Work order root cause positioning method and device, equipment and storage medium
By preprocessing and classifying work orders and matching the root cause positioning process, the problem of low positioning efficiency of work orders in the telecommunications industry is solved, and automated and efficient root cause positioning is achieved.
Patent Information
- Application Number
- CN202311723928.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2023-12-14
- Publication Date
- 2025-06-17
AI Technical Summary
The work orders in the telecom industry are low in positioning efficiency, rely on manual processing and long process time, resulting in low processing efficiency.
By obtaining work orders and preprocessing them, a work order feature vector is generated and inputted into the classification model for classification. If work orders are abnormal, the corresponding root cause positioning process will be matched to generate the cause of the exception.
It improves the efficiency of working order root cause positioning, reduces manual misjudgment, and realizes an automated root cause positioning process.
Smart Images
Figure CN120163308A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of artificial intelligence, and particularly to a method, device, equipment, and storage medium for locating the root cause of work orders. Background Art
[0002] Currently, the work order types in the telecommunications industry include more than a dozen major categories such as billing, account management, credit control, and business operations, as well as hundreds of minor categories. In order to better understand the problems encountered by users and take effective measures to solve and prevent the occurrence of similar problems, a large amount of work order data is analyzed through a work order location method to find the causes of faults and problems.
[0003] However, due to various problems in the telecommunications industry such as complex business rules, a large amount of redundant historical operations, decentralized system construction, and inconsistent data, when staff perform work order location, they need to judge the work order type based on their own business experience, then query information in the corresponding database according to the work order type, and finally, after manually synthesizing and calculating various basic data, they can find the root cause of the problem. The method of manually locating the root cause of work orders highly depends on the work experience of the processing personnel, and the manual processing process takes a long time, resulting in low processing efficiency.
[0004] Therefore, there is an urgent need for a method to improve the efficiency of locating the root cause of work orders. Summary of the Invention
[0005] This application provides a method, device, equipment, and storage medium for locating the root cause of work orders to solve the technical problem of low efficiency in locating the root cause of work orders.
[0006] In a first aspect, this application provides a method for locating the root cause of work orders, including:
[0007] Obtain a work order and process the work order to obtain a work order feature vector;
[0008] Input the work order feature vector into a classification model to obtain a work order classification result output by the classification model; wherein, the classification model includes multiple major category models, and each major category model is correspondingly provided with multiple minor category models, and the work order classification result is used to indicate whether a minor category in the major category to which the work order belongs is abnormal;
[0009] If the work order classification result is that the work order is abnormal, match a corresponding root cause location process according to the work order classification result, and the root cause location process is used to generate the cause of the abnormality; wherein, each minor category in each major category corresponds to its own root cause location judgment logic;
[0010] Obtain a root cause location result according to the root cause location process, and the root cause location result is the cause of the work order.
[0011] In a possible design, obtaining the work order and processing the work order to obtain a work order feature vector includes:
[0012] Preprocessing the work order, where the preprocessing includes at least one of invalid work order elimination, sensitive word replacement, stop word replacement, or symbol clearing;
[0013] Performing word segmentation on the preprocessed work order to obtain a word segmentation result;
[0014] Performing vectorization processing on the word segmentation result to obtain a work order feature vector.
[0015] In a possible design, inputting the work order feature vector into a classification model to obtain a work order classification result output by the classification model includes:
[0016] Inputting the work order feature vector into each major category model, obtaining a first classification probability output by each major category model, and determining the target major category model to which the work order belongs according to the first classification probability; wherein, the major category model is a classification model;
[0017] Inputting the work order into multiple sub-category models corresponding to the target major category model, and obtaining a second classification probability output by each sub-category model; wherein, the sub-category model is a classification model;
[0018] Obtaining the work order classification result output by the classification model according to the target major category model and the second classification probabilities output by each sub-category model.
[0019] In a possible design, obtaining a root cause location result according to the root cause location process includes:
[0020] Obtaining a target data source according to the work order classification result; wherein, different classifications correspond to different data sources;
[0021] Obtaining the business information of the user in the work order from the target data source;
[0022] Obtaining a root cause location result according to the business information and the root cause location process.
[0023] In a possible design, obtaining a root cause location result according to the business information and the root cause location process includes:
[0024] Obtaining the data types and judgment logics required by multiple nodes sorted in time sequence in the root cause location process;
[0025] Obtaining corresponding data from the business information according to the data type required by each node;
[0026] Process the data according to the judgment logic corresponding to the node, and obtain the root cause location information of each node;
[0027] According to the root cause location information of each node, obtain the abnormal root cause node, and obtain the root cause location result according to the abnormal root cause node.
