CN113705199A - Work order priority confirmation method, work order priority confirmation device, work order priority confirmation electronic equipment, work order priority confirmation medium and work order priority confirmation product - Google Patents
Work order priority confirmation method, work order priority confirmation device, work order priority confirmation electronic equipment, work order priority confirmation medium and work order priority confirmation product Download PDFInfo
- Publication number
- CN113705199A CN113705199A CN202110968794.0A CN202110968794A CN113705199A CN 113705199 A CN113705199 A CN 113705199A CN 202110968794 A CN202110968794 A CN 202110968794A CN 113705199 A CN113705199 A CN 113705199A
- Authority
- CN
- China
- Prior art keywords
- text data
- priority
- work order
- processed
- sample text
- 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
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/205—Parsing
- G06F40/216—Parsing using statistical methods
-
- 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
- G06Q30/016—After-sales
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Marketing (AREA)
- Probability & Statistics with Applications (AREA)
- Finance (AREA)
- General Business, Economics & Management (AREA)
- Economics (AREA)
- Development Economics (AREA)
- Accounting & Taxation (AREA)
- Strategic Management (AREA)
- Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Computational Linguistics (AREA)
- General Health & Medical Sciences (AREA)
- General Engineering & Computer Science (AREA)
- Machine Translation (AREA)
Abstract
The invention provides a work order priority confirmation method, a work order priority confirmation device, electronic equipment, a medium and a product, wherein the method comprises the following steps: acquiring text data to be processed; vectorizing the text data to be processed to obtain a feature vector of the text data to be processed; inputting the feature vector into a pre-trained priority confirmation model, and acquiring probability distribution of the feature vector on a plurality of priority classes; determining the priority of the work order text data according to the probability distribution; the priority confirmation model is obtained by training according to sample text data and priority label data corresponding to the sample text data; and the sample text data is the work order text data with determined priority. According to the invention, the priority of the work order to be processed can be automatically obtained through the pre-trained priority confirmation model, so that the working efficiency of customer service personnel can be improved, and the customer experience is improved.
Description
Technical Field
The present invention relates to the field of data processing technologies, and in particular, to a method, an apparatus, an electronic device, a medium, and a product for determining a priority of a work order.
Background
With the continuous application of artificial intelligence, the demand of people on intelligent analysis and processing of the complaint work orders of customers is increasing day by day, and the data of the complaint work orders are intelligently analyzed and processed, so that the workload of customer service staff can be greatly reduced, and the working efficiency is improved.
At present, in a mainstream customer service work order processing product, the priority of work order processing is often set manually by a customer service person, the customer service person sets the priority according to the industry experience and personal ability of the customer service person and referring to attribute information such as the customer level, the work order type and whether the work order is hastened or not, and the priority confirmation mode is limited by the industry experience and personal ability of the customer service person, so that inaccurate priority confirmation is easy to occur, the problem of important work order feedback is delayed to be solved, and the problem of poor customer experience is solved.
Disclosure of Invention
The invention provides a work order priority confirmation method, a work order priority confirmation device, electronic equipment, a medium and a product, which are used for solving the technical problem of poor customer experience caused by inaccurate work order priority confirmation in the prior art and achieving the purposes of improving the working efficiency of customer service staff and improving the customer experience.
In a first aspect, a method for confirming work order priority includes:
acquiring text data to be processed; the text data to be processed refers to work order text data with priority to be determined;
vectorizing the text data to be processed to obtain a feature vector of the text data to be processed;
inputting the feature vector into a pre-trained priority confirmation model, and acquiring probability distribution of the feature vector on a plurality of priority classes;
determining the priority of the work order text data according to the probability distribution;
the priority confirmation model is obtained by training according to sample text data and priority label data corresponding to the sample text data; and the sample text data is the work order text data with determined priority.
Further, according to the work order priority confirmation method provided by the present invention, after confirming the priority of the work order text data, the method includes:
and automatically generating a service level protocol according to the priority of the work order text data.
