:
step 1, obtaining dialogue interaction text data;
step 2, extracting and synthesizing the record list associated data through an intelligent record list associated algorithm model;
the process of the intelligent recording list association algorithm model comprises the following steps:
21) performing data cleaning on historical work order data, and selecting a sufficient number of standard samples to form a historical sample set;
22) performing work order data by using a natural language processing technology, performing word segmentation on the unstructured text, calculating word frequency and forming word vectors;
23) for the work order sample which forms the word vector and is formatted, carrying out classification training on information such as products, purchase channels, problem description, user appeal and the like by using a machine learning algorithm, and finally obtaining a standardized entry classification template through multiple training and optimization;
24) when a new client consults, voice data is converted into text data, user information and consultation contents are used as parameters and are transmitted into a trained classification model, and key information required by work order entry is extracted;
25) and the customer service checks whether the data generated by the intelligent record list is accurate or not, and performs corresponding optimization and supplementary adjustment.
The natural language processing technology in the step 22) is specifically as follows: judging words in text segment by dictionary generation, word segmentation and text feature extraction, quantitatively counting the occurrence frequency, degree of condensation and degree of freedom to generate special corpus dictionary, segmenting word data by dictionary and general dictionary newly generated by specific corpus, and vectorizing Chinese text
Step 3, automatically filling in work order content items;
step 4, checking the work order content and optimizing and supplementing adjustment;
step 5, submitting and generating a work order;
step 6, inputting the work order into an intelligent order dispatching identification algorithm model, and calculating and outputting work order processing category data;
the flow of the intelligent order dispatching identification algorithm model comprises the following steps:
61) after the customer service submits the work order, the content of the work order is automatically transmitted and input into an intelligent order delivery identification algorithm model, and the assignment identification of the work order problem is matched with the order delivery path flow;
62) the system extracts and submits work order text content as intelligent order delivery recognition algorithm model input;
63) measuring, calculating and classifying the theme and the distance through the trained online intelligent order delivery recognition algorithm model;
64) the intelligent order dispatching recognition algorithm model outputs the service problem category result matched by the work order;
65) triggering an intelligent dispatching model, and dispatching according to the configured work order acceptance flow.
And 7, inputting the work order processing type data into an intelligent order dispatching model to automatically dispatch the work order.
The training process of the intelligent dispatch model in the step 7 comprises the following steps:
71) inputting a large amount of historical work order content texts, extracting text data, performing word segmentation processing on the texts, and the like;
72) performing similarity characteristic analysis on the processed participles, vectorizing an analysis object, realizing synonym recommendation by using methods of 'tfidf', 'counts' and 'hashing', generating a thematic dictionary, and performing dictionary learning and updating supplementation;
73) clustering texts by using a clustering algorithm to generate a feature dictionary, and performing clustering learning on text features by using a plurality of algorithms such as an LDA theme algorithm, a k mean algorithm, distance calculation and the like to serve as a corpus training basis of a business problem analysis model;
74) analyzing the category of the service problems after the algorithm clustering is manually checked, and renaming related categories;
75) continuously training the linguistic data and the configuration rule of the intelligent order identification model, quickly supplementing the content of a newly input work order text, updating and training the key linguistic data in the identification model, and improving the identification redundancy and accuracy of the system model;
76) and configuring the intelligent dispatching model, setting the corresponding relation between the service problem category and the undertaking processing flow, and realizing the automatic dispatching of the work order.
Example 4:
an automatic order recording and intelligent order dispatching method for a customer service system comprises the following steps:
step 1, obtaining dialogue interaction text data;
step 2, extracting and synthesizing the record list associated data through an intelligent record list associated algorithm model;
step 3, automatically filling in work order content items;
step 4, checking the work order content and optimizing and supplementing adjustment;
step 5, submitting and generating a work order;
step 6, inputting the work order into an intelligent order dispatching identification algorithm model, and calculating and outputting work order processing category data;
and 7, inputting the work order processing type data into an intelligent order dispatching model to automatically dispatch the work order.
The intelligent order recording association algorithm model, the intelligent order dispatching identification algorithm model and the intelligent order dispatching model are built by the following steps: the method comprises the steps of classifying a standardized data sample set through data cleaning, establishing a classification model after characteristic selection, enabling a model main body to comprise LightGBM, xgboost and GBDT models, and performing aggregation learning through various algorithms such as an LDA theme algorithm, a k-means algorithm and distance calculation.
The intelligent order-recording correlation algorithm model, the intelligent order-dispatching identification algorithm model and the data model identification matching realized by the intelligent order-dispatching model utilize various integrated algorithm models to judge a threshold value of a sample and optimize parameters, and utilize methods such as decision trees, combination characteristic analysis and the like to select a model with higher matching degree.