[0028] In a possible design, before matching the corresponding root cause location process according to the work order classification result, the method further includes:
[0029] In response to a configuration request from the user terminal, feedback a judgment logic configuration interface to the user terminal, where the judgment logic configuration interface is used to input judgment information, and the judgment information includes a judgment name, a judgment type, available attributes, and judgment conditions;
[0030] Generate the judgment logic of each node according to the judgment information fed back by the user terminal;
[0031] Generate the root cause location process according to the judgment type and the judgment logic of each node; wherein, the judgment type is used to indicate whether the judgment logic is a starting judgment or a process judgment.
[0032] In a possible design, after obtaining the root cause location result according to the abnormal root cause node, the method further includes:
[0033] Obtain the solution corresponding to the abnormal root cause node;
[0034] Feedback abnormal information to the user terminal, where the abnormal information includes the abnormal root cause node and the corresponding solution.
[0035] In a second aspect, the present application provides a work order root cause location device, including:
[0036] An acquisition module, configured to acquire a work order and process the work order to obtain a work order feature vector;
[0037] A classification module, configured to input the work order feature vector into a classification model to obtain a work order classification result output by the classification model; wherein, the classification model includes multiple major category models, and each major category model is correspondingly provided with multiple minor category models, and the work order classification result is used to indicate whether the minor category in the major category to which the work order belongs is abnormal;
[0038] If the work order classification result is that the work order is abnormal, match the corresponding root cause location process according to the work order classification result, where the root cause location process is used to generate the cause of the abnormality; wherein, each minor category in each major category corresponds to its own root cause location judgment logic;
[0039] An execution module, configured to obtain a root cause location result according to the root cause location process, where the root cause location result is the cause of the work order.
[0040] In a third aspect, the present application provides a work order root cause location device, including: a processor, and a memory communicatively connected to the processor;
[0041] The memory stores computer execution instructions;
[0042] The processor executes the computer execution instructions stored in the memory, so that the at least one processor executes the work order root cause location method described in the first aspect and various possible designs of the first aspect as above.
[0043] In a fourth aspect, the present application provides a computer-readable storage medium, in which computer execution instructions are stored. When the processor executes the computer execution instructions, the work order root cause location method described in the first aspect and various possible designs of the first aspect as above is implemented.
[0044] The work order root cause location method, device, equipment and storage medium provided by the present application classify the work orders to obtain a classification result, indicate whether there is an abnormality in the sub-category in the major category to which the work order belongs according to the classification result, and match the corresponding root cause location process according to the work order classification result, so as to be able to obtain the cause of the abnormal work order, avoid misjudgment of the root cause caused by manual work order root cause location, and thus improve the work order root cause location efficiency. BRIEF DESCRIPTION OF THE DRAWINGS
[0045] The drawings herein are incorporated into the specification and constitute a part of this specification, showing embodiments consistent with the present application, and are used together with the specification to explain the principles of the present application.
[0046] Figure 1 It is the flowchart of the work order root cause location method provided by the embodiment of the present application Figure 1 ;
[0047] Figure 2 It is the flowchart of the work order root cause location method provided by the embodiment of the present application Figure 2 ;
[0048] Figure 3 It is the flowchart of the work order root cause location method provided by the embodiment of the present application Figure 3 ;
[0049] Figure 4 It is the structural schematic diagram of the work order root cause location device provided by the embodiment of the present application Figure 1 ;
[0050] Figure 5 It is the hardware schematic diagram of the work order root cause location device provided by the embodiment of the present application.
[0051] Through the above-mentioned drawings, specific embodiments of the present application have been shown, and there will be a more detailed description hereinafter. These drawings and written descriptions are not intended to limit the scope of the concept of the present application in any way, but to illustrate the concept of the present application to those skilled in the art by referring to specific embodiments. Detailed Description of Specific Embodiments
[0052] Here, exemplary embodiments will be described in detail, and examples are shown in the drawings. When the following description refers to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present application. On the contrary, they are merely examples of devices and methods consistent with some aspects of the present application as detailed in the appended claims.
[0053] In the prior art, when performing root cause location of telecom service work orders, staff need to judge the work order type based on their own business experience. However, there are problems in the telecom service such as complex business rules, a large amount of redundant historical services, decentralized system construction, and inconsistent data. As a result, staff need to query corresponding materials in numerous databases during the process of root cause location of work orders, and need to manually synthesize and calculate various basic data to find the root cause of abnormal work orders. This method has problems of long processing time and low processing efficiency.
[0054] In order to effectively improve the efficiency of root cause location of work orders, the present invention designs a method for root cause location of work orders. The root cause location system of work orders obtains work orders and processes the work orders to obtain work order feature vectors, and inputs the work order feature vectors into a classification model to obtain the work order classification result output by the classification model. Among them, the classification model includes multiple major category models, and each major category model is correspondingly provided with multiple minor category models. The work order classification result is used to indicate whether the minor category in the major category to which the work order belongs is abnormal. If the work order classification result indicates that the work order is abnormal, a corresponding root cause location process is matched according to the work order classification result. The root cause location process is used to generate the cause of the abnormality. Among them, each minor category in each major category corresponds to its own root cause location judgment logic, and the root cause location result is obtained according to the root cause location process. The root cause location result is the cause of the work order.