Further, according to the work order priority confirmation method provided by the present invention, before the acquiring text data to be processed, the method includes:
acquiring sample text data and priority label data of the sample text data;
preprocessing and word embedding processing are carried out on the sample text data to obtain a feature vector of the sample text data;
training a priority confirmation model based on the feature vectors of the sample text data and the priority label data of the sample text data.
Further, according to the work order priority level confirmation method provided by the present invention, the training of the priority level confirmation model based on the feature vector of the sample text data and the priority level label data of the sample text data includes:
step S1: carrying out priority identification on the feature vectors of the sample text data by using a priority confirmation model to be trained to obtain an identification result of the sample text data;
step S2: judging whether a model training termination condition is met or not according to the recognition result and the priority label data of the sample text data, adjusting the parameters of the model to be trained when the model training termination condition is not met, and re-executing the step S1 by using the adjusted priority confirmation model; and when the model training termination condition is met, obtaining a trained priority confirmation model.
Further, according to the work order priority confirmation method provided by the present invention, the preprocessing the sample text data includes:
and carrying out data shape transformation processing on the sample text data to obtain a matrix of the sample text data.
Further, according to the work order priority confirmation method provided by the present invention, before the vectorization processing is performed on the text data to be processed, the method includes:
and performing complex and simple conversion, text error correction and voice-to-text processing on the text data to be processed.
In a second aspect, the present invention further provides a work order priority confirming device, including:
the acquisition module is used for acquiring text data to be processed; the text data to be processed is work order text data with priority to be determined;
the processing module is used for vectorizing the text data to be processed to obtain a feature vector of the text data to be processed;
the input module is used for inputting the feature vector into a pre-trained priority confirmation model and acquiring probability distribution of the feature vector on a plurality of priority classes;
the confirming module is used for confirming the priority of the work order text data according to the probability distribution;
the priority confirmation model is obtained by training according to the sample text data and the priority label data of the sample text data; and the sample text data is the work order text data with determined priority.
In a third aspect, the present invention also provides an electronic device, including:
a processor, a memory, and a bus, wherein,
the processor and the memory are communicated with each other through the bus;
the memory stores program instructions executable by the processor to invoke steps of the method for work order priority validation as described in any of the above.
In a fourth aspect, the present invention also provides a non-transitory computer readable storage medium storing computer instructions for causing the computer to perform the steps of the work order priority validation method as described above.
In a fifth aspect, the present invention also provides a computer program product comprising a computer program, wherein the computer program is configured to, when executed by a processor, implement the steps of the work order priority validation method as described in any one of the above.
The invention provides a work order priority confirmation method, a work order priority confirmation device, electronic equipment, a work order priority confirmation medium and a work order priority confirmation product, wherein the work order priority confirmation method comprises the following steps: acquiring text data to be processed; vectorizing the text data to be processed to obtain a feature vector of the text data to be processed; inputting the feature vector into a pre-trained priority confirmation model, and acquiring probability distribution of the feature vector on a plurality of priority classes; determining the priority of the work order text data according to the probability distribution; the priority confirmation model is obtained by training according to sample text data and priority label data corresponding to the sample text data; and the sample text data is the work order text data with determined priority. According to the invention, the priority of the work order to be processed can be automatically obtained through the pre-trained priority confirmation model, so that the working efficiency of customer service personnel can be improved, and the customer experience is improved.
Drawings
In order to more clearly illustrate the technical solutions of the present invention or the prior art, the drawings needed for the description of the embodiments or the prior art will be briefly described below, and it is obvious that the drawings in the following description are some embodiments of the present invention, and those skilled in the art can also obtain other drawings according to the drawings without creative efforts.
