Work order assignment method and device, storage medium and electronic equipment

By using a target model based on convolutional neural network in the banking system to extract and classify work orders, the work order attributes are automatically predicted, which solves the problem of low accuracy of work order distribution caused by manual filling, and improves the accuracy of work order distribution. and user experience.

CN118941010APending Publication Date: 2024-11-12INDUSTRIAL AND COMMERCIAL BANK OF CHINA
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202410976192.3
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-19
Publication Date
2024-11-12

AI Technical Summary

Technical Problem

In the prior art, the distribution is based on manually filled-in work order information, resulting in low accuracy of work order distribution, resulting in an extended work order circulation time, affecting user experience and processing efficiency.

Method used

The target model that trains the convolutional neural network based on the training sample set is adopted. Through feature extraction and classification, the attribution system of work orders is automatically predicted, the error allocation rate is reduced, and the accuracy of work order dispatch is improved.

Benefits of technology

It effectively reduces the error rate of first-time assignment of work orders, reduces the number of transfers, improves the accuracy of work order assignments, and improves the user experience.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118941010A_ABST
    Figure CN118941010A_ABST
Patent Text Reader

Abstract

The invention discloses a work order dispatching method and device, a storage medium and electronic equipment, and relates to the field of financial science and technology or other related fields. The method comprises the following steps: acquiring work order information of each to-be-dispatched work order, wherein the work order information is used for describing problem information corresponding to the to-be-dispatched work order; performing feature extraction on the work order information of each to-be-dispatched work order through a target model to obtain a feature vector corresponding to each to-be-dispatched work order, and determining a target work order processing system corresponding to each to-be-dispatched work order according to the feature vector corresponding to each to-be-dispatched work order, the target model is obtained by training a convolutional neural network for text classification based on a training sample set; and each work order to be dispatched is dispatched to the target work order processing system corresponding to each work order to be dispatched. The technical problem that in the prior art, work order dispatching is carried out according to related information filled manually, and the accuracy of work order dispatching is low is solved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of financial technology or other related fields, and in particular, to a work order dispatching method, device, storage medium and electronic device. Background Art

[0002] With the continuous increase of new business functions, the number of work orders generated in the banking system every day is also growing. At present, the existing technology mainly assigns work orders based on the system to which the work order filled in by the business personnel belongs. The recognition accuracy of the system to which the work order belongs is not high, and work order assignment errors often occur. The accuracy of work order assignment is low, and the assignment of wrong work orders prolongs the work order circulation time, affects the user experience, and increases the burden on the work order processor, affecting the work efficiency of the processor.

[0003] To address the above-mentioned problems, no effective solution has been proposed yet. Summary of the invention

[0004] The embodiments of the present invention provide a work order dispatching method, device, storage medium and electronic device to at least solve the technical problem in the prior art that work order dispatching is performed based on manually filled relevant information, and the work order dispatching has low accuracy.

[0005] According to one aspect of an embodiment of the present invention, there is provided a work order dispatching method, comprising: obtaining work order information of each work order to be dispatched, wherein the work order information is used to describe problem information corresponding to the work order to be dispatched; performing feature extraction on the work order information of each work order to be dispatched through a target model to obtain a feature vector corresponding to each work order to be dispatched, and determining a target work order processing system corresponding to each work order to be dispatched based on the feature vector corresponding to each work order to be dispatched, wherein the target model is obtained by training a convolutional neural network for text classification based on a training sample set; and dispatching each work order to be dispatched to the target work order processing system corresponding to each work order to be dispatched.

[0006] Furthermore, the target work order processing system corresponding to each work order to be dispatched is determined based on the feature vector corresponding to each work order to be dispatched, including: determining the target business category corresponding to each work order to be dispatched based on the feature vector corresponding to each work order to be dispatched; determining the target work order processing system corresponding to each work order to be dispatched based on the target business category corresponding to each work order to be dispatched.

[0007] Furthermore, based on the feature vector corresponding to each work order to be dispatched, the target business category corresponding to each work order to be dispatched is determined, including: inputting the feature vector corresponding to each work order to be dispatched into a target classifier; classifying the feature vector corresponding to each work order to be dispatched through the target classifier to obtain the target business category corresponding to each work order to be dispatched.

[0008] Furthermore, before obtaining the work order information of each work order to be assigned, the method also includes: obtaining the work order description text of each work order to be assigned, and performing word segmentation processing on the work order description text of each work order to be assigned to obtain multiple words corresponding to each work order to be assigned; filtering the multiple words corresponding to each work order to be assigned according to a preset word set to obtain the filtered words corresponding to each work order to be assigned; processing the filtered words corresponding to each work order to be assigned according to a neural network model for generating word vectors to obtain the word vector representation corresponding to each work order to be assigned; and determining the work order information of each work order to be assigned based on the word vector representation corresponding to each work order to be assigned.

[0009] Furthermore, a target model is generated through the following steps: historical work order processing data of multiple work order processing systems are obtained, and the historical work order processing data are preprocessed to obtain a training sample set, wherein the training samples in the training sample set include a first label and a second label, the first label is used to identify the business category corresponding to the training sample, and the second label is used to identify the work order processing system corresponding to the training sample; the initial model is trained based on the training sample set to obtain a target model.