[0055] The method for root cause location of work orders provided by the present application aims to solve the above technical problems in the prior art.
[0056] The following uses specific embodiments to detail the technical solution of the present application and how the technical solution of the present application solves the above technical problems. The following several specific embodiments can be combined with each other, and the same or similar concepts or processes may not be repeated in some embodiments. The embodiments of the present application will be described below in conjunction with the drawings.
[0057] Figure 1 The process flow of the work order root cause location method provided by the embodiments of this application Figure 1 As Figure 1 shown, the method includes:
[0058] S101. Obtain a work order and process the work order to obtain a work order feature vector.
[0059] Specifically, the work order root cause location system can obtain a work order from a work order system or a relevant database. The work order includes work order data, and the work order data can contain various attributes of the work order, such as work order number, work order creation time, work order processing status, work order description, etc. The processing of the work order is that the work order root cause location system performs invalid work order elimination, sensitive word replacement, stop word replacement, or symbol clearing on the work order. The work order root cause location system performs word segmentation processing on the processed work order to obtain a word segmentation result, and then performs vectorization processing on the word segmentation result to obtain a work order feature vector.
[0060] In a possible design, when performing invalid work order elimination, the root cause location system first defines the criteria for invalid work orders, determines invalid work orders according to the status and creation time of the work order, such as identifying expired work orders and overdue work orders as invalid work orders. Screen the work orders according to the above criteria, and remove the invalid work orders from the work order set. Perform sensitive word replacement on the work order set after removing the invalid work orders. Retrieve each work order in the work order set according to the sensitive word list, check whether there are sensitive words in each work order, and perform sensitive word replacement on the work orders with sensitive words, such as replacing the sensitive words with specific placeholders or other appropriate words. Perform stop word replacement on the work order set after replacing the sensitive words. Retrieve each work order in the work order set according to the stop word list, check whether there are stop words in each work order, and perform stop word replacement on the work orders with stop words, such as replacing the stop words with specific placeholders or other appropriate words, or directly deleting the stop words. Perform symbol clearing on the work order set after replacing the stop words. The symbols can be punctuation marks, special characters, or other non-alphabetic characters, and regular expressions or string operations are used for symbol clearing.
[0061] In the specific implementation process, the work order root cause location system obtains a work order and processes the work order to obtain a work order feature vector.
[0062] S102. Input the work order feature vector into a classification model to obtain a work order classification result output by the classification model; wherein, the classification model includes multiple major category models, and each major category model is correspondingly provided with multiple minor category models, and the work order classification result is used to indicate whether the minor category in the major category to which the work order belongs is abnormal.
[0063] Specifically, the classification model is trained through historical work order information. Among them, the classification model has a major category model and a minor category model corresponding to the major category model. The training process of the classification model is as follows: first, train the major category model, and then train the models under each major category separately to obtain multiple minor category models. The minor category models can distinguish the minor category business scenarios under the corresponding major category business scenarios, and thus can classify work orders more accurately.
[0064] The classification model includes multiple major category models, such as a billing classification model, an account management classification model, a credit control classification model, a business classification model, etc. Each major category model corresponds to multiple minor category models. For example, under the credit control classification model, there are corresponding models such as a model for non-booting after payment, a model for payment failure, and a model for abnormal deduction. Each major category model corresponds to multiple minor category models, and among them, the multiple minor category models can include normal business models and abnormal business models. If a work order is classified into a normal business model, for example, the work order is classified into a call fee query model, then the above work order is considered a normal work order, and the call fee query model outputs the work order classification result, and this classification result indicates that the minor category model to which the above work order belongs is a normal business model. If a work order is classified into an abnormal business model, for example, the work order is classified into a model for complaining about the service attitude of payment, then the above work order is considered an abnormal work order, and the model for complaining about the service attitude of payment outputs the work order classification result, and this classification result indicates that the minor category model to which the above work order belongs is an abnormal business model.
[0065] In a possible design, the work order classification model can be trained through the following steps:
[0066] First, obtain historical work orders from the work order system or relevant databases, obtain the full amount of work order information based on the historical work orders, sort out the full amount of work order information according to the required format of the system, and store the sorted full amount of work order information in the system for subsequent model training processes.
[0067] Secondly, preprocess the full amount of work order information. The preprocessing includes eliminating invalid work orders, replacing sensitive words, replacing stop words, or clearing symbols.
[0068] Thirdly, perform word segmentation on the preprocessed full amount of work order information. For example, use pkuseg to perform word segmentation on the full amount of work order information. Pkuseg splits the input full amount of work order information into words to form a word sequence. Then vectorize the obtained word segmentation results, convert the segmented words into word vector representations, so that the neural network can accept them as inputs. The word segmentation vectorization can be obtained using the Glove word vector model.