FIG. 1 is a flow chart of a work order priority validation method according to the present invention;
FIG. 2 is a schematic structural diagram of a work order priority confirmation apparatus provided in the present invention;
fig. 3 is a schematic structural diagram of an electronic device provided in the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention clearer, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings, and it is obvious that the described embodiments are some, but not all embodiments of the present invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Fig. 1 is a schematic flow chart of a work order priority level confirmation method provided by the present invention, and as shown in fig. 1, the work order priority level confirmation method provided by the present invention includes the following steps:
step 101: acquiring text data to be processed; the text data to be processed refers to work order text data with priority to be determined;
step 102: vectorizing the text data to be processed to obtain a feature vector of the text data to be processed;
step 103: inputting the feature vector into a pre-trained priority confirmation model, and acquiring probability distribution of the feature vector on a plurality of priority classes;
step 104: determining the priority of the work order text data according to the probability distribution;
the priority confirmation model is obtained by training according to sample text data and priority label data corresponding to the sample text data; and the sample text data is the work order text data with determined priority.
Specifically, the priority (priority) is a convention that a high priority is made first, and a low priority is made later, and is a parameter for determining the priority level of each operating program for accepting system resources when the computer time-sharing operating system processes a plurality of operating programs.
In step 101, the obtained text data to be processed refers to work order text data whose priority has not been determined, and it should be noted that the text data to be processed includes client information, work order information, history call records, chat record information, and the like, where the history call records are audio information and need to be converted into text information by a speech recognition system.
In step 102, performing vectorization processing on the text data to be processed to obtain a feature vector of the text data to be processed, where the vectorization processing includes multiple modes, Word vector models such as Word2Vec, Glove, BERT, and the like may be used to perform vectorization processing on the text data to be processed, and a Sliding Window (Sliding Window) method may be used to process similar long texts in history call records and chat records included in the text data to be processed. It should be noted that, the user may select different vectorization processing manners according to actual needs, and is not limited in this time.
In step 103, the feature vector obtained in step 102 is input into a pre-trained priority confirmation model, so as to obtain probability distributions on priority classes under the feature vector, where the probability distributions are, for example, 1-level (emergency), 2-level (emergency), 3-level (general), and 4-level (non-emergency), and the probability distributions obtained by the priority confirmation model are: the probability value is 0.5 at level 1, 0.2 at level 2, 0.2 at level 3 and 0.1 at level 4. It should be noted that the classification results of the priority classes are normalized, so that the probability value of each class is scaled between 0 and 1, and the sum of the probability values of the priority classes is ensured to be 1.
In step 104, according to the probability distribution of the obtained feature vector on each category of the priority, the priority category corresponding to the maximum probability value is determined as the priority of the simplex text data. It should be noted that, in this embodiment, the priority class corresponding to the maximum probability value is determined as the priority of the worksheet text data, and the user may select different determination manners according to actual needs, which is not specifically limited herein.
In the embodiment of the invention, the characteristic vector of the text data to be processed is obtained by vectorizing the text data to be processed, then the characteristic vector is input into a pre-trained priority confirmation model to obtain the probability distribution of the characteristic vector on a plurality of priority classes, and the priority of the simplex text data is determined according to the probability distribution. The method provided by the invention can accurately determine the priority of the work order data to be processed, improve the working efficiency of customer service personnel and improve the customer experience.
In another embodiment of the present invention, after the determining the priority of the work order text data, the method includes:
and automatically generating a service level protocol according to the priority of the work order text data.
Specifically, a Service Level Agreement (SLA) refers to an Agreement or contract that is agreed between an enterprise providing a Service and a client and is commonly accepted by both parties with respect to the quality, Level, performance, and the like of the Service.
In the embodiment of the present invention, after the priority of the document data of the worksheet is confirmed, an SLA is automatically generated according to the priority, where the SLA may be: class 1 (very urgent) requires 2 hours to complete, class 2 (urgent) requires 8 hours to complete, class 3 (general) requires 24 hours to complete, class 4 (non-urgent) requires 72 hours to complete. It should be noted that the user may set the setting according to actual needs, and is not limited in particular here.
The service level protocol is automatically generated according to the priority of the work order in the embodiment of the invention, so that the working basis can be provided for customer service personnel, and the working efficiency is improved.
In an embodiment of the present invention, before the acquiring text data to be processed, the method includes:
acquiring sample text data and priority label data of the sample text data;
preprocessing and word embedding processing are carried out on the sample text data to obtain a feature vector of the sample text data;
training a priority confirmation model based on the feature vectors of the sample text data and the priority label data of the sample text data.