The above embodiments are only exemplary embodiments of the present invention, and are not intended to limit the present invention, and the scope of the present invention is defined by the claims. Various modifications and equivalents may be made thereto by those skilled in the art within the spirit and scope of the present invention, and such modifications and equivalents should be considered as falling within the scope of the present invention.
Claims (8)
1. The utility model provides an automatic list of customer service system and intelligence dispatch method which characterized in that includes:
step 1, obtaining dialogue interaction text data;
step 2, extracting and synthesizing the record list associated data through an intelligent record list associated algorithm model;
step 3, automatically filling in work order content items;
step 4, checking the work order content and optimizing and supplementing adjustment;
step 5, submitting and generating a work order;
step 6, inputting the work order into an intelligent order dispatching identification algorithm model, and calculating and outputting work order processing category data;
and 7, inputting the work order processing type data into an intelligent order dispatching model to automatically dispatch the work order.
2. The customer service system automatic order recording and intelligent order dispatching method of claim 1, wherein: the intelligent order recording association algorithm model, the intelligent order dispatching identification algorithm model and the intelligent order dispatching model are independently and locally deployed and independently operate with the existing customer service system, interface calling is provided as required, and training and operation are not interfered with each other.
3. The customer service system automatic order recording and intelligent order dispatching method of claim 2, wherein: the process of the intelligent recording list association algorithm model in the step 2 comprises the following steps:
21) performing data cleaning on historical work order data, and selecting a sufficient number of standard samples to form a historical sample set;
22) performing work order data by using a natural language processing technology, performing word segmentation on the unstructured text, calculating word frequency and forming word vectors;
23) for the work order sample which forms the word vector and is formatted, carrying out classification training on information such as products, purchase channels, problem description, user appeal and the like by using a machine learning algorithm, and finally obtaining a standardized entry classification template through multiple training and optimization;
24) when a new client consults, voice data is converted into text data, user information and consultation contents are used as parameters and are transmitted into a trained classification model, and key information required by work order entry is extracted;
25) and the customer service checks whether the data generated by the intelligent record list is accurate or not, and performs corresponding optimization and supplementary adjustment.
4. The customer service system automatic order recording and intelligent order dispatching method of claim 3, wherein: the natural language processing technology in the step 22) is specifically as follows: the method comprises the steps of judging words in text segments by dictionary generation, word segmentation and text feature extraction methods, carrying out quantitative statistics on the occurrence frequency, the degree of condensation and the degree of freedom to generate a thematic corpus dictionary, carrying out text data word segmentation by using a dictionary and a general dictionary newly generated by a specific corpus, and vectorizing a Chinese text.
5. The customer service system automatic order recording and intelligent order dispatching method of claim 3, wherein: the process of the intelligent dispatch identification algorithm model in the step 6 comprises the following steps:
61) after the customer service submits the work order, the content of the work order is automatically transmitted and input into an intelligent order delivery identification algorithm model, and the assignment identification of the work order problem is matched with the order delivery path flow;
62) the system extracts and submits work order text content as intelligent order delivery recognition algorithm model input;
63) measuring, calculating and classifying the theme and the distance through the trained online intelligent order delivery recognition algorithm model;
64) the intelligent order dispatching recognition algorithm model outputs the service problem category result matched by the work order;
65) triggering an intelligent dispatching model, and dispatching according to the configured work order acceptance flow.
6. The customer service system automatic order recording and intelligent order dispatching method of claim 5, wherein: the training process of the intelligent dispatch model in the step 7 comprises the following steps:
71) inputting a large amount of historical work order content texts, extracting text data, performing word segmentation processing on the texts, and the like;
72) performing similarity characteristic analysis on the processed participles, vectorizing an analysis object, realizing synonym recommendation by using methods of 'tfidf', 'counts' and 'hashing', generating a thematic dictionary, and performing dictionary learning and updating supplementation;
73) clustering texts by using a clustering algorithm to generate a feature dictionary, and performing clustering learning on text features by using a plurality of algorithms such as an LDA theme algorithm, a k mean algorithm, distance calculation and the like to serve as a corpus training basis of a business problem analysis model;
74) analyzing the category of the service problems after the algorithm clustering is manually checked, and renaming related categories;
75) continuously training the linguistic data and the configuration rule of the intelligent order identification model, quickly supplementing the content of a newly input work order text, updating and training the key linguistic data in the identification model, and improving the identification redundancy and accuracy of the system model;
76) and configuring the intelligent dispatching model, setting the corresponding relation between the service problem category and the undertaking processing flow, and realizing the automatic dispatching of the work order.
7. The customer service system automatic order recording and intelligent order dispatching method of claim 6, wherein: the intelligent order recording association algorithm model, the intelligent order dispatching identification algorithm model and the intelligent order dispatching model are built by the following steps: the method comprises the steps of classifying a standardized data sample set through data cleaning, establishing a classification model after characteristic selection, enabling a model main body to comprise LightGBM, xgboost and GBDT models, and performing aggregation learning through various algorithms such as an LDA theme algorithm, a k-means algorithm and distance calculation.