[0069] Then, manually classify and label the word segmentation vectors, classify and label the historical work orders to be included in the training, and obtain word segmentation vectors representing different business scenarios.
[0070] Furthermore, a supervised neural network model is selected, and the tokenized vectors processed above are divided into a test set and a training set according to a ratio, and then imported into the model for training.
[0071] Finally, the classification model after model training and parameter tuning is exported and saved for subsequent use in the work order root cause location process.
[0072] In the specific implementation process, the work order root cause location system inputs the work order feature vector into the classification model to obtain the major model to which the work order feature vector belongs, then classifies the work order feature vector according to the minor models under the major model, and obtains the work order classification result according to the minor model to which the work order feature vector belongs. Among them, the work order classification result is used to indicate whether the minor class in the major class to which the work order belongs is abnormal.
[0073] S103. If the work order classification result indicates that the work order is abnormal, then match the corresponding root cause location process according to the work order classification result. The root cause location process is used to generate the cause of the abnormality; among them, each minor class in the major class corresponds to its own root cause location judgment logic.
[0074] Specifically, the root cause location process is composed of multiple nodes, each node has a corresponding judgment logic, and the judgment logic in the node can process data and output the root cause location information of each node.
[0075] In a possible design, the root cause location process can be generated through the following steps:
[0076] First, in response to a configuration request from the user terminal, the work order root cause location system feeds back a judgment logic configuration interface to the user terminal. The judgment logic configuration interface is used to input judgment information, and the judgment information includes judgment name, judgment type, available attributes, and judgment conditions. In a possible design, the judgment name can be whether the work order label is consistent with the business label, so as to confirm whether the minor model to which the work order belongs is the business scenario to which it belongs, thereby ensuring the accuracy of the work order root cause location. The judgment type includes start judgment and process judgment. The start judgment is the starting node of the work order root cause location process, and the process judgment is the intermediate node of the work order root cause location process. The available attributes include serial Number, acctld, userld, provinceCode, domain, frame Type, logic Phone, etc. The judgment condition is a relational expression between the available attributes of the work order. For example, ‘#{logic Phone}’ = ‘#{serial Number}’.
[0077] Second, the work order root cause location system receives the above judgment information of multiple nodes configured by the user terminal and generates the judgment logic of each node.
[0078] Finally, according to the judgment type and the judgment logic of each node, a root cause location process is generated; among them, the judgment type is used to indicate whether the judgment logic is a starting judgment or a process judgment.
[0079] In the specific implementation process, if the work order classification result is a work order anomaly, the work order root cause location system matches the corresponding root cause location process according to the work order classification result, and the root cause location process is used to generate the cause of the anomaly; among them, the sub-categories in each major category correspond to their respective root cause location judgment logics.
[0080] S104. Obtain a root cause location result according to the root cause location process, and the root cause location result is the cause of the work order.
[0081] The root cause location result is the judgment result of the node when the node logic judgment is abnormal in the work order root cause location process. The work order root cause location system can obtain the cause of the abnormal work order according to the judgment result output when the node logic judgment is abnormal.
[0082] In the specific implementation process, the work order root cause location system obtains a root cause location result according to the root cause location process, and the root cause location result is the cause of the work order.
[0083] The work order root cause location method provided in this embodiment classifies the work order to obtain a classification result, indicates whether the sub-category in the major category to which the work order belongs is abnormal according to the classification result, and matches the corresponding root cause location process according to the work order classification result, so as to be able to obtain the cause of the abnormal work order, avoid misjudgment of the root cause caused by manual work order root cause location, and thus improve the work order root cause location efficiency.
[0084] Figure 2 It is the work order root cause location method flow provided by the embodiment of the present application Figure 2 . This embodiment elaborates on the work order root cause location method in detail on the basis of the above Figure 1 embodiment. As Figure 2 shown, the method includes:
[0085] S201. Preprocess the work order, and the preprocessing includes at least one of invalid work order elimination, sensitive word replacement, stop word replacement, or symbol clearing.
[0086] S202. Perform word segmentation processing on the preprocessed work order to obtain a word segmentation result.
[0087] S203. Perform vectorization processing on the word segmentation result to obtain a work order feature vector.
[0088] S201 - S203 is similar to Figure 1 the steps S101 - S102 in the
[0089] S204. Input the work order feature vector into each major category model, obtain the first classification probability output by each major category model, and determine the target major category model to which the work order belongs according to the first classification probability; wherein, the major category model is a classification model.
[0090] Specifically, input the work order feature vector into each major category model. Each major category model uses the Softmax function to obtain the prediction result distribution probability of the work order feature vector. The probability represents the possibility that the work order belongs to each major category model. If a certain prediction result distribution probability exceeds the preset threshold, it is considered that the work order belongs to the major category model represented by the prediction result distribution probability. In a possible design, the preset threshold is 80%.
[0091] S205. Input the work order into multiple sub-category models corresponding to the target major category model, and obtain the second classification probability output by each sub-category model; wherein, the sub-category model is a classification model.