Specifically, the priority label data is data obtained by setting a priority label to sample text data.
In the embodiment of the present invention, the sample text data refers to collected historical work order text data, wherein the priority label data refers to label data of which the historical work order text data has determined priority, an SLA rule corresponding to the priority is set according to the obtained data, and through data analysis and statistics, the influence factors of the priority and the SLA of the work order are mainly obtained as follows: the method comprises the steps of customer level, customer industry, customer scale, work order title, problem description, completion duration of the work order, work order handling times, satisfaction degree of the work order, historical call records and chat records of customer ordering through a call center or an online chat mode, converting voice of voice or online chat into text through a voice recognition system (ASR), and storing obtained sample text data into a database, wherein the specific storage mode is shown in the following table 1 and table 2.
Table 1 sample text data
Table 2 sample text data
The sample text data according to the storage format is preprocessed and Word embedded to obtain the feature vector of the sample text data, wherein Word vector models such as Word2Vec, Glove, BERT and the like can be adopted to carry out Word embedded processing on the sample text data, and a Sliding Window (Sliding Window) method can be adopted to process long texts such as historical call records and chat records contained in the sample text data, namely, a document is divided into a plurality of overlapped sections, and then each section is taken as an independent document and sent into the Word vector model to be processed to obtain the feature vector. It should be noted that, the user may select different vectorization processing manners according to actual needs, and is not limited in this time.
On the basis of the processing of the sample text data, the obtained feature vectors and the priority label data of the sample text data form a training set to train a priority confirmation model.
In the embodiment of the invention, a data set is formed according to historical sample text data and label data with determined priority, firstly, the sample text data in a training set is preprocessed and word embedded to obtain the characteristic vector of the sample text data, and then a priority confirmation model is trained on the basis of the characteristic vector of the sample text data and the label data with the priority, so that the accuracy of a model identification result is improved.
In another embodiment of the present invention, the training of the priority confirmation model based on the feature vector of the sample text data and the priority label data of the sample text data includes:
step S1: carrying out priority identification on the feature vectors of the sample text data by using a priority confirmation model to be trained to obtain an identification result of the sample text data;
step S2: judging whether a model training termination condition is met or not according to the recognition result and the priority label data of the sample text data, adjusting the parameters of the model to be trained when the model training termination condition is not met, and re-executing the step S1 by using the adjusted priority confirmation model; and when the model training termination condition is met, obtaining a trained priority confirmation model.
Specifically, the training termination condition is that the training of the model is stopped when the accuracy of the model identification result reaches a preset threshold. The setting can be specifically carried out according to actual needs, and no specific requirements are made here.
In the embodiment of the invention, the feature vector of the sample text data acquired by the embodiment is input into a priority confirmation model to be trained to obtain the identification result of the sample text data, wherein, the priority confirmation model to be trained selects a BERT model, then judges whether the training termination condition of the model is satisfied according to the recognition result and the priority label data of the sample text data, if the identification result obtained by the feature vector of the sample text data is compared with the priority label data for confirmation, the accuracy of the model identification is 70 percent, and the preset threshold value of model termination training is 90%, that is, the training termination condition is not satisfied at this time, some hyper-parameters in the model need to be subjected to fine tuning processing, and the training of the model is stopped until the accuracy of the model meets the requirement of the preset threshold value, so as to obtain the priority confirmation model. When it needs to be described, the priority confirmation model to be trained may select BERT, ELMo, GPT, or the like as the pre-training model, and may be specifically set according to actual needs, which is not specifically limited herein.
In the embodiment of the invention, the priority confirmation model is obtained by the provided model training method, so that the accuracy of the model identification result is ensured.
In one embodiment of the present invention, the preprocessing the sample text data includes:
and carrying out data shape transformation processing on the sample text data to obtain a matrix of the sample text data.
Specifically, the data shape transformation refers to a manner of converting one-dimensional sample text data into a two-dimensional matrix form.