8. The customer service system automatic order recording and intelligent order dispatching method of claim 1, wherein: the intelligent order-recording correlation algorithm model, the intelligent order-dispatching identification algorithm model and the data model identification matching realized by the intelligent order-dispatching model utilize various integrated algorithm models to judge a threshold value of a sample and optimize parameters, and utilize methods such as decision trees, combination characteristic analysis and the like to select a model with higher matching degree.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111609645.1A CN114418327A (en) | 2021-12-27 | 2021-12-27 | Automatic order recording and intelligent order dispatching method for customer service system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111609645.1A CN114418327A (en) | 2021-12-27 | 2021-12-27 | Automatic order recording and intelligent order dispatching method for customer service system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114418327A true CN114418327A (en) | 2022-04-29 |
Family
ID=81268954
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111609645.1A Pending CN114418327A (en) | 2021-12-27 | 2021-12-27 | Automatic order recording and intelligent order dispatching method for customer service system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114418327A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116611453A (en) * | 2023-07-19 | 2023-08-18 | 天津奇立软件技术有限公司 | Intelligent order-distributing and order-following method and system based on big data and storage medium |
| CN116777148A (en) * | 2023-05-31 | 2023-09-19 | 江苏瑞德信息产业有限公司 | Intelligent distribution processing system for service work orders based on data analysis |
| CN117829494A (en) * | 2023-12-27 | 2024-04-05 | 合肥工业大学 | Intelligent service heat line work order identification and distribution platform based on domain knowledge graph |
-
2021
- 2021-12-27 CN CN202111609645.1A patent/CN114418327A/en active Pending
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116777148A (en) * | 2023-05-31 | 2023-09-19 | 江苏瑞德信息产业有限公司 | Intelligent distribution processing system for service work orders based on data analysis |
| CN116777148B (en) * | 2023-05-31 | 2023-12-05 | 江苏瑞德信息产业有限公司 | Intelligent distribution processing system for service work orders based on data analysis |
| CN116611453A (en) * | 2023-07-19 | 2023-08-18 | 天津奇立软件技术有限公司 | Intelligent order-distributing and order-following method and system based on big data and storage medium |
| CN116611453B (en) * | 2023-07-19 | 2023-10-03 | 天津奇立软件技术有限公司 | Intelligent order-distributing and order-following method and system based on big data and storage medium |
| CN117829494A (en) * | 2023-12-27 | 2024-04-05 | 合肥工业大学 | Intelligent service heat line work order identification and distribution platform based on domain knowledge graph |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109189901B (en) | Method for automatically discovering new classification and corresponding corpus in intelligent customer service system | |
| CN114418327A (en) | Automatic order recording and intelligent order dispatching method for customer service system | |
| CN109492109B (en) | Information hotspot mining method and device | |
| CN113468296B (en) | Model self-iteration type intelligent customer service quality inspection system and method capable of configuring business logic | |
| CN107886231B (en) | Service quality evaluation method and system for customer service | |
| CN107833059B (en) | Service quality evaluation method and system for customer service | |
| CN111462752B (en) | Attention mechanism, feature embedding and BI-LSTM (business-to-business) based customer intention recognition method | |
| US20130325560A1 (en) | System, Method and Apparatus for Voice Analytics of Recorded Audio | |
| CN111191051B (en) | Method and system for constructing emergency knowledge map based on Chinese word segmentation technology | |
| CN111858842A (en) | Judicial case screening method based on LDA topic model | |
| CN116628173B (en) | Intelligent customer service information generation system and method based on keyword extraction | |
| CN107766560B (en) | Method and system for evaluating customer service flow | |
| CN113051380A (en) | Information generation method and device, electronic equipment and storage medium | |
| CN112632982A (en) | Dialogue text emotion analysis method capable of being used for supplier evaluation | |
| CN114722191A (en) | Automatic call clustering method and system based on semantic understanding processing | |
| TWI734085B (en) | Dialogue system using intention detection ensemble learning and method thereof | |
| CN111460114A (en) | Retrieval method, device, equipment and computer readable storage medium | |
| CN111274390A (en) | Emotional reason determining method and device based on dialogue data | |
| CN111062211A (en) | Information extraction method and device, electronic equipment and storage medium | |
| CN114239579A (en) | Electric power searchable document extraction method and device based on regular expression and CRF model | |
| CN113037934A (en) | Hot word analysis system based on call recording of call center | |
| CN116600052A (en) | Intelligent customer service assistant intention recognition method and system based on knowledge calculation | |
| CN107886233B (en) | Service quality evaluation method and system for customer service | |
| CN116501844A (en) | Voice keyword retrieval method and system | |
| CN115936011A (en) | Multi-intention semantic recognition method in intelligent conversation |
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 |