[0092] Specifically, input the work order feature vector into each sub-category model in the major category model to which it belongs. Each sub-category model uses the Softmax function to obtain the prediction result distribution probability of the work order feature vector. The probability represents the possibility that the work order belongs to each sub-category model. If a certain prediction result distribution probability exceeds the preset threshold, it is considered that the work order belongs to the sub-category model represented by the prediction result distribution probability. In a possible design, the preset threshold is 80%.
[0093] S206. Obtain the work order classification result output by the classification model according to the target major category model and the second classification probability output by each sub-category model.
[0094] In the specific implementation process, the work order root cause location system inputs the work order feature vector into each major category model, obtains the first classification probability output by each major category model, determines the target major category model to which the work order belongs according to the first classification probability, and then inputs the work order feature vector into multiple sub-category models corresponding to the target major category model to which it belongs, obtains the second classification probability output by each sub-category model, and obtains the work order classification result output by the classification model according to the target major category model and the second classification probability output by each sub-category model.
[0095] S207. Determine whether the work order classification result is normal. If the work order classification result is that the work order is normal, execute S209 and stop root cause location; if the work order classification result is that the work order is abnormal, execute S208.
[0096] Specifically, it is determined whether the work order classification result is normal according to the sub - category model to which the work order belongs. The sub - category model can include a normal business model and an abnormal business model. If the work order is classified into the normal business model, for example, the work order is classified into the call fee query model, then the above - mentioned work order is considered a normal work order, and the call fee query model outputs the work order classification result, which indicates that the sub - category model to which the work order belongs is a normal business model. If the work order is classified into the abnormal business model, for example, the work order is classified into the complaint model for payment service attitude, then the above - mentioned work order is considered an abnormal work order, and the complaint model for payment service attitude outputs the work order classification result, which indicates that the sub - category model to which the work order belongs is an abnormal business model.
[0097] In the specific implementation process, the work order root cause location system determines whether the work order classification result is normal according to the business type represented by the sub - category model to which the work order belongs. If the work order classification result is that the work order is normal, the root cause location stops; if the work order classification result is that the work order is abnormal, the root cause location result is obtained according to the root cause location process.
[0098] S208. Obtain the root cause location result according to the root cause location process, and the root cause location result is the cause of the work order generation.
[0099] S209. Stop the root cause location.
[0100] Figure 3 This is the work order root cause location method process provided by the embodiment of the present application. Figure 3 。This embodiment is based on the above Figure 2 embodiment, and details the obtaining of the root cause location result according to the root cause location process in the root cause location method. As Figure 3 shown, the method includes:
[0101] S301. Match the corresponding root cause location process according to the work order classification result. The root cause location process is used to generate the cause of the abnormality; among them, each sub - category in each major category corresponds to its own root cause location judgment logic.
[0102] Specifically, the root cause location process consists of multiple nodes, each node having corresponding judgment logic. The judgment logic in the node can process data and output the root cause location information for each node. For example, a work order is classified into the sub-category of "failure to power on when paying fees" in the major category of information control through the work order root cause location system. The work order root cause location system matches the root cause location process in the scenario of "failure to power on when paying fees" according to the work order classification result. This root cause location process consists of 10 nodes, namely, judging the number of users with abnormal billing flags and cancellation flags, judging the number of users' effective default payment relationships, judging the number of data in the basic information table corresponding to the number, judging whether the logical number is consistent with the service number, judging whether there is an effective main service for the user, judging whether the customer in the user information table is consistent with the account customer, judging whether the user number information and account information are normal, judging whether the service status of the user's main service is normal, judging whether the payment flags in the user information table and the basic information table are consistent, and judging whether the service status in the user information table is consistent with the service status in the service status table. When performing root cause judgment, each node generates root cause location information, which consists of normal root causes or abnormal root causes. For example, in the case of judging whether the logical number is consistent with the service number, if the judgment result is that the logical number is consistent with the service number, then this judgment result is used as the normal root cause and the root cause location information is output; if the judgment result is that the logical number is inconsistent with the service number, then this judgment result is used as the abnormal root cause and the root cause location information is output. The work order root cause location system judges the cause of "failure to power on when paying fees" based on multiple nodes in the above root cause location process.
[0103] In the specific implementation process, the work order root cause location system matches the corresponding root cause location process according to the work order classification result, and the root cause location process is used to generate the cause of the exception; among them, the sub-categories in each major category correspond to their respective root cause location judgment logics.
[0104] S302. Obtain the target data source according to the work order classification result; among them, different classifications correspond to different data sources.
[0105] Specifically, the target data source includes the user center data source, the information control center data source, the accounting center data source, the charging center data source, the account center data source, other data sources, etc. For example, when the work order classification result is related to information control, the work order root cause location system obtains the business information in the information control center data source.
[0106] S303. Obtain the business information of the user in the work order from the target data source.