In the embodiment of the invention, the acquired sample text data is one-dimensional data, all the data are connected together and cannot be further processed, the sample text data is subjected to data shape transformation, each row is a data sample (11 value), each data sample comprises 10 influencing factors influencing priority and priority data to form a 2-dimensional matrix, and data shape transformation processing is performed on the sample text data to provide data support for subsequent further processing.
In another embodiment of the present invention, before the vectorizing the text data to be processed, the method includes:
and performing complex and simple conversion, text error correction and voice-to-text processing on the text data to be processed.
Specifically, the simplified and simplified conversion refers to the conversion between simplified and simplified types.
In the embodiment of the invention, before vectorization processing is carried out on the text data to be processed, complex and simple conversion, text error correction and speech-to-text processing are also carried out on the text data to be processed, wherein the speech-to-text processing refers to converting historical call records and chat records of customer invoicing in a call center or online chat mode into text data through a speech recognition system (ASR). It should be noted that, the user may perform corresponding processing on the text to be processed according to actual needs, and is not specifically limited herein.
In another embodiment of the present invention, training the priority validation model comprises the steps of:
collecting historical sample text data of a client submitted by the work order text data, setting the priority of the work order text data into four level types, namely emergency, urgent, general and non-emergency, and obtaining four SLA rules according to the set priority, wherein the four SLA rules are completed within 2 hours (the priority is emergency), completed within 8 hours (the priority is emergency), completed within 24 hours (the priority is general) and completed within 72 hours (the priority is non-emergency); through analysis and statistics of sample text data, the priority and SLA of the work order text data are confirmed to be mainly related to the following influence factors: the method comprises the steps of customer level, customer industry, customer scale, work order title, problem description, work order completion duration, work order handling times, work order satisfaction degree and historical call records and chat records of customer ordering through a call center or an online chat mode.
And preprocessing the stored sample text data. Firstly, importing sample text data of each key factor (customer level, customer industry, customer scale, work order title, problem description, work order completion time, work order handling times, work order satisfaction degree, and historical call records and chat records of customer handling orders through a call center or an online chat mode), wherein the sample text data is 1-dimensional, all data are connected together and cannot be further processed, data shape transformation needs to be carried out on the sample text data, each row of data is a data sample (11 values), each data sample comprises 10 key factors influencing priority and priority label data, and a 2-dimensional matrix is formed.
And then carrying out Word embedding processing on the preprocessed sample text data, wherein the Word embedding processing method comprises but is not limited to Word2Vec, Glove, BERT and the like, wherein for long texts such as call records and chat records, a Sliding Window (Sliding Window) method can be adopted to divide the document into a plurality of overlapped sections, and then each section is used as an independent document and is sent into a Word embedding model to be processed, so as to obtain the feature vector of the sample text data. And forming a data set by the feature vectors of the plurality of sample text data and the corresponding priority label data.
And dividing the data set obtained in the last step into a training set and a test set, wherein the training set is used for determining parameters of the priority confirmation model, and the test set is used for judging the effect of the priority confirmation model. Based on a general division standard, 80% of data in a data set is used as a training set, 20% of data is used as a test set, a priority confirmation model to be trained is trained by utilizing the data in the training set, wherein the priority confirmation model can select BERT, ELMo, GPT and the like as pre-training models. It should be noted that the classification result needs to be normalized, so that the probability value of each feature is scaled to 0-1, the output result of the model is directly used as the probability distribution of the sample text data on multiple priority classes, and the probability sum is 1.
Fig. 2 is a diagram illustrating a work order priority confirmation apparatus according to the present invention, and as shown in fig. 2, the work order priority confirmation apparatus according to the present invention includes:
an obtaining module 201, configured to obtain text data to be processed; the text data to be processed is work order text data with priority to be determined;
the processing module 202 is configured to perform vectorization processing on the text data to be processed to obtain a feature vector of the text data to be processed;
an input module 203, configured to input the feature vector into a pre-trained priority confirmation model, and obtain probability distribution of the feature vector on multiple priority categories;
a confirming module 204, configured to confirm the priority of the work order text data according to the probability distribution;
the priority confirmation model is obtained by training according to the sample text data and the priority label data of the sample text data; and the sample text data is the work order text data with determined priority.
The work order priority confirming device provided by the invention can improve the working efficiency of customer service personnel and improve the customer experience.
Further, the apparatus further comprises a generating module configured to:
and automatically generating a service level protocol according to the priority of the work order text data.
Further, the apparatus further comprises a training module configured to:
acquiring sample text data and priority label data of the sample text data;
preprocessing and word embedding processing are carried out on the sample text data to obtain a feature vector of the sample text data;
training a priority confirmation model based on the feature vectors of the sample text data and the priority label data of the sample text data.
Further, the training module comprises an identification unit and a judgment unit, wherein,
the identification unit is used for carrying out priority identification on the feature vectors of the sample text data by utilizing a priority confirmation model to be trained to obtain an identification result of the sample text data;
the judging unit is used for judging whether a model training termination condition is met or not according to the recognition result and the priority label data of the sample text data, adjusting the parameters of the model to be trained when the model training termination condition is not met, and confirming the model to be re-executed by using the adjusted priority; and when the model training termination condition is met, obtaining a trained priority confirmation model.
Further, the processing module is further configured to:
and carrying out data shape transformation processing on the sample text data to obtain a matrix of the sample text data.
Further, the apparatus further comprises a preprocessing module configured to:
and performing complex and simple conversion, text error correction and voice-to-text processing on the text data to be processed.
Since the principle of the apparatus according to the embodiment of the present invention is the same as that of the method according to the above embodiment, further details are not described herein for further explanation.
Fig. 3 is a schematic structural diagram of an electronic device according to an embodiment of the present invention, and as shown in fig. 3, the present invention provides an electronic device, including: a processor (processor)301, a memory (memory)302, and a bus 303;
wherein, the processor 301 and the memory 302 complete the communication with each other through the bus 303;
The present embodiments provide a non-transitory computer-readable storage medium storing computer instructions that cause the computer to perform the methods provided by the above method embodiments, for example, including: acquiring text data to be processed; the text data to be processed refers to work order text data with priority to be determined; inputting the text data to be processed into a word vector model to obtain a feature vector of the text data to be processed; inputting the feature vector into a pre-trained priority confirmation model, and acquiring probability distribution of the feature vector on a plurality of priority classes; determining the priority of the work order text data according to the probability distribution; the priority confirmation model is obtained by training according to sample text data and priority label data corresponding to the sample text data; and the sample text data is the work order text data with determined priority.
The present invention also provides a computer program product comprising a computer program stored on a non-transitory computer readable storage medium, the computer program comprising program instructions which, when executed by a computer, enable the computer to perform a method of work order priority validation as provided by the above methods, the method comprising: acquiring text data to be processed; the text data to be processed refers to work order text data with priority to be determined; inputting the text data to be processed into a word vector model to obtain a feature vector of the text data to be processed; inputting the feature vector into a pre-trained priority confirmation model, and acquiring probability distribution of the feature vector on a plurality of priority classes; determining the priority of the work order text data according to the probability distribution; the priority confirmation model is obtained by training according to sample text data and priority label data corresponding to the sample text data; and the sample text data is the work order text data with determined priority.
Those of ordinary skill in the art will understand that: all or part of the steps for implementing the method embodiments may be implemented by hardware related to program instructions, and the program may be stored in a computer readable storage medium, and when executed, the program performs the steps including the method embodiments; and the aforementioned storage medium includes: various media that can store program codes, such as ROM, RAM, magnetic or optical disks.
Finally, it should be noted that: the above examples are only intended to illustrate the technical solution of the present invention, but not to limit it; although the present 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 solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention.
Claims (10)
1. A method for work order priority validation, comprising:
acquiring text data to be processed; the text data to be processed refers to work order text data with priority to be determined;
vectorizing the text data to be processed to obtain a feature vector of the text data to be processed;
inputting the feature vector into a pre-trained priority confirmation model, and acquiring probability distribution of the feature vector on a plurality of priority classes;
determining the priority of the work order text data according to the probability distribution;
the priority confirmation model is obtained by training according to sample text data and priority label data corresponding to the sample text data; and the sample text data is the work order text data with determined priority.