[0107] Specifically, the user business information is the user information required for each node in the root cause location process to perform logical judgment.
[0108] In a possible design, before the work order root cause location system obtains the business information of the user in the work order from the target data source, it further includes: extracting the user feature identifier and the business scenario feature identifier in the work order, and comparing the above feature identifiers with the user feature identifier and the business scenario feature identifier in the target data source to verify the accuracy of the classification of the work order root cause location system.
[0109] S304. Obtain the data types and judgment logics required for multiple nodes sorted in time sequence in the root cause location process.
[0110] S305. Obtain the corresponding data from the business information according to the data type required for each node.
[0111] S306. Process the data according to the judgment logic corresponding to the node to obtain the root cause location information of each node.
[0112] S307. Obtain the abnormal root cause node according to the root cause location information of each node, and obtain the root cause location result according to the abnormal root cause node.
[0113] In a possible design, in the scenario of non-booting after payment, the abnormal root cause nodes are to judge the payment credit control processing process, judge the description of the first payment credit control processing process, and judge the description of the second payment credit control processing process. The root cause location information corresponding to the above abnormal root cause nodes is {number of business delayed work orders: 1}, {the payment boot storage process is correctly executed}, {CRM processing error, work order postponed}. According to the above abnormal root cause nodes and their corresponding root cause location information, the work order root cause location system can obtain the root cause location result as: there are business in-transit orders.
[0114] In the specific implementation process, the work order root cause location system obtains the root cause location information of each node in the root cause location process, obtains the root cause node corresponding to the abnormal root cause in the root cause location information, and obtains the root cause location result according to the abnormal root cause node.
[0115] S308. Obtain the solution corresponding to the abnormal root cause node.
[0116] In a possible design, in the above scenario of non-booting after payment, the work order root cause location system obtains the root cause location result as there are business in-transit orders. The work order root cause location system performs keyword matching in the database according to the above location result to obtain the corresponding solution as: Please contact the business for processing and then perform zero payment operation.
[0117] S309. Feedback the abnormal information to the user terminal, and the abnormal information includes the abnormal root cause node and the corresponding solution.
[0118] Figure 4 Structural schematic of the work order root cause location device provided by the embodiment of the present applicationFigure 1 As shown in Figure 4 Figure 40, the root cause location device 40 of the work order includes: an acquisition module 401, a classification module 402, and an execution module 403.
[0119] The acquisition module 401 is used to acquire a work order and process the work order to obtain a work order feature vector.
[0120] The classification module 402 is used to input the work order feature vector into a classification model to obtain a work order classification result output by the classification model; wherein, the classification model includes multiple major category models, and each major category model is correspondingly provided with multiple minor category models, and the work order classification result is used to indicate whether the minor category in the major category to which the work order belongs is abnormal;
[0121] If the work order classification result is that the work order is abnormal, then match the corresponding root cause location process according to the work order classification result, and the root cause location process is used to generate the cause of the abnormality; wherein, each minor category in each major category corresponds to its own root cause location judgment logic.
[0122] The execution module 403 is used to obtain a root cause location result according to the root cause location process, and the root cause location result is the cause of the work order.
[0123] In a possible design, the acquisition module 401 is further used to:
[0124] Preprocess the work order, and the preprocessing includes at least one of invalid work order elimination, sensitive word replacement, stop word replacement, or symbol clearing;
[0125] Perform word segmentation on the preprocessed work order to obtain a word segmentation result;
[0126] Perform vectorization processing on the word segmentation result to obtain a work order feature vector.
[0127] In a possible design, the acquisition module 401 is further used to:
[0128] In response to a configuration request from a user terminal, feedback a judgment logic configuration interface to the user terminal, and the judgment logic configuration interface is used to input judgment information, and the judgment information includes a judgment name, a judgment type, available attributes, and judgment conditions;
[0129] Generate the judgment logic of each node according to the judgment information fed back by the user terminal;
[0130] Generate a root cause location process according to the judgment type and the judgment logic of each node; wherein, the judgment type is used to indicate that the judgment logic is a starting judgment or a process judgment.
[0131] In a possible design, the classification module 402 is further used to:
[0132] Input the work order feature vector into each major category model, obtain the first classification probability output by each major category model, and determine the target major category model to which the work order belongs according to the first classification probability; wherein, the major category model is a classification model;
[0133] Input the work order into multiple sub-category models corresponding to the target major category model, and obtain the second classification probability output by each sub-category model; wherein, the sub-category model is a classification model;
[0134] According to the target major category model and the second classification probabilities output by each sub-category model, obtain the work order classification result output by the classification model.
[0135] In a possible design, the execution module 403 is further configured to:
[0136] Obtain the target data source according to the work order classification result; wherein, different classifications correspond to different data sources;
[0137] Obtain the business information of the user in the work order from the target data source;
[0138] Obtain the root cause location result according to the business information and the root cause location process.