2. The method of claim 1, wherein the determining the priority of the work order text data comprises:
and automatically generating a service level protocol according to the priority of the work order text data.
3. The method for confirming work order priority as claimed in claim 1, wherein before the obtaining text data to be processed, the method comprises:
acquiring sample text data and priority label data of the sample text data;
preprocessing and word embedding processing are carried out on the sample text data to obtain a feature vector of the sample text data;
training a priority confirmation model based on the feature vectors of the sample text data and the priority label data of the sample text data.
4. The method of claim 3, wherein training a priority validation model based on the feature vector of the sample text data and the priority label data of the sample text data comprises:
step S1: carrying out priority identification on the feature vectors of the sample text data by using a priority confirmation model to be trained to obtain an identification result of the sample text data;
step S2: judging whether a model training termination condition is met or not according to the recognition result and the priority label data of the sample text data, adjusting the parameters of the model to be trained when the model training termination condition is not met, and re-executing the step S1 by using the adjusted priority confirmation model; and when the model training termination condition is met, obtaining a trained priority confirmation model.
5. The work order priority validation method of claim 3, wherein the pre-processing the sample text data comprises:
and carrying out data shape transformation processing on the sample text data to obtain a matrix of the sample text data.
6. The work order priority validation method of claim 1, wherein prior to the entering the text data to be processed into a word vector model, comprising:
and performing complex and simple conversion, text error correction and voice-to-text processing on the text data to be processed.
7. A work order priority confirmation apparatus, comprising:
the acquisition module is used for acquiring text data to be processed; the text data to be processed is work order text data with priority to be determined;
the processing module is used for vectorizing the text data to be processed to obtain a feature vector of the text data to be processed;
the input module is used for inputting the feature vector into a pre-trained priority confirmation model and acquiring probability distribution of the feature vector on a plurality of priority classes;
the confirming module is used for confirming the priority of the work order text data according to the probability distribution;
the priority confirmation model is obtained by training according to the sample text data and the priority label data of the sample text data; and the sample text data is the work order text data with determined priority.
8. An electronic device, comprising: a processor, a memory, and a bus, wherein,
the processor and the memory are communicated with each other through the bus;
the memory stores program instructions executable by the processor to invoke steps of a method of work order priority validation as claimed in any one of claims 1 to 6.
9. A non-transitory computer readable storage medium storing computer instructions for causing a computer to perform the steps of the work order priority validation method of any of claims 1-6.
10. A computer program product comprising a computer program, wherein the computer program when executed by a processor implements the steps of the work order priority validation method of any of claims 1 to 6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110968794.0A CN113705199A (en) | 2021-08-23 | 2021-08-23 | Work order priority confirmation method, work order priority confirmation device, work order priority confirmation electronic equipment, work order priority confirmation medium and work order priority confirmation product |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110968794.0A CN113705199A (en) | 2021-08-23 | 2021-08-23 | Work order priority confirmation method, work order priority confirmation device, work order priority confirmation electronic equipment, work order priority confirmation medium and work order priority confirmation product |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113705199A true CN113705199A (en) | 2021-11-26 |
Family
ID=78654099