[0139] In a possible design, the execution module 403 is further configured to:
[0140] Obtain the data types and judgment logics required by multiple nodes sorted in time sequence in the root cause location process;
[0141] According to the data types required by each node, obtain the corresponding data from the business information;
[0142] Process the data according to the judgment logic corresponding to the node, and obtain the root cause location information of each node;
[0143] According to the root cause location information of each node, obtain the abnormal root cause node, and obtain the root cause location result according to the abnormal root cause node.
[0144] In a possible design, the execution module 403 is further configured to:
[0145] Obtain the solution corresponding to the abnormal root cause node;
[0146] Feedback the abnormal information to the user terminal, and the abnormal information includes the abnormal root cause node and the corresponding solution.
[0147] Figure 5 This is the hardware schematic diagram of the work order root cause location device provided by the embodiments of the present application. As Figure 5As shown in the figure, the work order root cause positioning device 50 provided in this embodiment includes: at least one processor 501 and a memory 502. The device 50 also includes a communication component 503. Among them, the processor 501, the memory 502, and the communication component 503 are connected through a bus 504.
[0148] In a specific implementation process, at least one processor 501 executes the computer-executable instructions stored in the memory 502, so that at least one processor 501 executes the work order root cause positioning method as described above.
[0149] For the specific implementation process of the processor 501, reference can be made to the above method embodiments. Their implementation principles and technical effects are similar, and will not be elaborated here in this embodiment.
[0150] This application also provides an electronic device, including: a processor, and a memory communicatively connected to the processor. The memory stores computer-executable instructions, and the processor executes the computer-executable instructions stored in the memory to implement the above work order root cause positioning method.
[0151] This application also provides a computer-readable storage medium. Computer-executable instructions are stored in the computer-readable storage medium. When the processor executes the computer-executable instructions, the above work order root cause positioning method is implemented.
[0152] For the above computer-readable storage medium, the above-readable storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, a magnetic disk or an optical disc. The readable storage medium can be any available medium accessible by a general-purpose or special-purpose computer.
[0153] It should be noted that for the foregoing method embodiments, for the sake of simple description, they are all expressed as a series of action combinations. However, those skilled in the art should know that this application is not limited by the described action sequence, because according to this application, certain steps can be performed in other sequences or simultaneously. Secondly, those skilled in the art should also know that the embodiments described in the specification are all optional embodiments, and the actions and modules involved are not necessarily essential to this application.
[0154] It should be further noted that although the steps in the flowchart are shown in sequence according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless there is a clear description in this article, the execution of these steps has no strict order limit, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowchart may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily executed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be executed alternately or in turn with at least a part of other steps or sub-steps or stages of other steps.
[0155] It should be understood that the above device embodiments are illustrative only, and the devices of the present application can also be implemented in other ways. For example, the division of units / modules in the above embodiments is only a logical function division, and there can be other division methods in actual implementation. For example, multiple units, modules or components can be combined, or can be integrated into another system, or some features can be ignored or not executed.
[0156] In addition, without special instructions, in each embodiment of the present application, each functional unit / module can be integrated in one unit / module, or each unit / module can exist physically alone, or two or more units / modules can be integrated together. The above integrated unit / module can be implemented in the form of hardware or in the form of a software program module.
[0157] When the integrated unit / module is implemented in the form of hardware, the hardware can be a digital circuit, an analog circuit, etc. The physical implementation of the hardware structure includes but is not limited to transistors, memristors, etc. Without special instructions, the processor can be any suitable hardware processor, such as CPU, GPU, FPGA, DSP, and ASIC, etc. Without special instructions, the storage unit can be any suitable magnetic storage medium or magneto-optical storage medium, such as resistive random access memory RRAM (Resistive Random Access Memory), dynamic random access memory DRAM (Dynamic Random Access Memory), static random access memory SRAM (Static Random-Access Memory), enhanced dynamic random access memory EDRAM (Enhanced Dynamic Random Access Memory), high-bandwidth memory HBM (High-Bandwidth Memory), hybrid memory cube HMC (Hybrid Memory Cube), etc.
[0158] When the integrated unit / module is implemented in the form of a software program module and sold or used as an independent product, it can be stored in a computer-readable memory. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a memory and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods of various embodiments of this application. And the aforementioned memory includes: various media that can store program codes, such as USB flash drives, read-only memories (ROM, Read-Only Memory), random access memories (RAM, Random Access Memory), mobile hard disks, magnetic disks, or optical discs.
[0159] In the above embodiments, the descriptions of each embodiment have their own emphases. For the parts not detailed in a certain embodiment, reference can be made to the relevant descriptions of other embodiments. The technical features of the above embodiments can be combined arbitrarily. For the sake of brevity of description, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, it should be considered as within the scope described in this specification.
[0160] Those skilled in the art will readily conceive of other embodiments of this application after considering the specification and practicing the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of this application that follow the general principles of this application and include common general knowledge or conventional technical means in the technical field not disclosed in this application. The specification and the embodiments are only regarded as exemplary, and the true scope and spirit of this application are pointed out by the following claims.
[0161] It should be understood that this application is not limited to the exact structures already described and shown in the drawings, and various modifications and changes can be made without departing from its scope. The scope of this application is only limited by the appended claims.
Claims
1. A method for locating the root cause of a work order, characterized in that, The method includes: Obtaining a work order and processing the work order to obtain a work order feature vector; Inputting the work order feature vector into a classification model to obtain a work order classification result output by the classification model; wherein, the classification model includes multiple major category models, and each major category model is correspondingly provided with multiple minor category models, and the work order classification result is used to indicate whether the minor category in the major category to which the work order belongs is abnormal; If the work order classification result is that the work order is abnormal, then matching a corresponding root cause location process according to the work order classification result, and the root cause location process is used to generate the cause of the abnormality; wherein, each minor category in each major category corresponds to its own root cause location judgment logic; Obtaining a root cause location result according to the root cause location process, and the root cause location result is the cause of the work order.
2. The method according to claim 1, characterized in that, The obtaining a work order and processing the work order to obtain a work order feature vector includes: Preprocessing the work order, and the preprocessing includes at least one of invalid work order elimination, sensitive word replacement, stop word replacement, or symbol clearing; Performing word segmentation processing on the preprocessed work order to obtain a word segmentation result; Performing vectorization processing on the word segmentation result to obtain a work order feature vector.
3. The method according to claim 1, characterized in that, The inputting the work order feature vector into a classification model to obtain a work order classification result output by the classification model includes: Inputting the work order feature vector into each major category model, obtaining a first classification probability output by each major category model, and determining the target major category model to which the work order belongs according to the first classification probability; wherein, the major category model is a classification model; Inputting the work order into the multiple minor category models corresponding to the target major category model, and obtaining a second classification probability output by each minor category model; wherein, the minor category model is a classification model; Obtaining a work order classification result output by the classification model according to the target major category model and the second classification probabilities output by each minor category model.
4. The method according to claim 1, characterized in that, The obtaining a root cause location result according to the root cause location process includes: Obtaining a target data source according to the work order classification result; wherein, different classifications correspond to different data sources; Obtaining the business materials of the user in the work order from the target data source; Obtaining a root cause location result according to the business materials and the root cause location process.
5. The method according to claim 4, characterized in that, The obtaining a root cause location result according to the business materials and the root cause location process includes: Obtaining the data types and judgment logics required by multiple nodes sorted in time sequence in the root cause location process; Obtaining corresponding data from the business materials according to the data types required by each node; Processing the data according to the judgment logic corresponding to the node to obtain root cause location information of each node; Obtaining an abnormal root cause node according to the root cause location information of each node, and obtaining the root cause location result according to the abnormal root cause node.
6. The method according to claim 5, characterized in that, Before matching a corresponding root cause location process according to the work order classification result, the method further includes: In response to a configuration request from a user terminal, a judgment logic configuration interface is fed back to the user terminal. The judgment logic configuration interface is used to input judgment information, and the judgment information includes a judgment name, a judgment type, available attributes, and judgment conditions. Based on the judgment information fed back by the user terminal, the judgment logic for each node is generated. Based on the judgment type and the judgment logic for each node, the root cause location process is generated; wherein the judgment type is used to indicate whether the judgment logic is a starting judgment or a process judgment.
7. The method according to claim 5, characterized in that, After obtaining the root cause location result according to the abnormal root cause node, the method further includes: Obtaining a solution corresponding to the abnormal root cause node. Feeding back abnormal information to the user terminal, where the abnormal information includes the abnormal root cause node and the corresponding solution.
8. A device for locating the root cause of a work order, characterized in that, Includes: An acquisition module, configured to acquire a work order and process the work order to obtain a work order feature vector. A classification module, configured to input the work order feature vector into a classification model to obtain a work order classification result output by the classification model; wherein the classification model includes multiple major category models, and each major category model is correspondingly provided with multiple minor category models, and the work order classification result is used to indicate whether the minor category in the major category to which the work order belongs is abnormal. If the work order classification result indicates that the work order is abnormal, a corresponding root cause location process is matched according to the work order classification result. The root cause location process is used to generate the cause of the abnormality; wherein each minor category in each major category corresponds to its own root cause location judgment logic. An execution module, configured to obtain a root cause location result according to the root cause location process, and the root cause location result is the cause of the work order.
9. An electronic device, characterized in that, Includes: A processor and a memory communicatively connected to the processor. The memory stores computer execution instructions. The processor executes the computer execution instructions stored in the memory to implement the method according to any one of claims 1 to 7.
10. A computer-readable storage medium, characterized in that, Computer execution instructions are stored in the computer-readable storage medium, and when the computer execution instructions are executed by a processor, they are used to implement the method according to any one of claims 1 to 7.