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110968794.0A Pending CN113705199A (en) | 2021-08-23 | 2021-08-23 | Work order priority confirmation method, work order priority confirmation device, work order priority confirmation electronic equipment, work order priority confirmation medium and work order priority confirmation product |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113705199A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114372700A (en) * | 2022-01-07 | 2022-04-19 | 京东科技信息技术有限公司 | Data sampling detection method and device |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101038640A (en) * | 2007-04-09 | 2007-09-19 | 华为技术有限公司 | Worksheets priority level administration method and system, and worksheets process method and system |
| DE102007054674A1 (en) * | 2006-11-17 | 2008-05-21 | Avaya Technology Llc | Work knowledge algorithm for contact center agents |
| CN107094208A (en) * | 2016-09-28 | 2017-08-25 | 北京小度信息科技有限公司 | Worksheet method and device |
| CN108416375A (en) * | 2018-02-13 | 2018-08-17 | 中国联合网络通信集团有限公司 | Work order sorting technique and device |
| CN111325405A (en) * | 2020-03-02 | 2020-06-23 | 贵州电网有限责任公司 | Complaint sensitivity modeling prediction method |
| CN111667108A (en) * | 2020-05-29 | 2020-09-15 | 国网河北省电力有限公司电力科学研究院 | Distribution network first-aid repair time length prediction method |
-
2021
- 2021-08-23 CN CN202110968794.0A patent/CN113705199A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| DE102007054674A1 (en) * | 2006-11-17 | 2008-05-21 | Avaya Technology Llc | Work knowledge algorithm for contact center agents |
| CN101038640A (en) * | 2007-04-09 | 2007-09-19 | 华为技术有限公司 | Worksheets priority level administration method and system, and worksheets process method and system |
| CN107094208A (en) * | 2016-09-28 | 2017-08-25 | 北京小度信息科技有限公司 | Worksheet method and device |
| CN108416375A (en) * | 2018-02-13 | 2018-08-17 | 中国联合网络通信集团有限公司 | Work order sorting technique and device |
| CN111325405A (en) * | 2020-03-02 | 2020-06-23 | 贵州电网有限责任公司 | Complaint sensitivity modeling prediction method |
| CN111667108A (en) * | 2020-05-29 | 2020-09-15 | 国网河北省电力有限公司电力科学研究院 | Distribution network first-aid repair time length prediction method |
Non-Patent Citations (1)
| Title |
|---|
| 任莹: "基于预训练BERT模型的客服工单自动分类研究", 云南电力技术, vol. 48, no. 01, pages 2 - 7 * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114372700A (en) * | 2022-01-07 | 2022-04-19 | 京东科技信息技术有限公司 | Data sampling detection method and device |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109522556B (en) | Intention recognition method and device | |
| US11804216B2 (en) | Generating training datasets for a supervised learning topic model from outputs of a discovery topic model | |
| CN110890088B (en) | Voice information feedback method and device, computer equipment and storage medium | |
| CN110955770A (en) | Intelligent dialogue system | |
| CN111429157A (en) | Method, device and equipment for evaluating and processing complaint work order and storage medium | |
| CN113705199A (en) | Work order priority confirmation method, work order priority confirmation device, work order priority confirmation electronic equipment, work order priority confirmation medium and work order priority confirmation product | |
| CN117688165B (en) | Multi-edge collaborative customer service method, device, equipment and readable storage medium | |
| CN116308219B (en) | Generated RPA flow recommendation method and system based on Tranformer | |
| CN112860873A (en) | Intelligent response method, device and storage medium | |
| CN117116251A (en) | Repayment probability assessment method and device based on collection-accelerating record | |
| CN111353728A (en) | Risk analysis method and system | |
| CN116127011A (en) | Intention recognition method, device, electronic equipment and storage medium | |
| CN112929502B (en) | Voice recognition method and system based on electric power customer service | |
| CN112131369B (en) | Service class determining method and device | |
| CN115080732A (en) | Complaint work order processing method and device, electronic equipment and storage medium | |
| CN114254088A (en) | Method for constructing automatic response model and automatic response method | |
| CN113992806A (en) | Intelligent voice RPA robot outbound method and device | |
| CN115906844B (en) | Rule template-based information extraction method and system | |
| KR102507810B1 (en) | Voice-based sales information extraction and lead recommendation method using artificial intelligence, and data analysis apparatus therefor | |
| CN112714220B (en) | Business processing method and device, computing equipment and computer readable storage medium | |
| CN118626057A (en) | Modularized development method and system based on large language model | |
| CN116308735A (en) | Financial data prediction method, device, electronic equipment and storage medium | |
| CN117874225A (en) | Customer complaint work order intention recognition method and system | |
| CN115238057A (en) | Method and device for responding to sudden aggregated consultation | |
| CN117995199A (en) | Loan contract voice signing method, device 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 |