[0010] Furthermore, the initial model at least includes: an initial classifier and a convolutional neural network for text classification, wherein the initial model is trained according to the training sample set to obtain a target model, including: training the convolutional neural network for text classification and the initial classifier according to the training sample set respectively to obtain a trained model and a target classifier; and forming a target model based on the trained model and the target classifier.

[0011] Furthermore, the training sample set at least includes: a training set, a validation set and a test set, wherein the convolutional neural network and the initial classifier for text classification are trained respectively according to the training sample set to obtain a trained model and a target classifier, including: the convolutional neural network and the initial classifier for text classification are trained respectively according to the training set to obtain a first model and a first classifier; feature extraction is performed on the validation sample in the validation set according to the first model to obtain a feature vector corresponding to the validation sample, and the feature vector corresponding to the validation sample is classified according to the first classifier to obtain a predicted business category and a predicted work order processing system corresponding to the validation sample; the first classifier is optimized according to the first label, the second label, the predicted business category corresponding to the validation sample, the predicted work order processing system and the test set, and the model parameters of the first model are adjusted by feedback until the accuracy of the first classifier meets the preset threshold, and the trained model and the target classifier are obtained.

[0012] According to another aspect of an embodiment of the present invention, there is also provided a work order dispatching device, including: an acquisition module, used to obtain the work order information of each work order to be dispatched, wherein the work order information is used to describe the problem information corresponding to the work order to be dispatched; a determination module, used to perform feature extraction on the work order information of each work order to be dispatched through a target model, obtain a feature vector corresponding to each work order to be dispatched, and determine a target work order processing system corresponding to each work order to be dispatched based on the feature vector corresponding to each work order to be dispatched, wherein the target model is obtained by training a convolutional neural network for text classification based on a training sample set; and a processing module, used to dispatch each work order to be dispatched to the target work order processing system corresponding to each work order to be dispatched.

[0013] According to another aspect of an embodiment of the present invention, a computer program product is provided, including a computer program, and when the computer program is executed by a processor, the work order dispatching method described above is implemented.

[0014] According to another aspect of an embodiment of the present invention, a computer-readable storage medium is provided, in which a computer program is stored, wherein the computer program is configured to execute the above-mentioned work order dispatching method when running.

[0015] According to another aspect of an embodiment of the present invention, there is also provided an electronic device, which includes one or more processors; a memory for storing one or more programs, which enables the one or more processors to run the programs when the one or more programs are executed by the one or more processors, wherein the programs are configured to execute the above-mentioned work order dispatching method when running.

[0016] In an embodiment of the present invention, a target model obtained by training a convolutional neural network for text classification based on a training sample set is used to determine a target work order processing system corresponding to each work order to be dispatched. First, the work order information of each work order to be dispatched is obtained, wherein the work order information is used to describe the problem information corresponding to the work order to be dispatched; the work order information of each work order to be dispatched is feature extracted through the target model to obtain a feature vector corresponding to each work order to be dispatched, and the target work order processing system corresponding to each work order to be dispatched is determined based on the feature vector corresponding to each work order to be dispatched, wherein the target model is obtained by training a convolutional neural network for text classification based on a training sample set; each work order to be dispatched is dispatched to the target work order processing system corresponding to each work order to be dispatched.

[0017] In the above process, the target model can be used to automatically predict the system to which the work order belongs in combination with the work order problem information, thereby reducing the error rate of the first assignment of the work order and effectively reducing the number of misassignments, thereby reducing the number of circulation times and improving the accuracy of work order dispatch, thereby reducing the waiting time for work order issues and improving user experience.

[0018] It can be seen that the technical solution of the present invention has achieved the purpose of improving user experience, thereby realizing the technical effect of improving the accuracy of work order assignment, and further solving the technical problem of low accuracy of work order assignment in the prior art based on manually filled in relevant information. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] The drawings described herein are used to provide a further understanding of the present invention and constitute a part of this application. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings:

[0020] Figure 1 is a flow chart of an optional work order dispatching method according to an embodiment of the present invention;

[0021] Figure 2 is a schematic diagram of an optional process of intelligent work order dispatching according to an embodiment of the present invention;

[0022] Figure 3 is a schematic diagram of an optional work order dispatching device according to an embodiment of the present invention;

[0023] Figure 4 is a schematic diagram of an optional electronic device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0024] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0025] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0026] It should be noted that the relevant information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data for display, data for analysis, etc.) involved in the present invention are all information and data authorized by the user or fully authorized by all parties. For example, an interface is set between the system and the relevant user or organization. Before obtaining relevant information, it is necessary to send an acquisition request to the aforementioned user or organization through the interface, and obtain relevant information after receiving the consent information fed back by the aforementioned user or organization.

[0027] Example 1

[0028] According to an embodiment of the present invention, an embodiment of a work order dispatching method is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0029] Figure 1 is a flow chart of an optional work order dispatching method according to an embodiment of the present invention. Figure 1 As shown, the method comprises the following steps:

[0030] Step S101, obtaining the work order information of each work order to be assigned, wherein the work order information is used to describe the problem information corresponding to the work order to be assigned.

[0031] In the above steps, the work order information of each work order to be dispatched can be obtained through application systems, processors, electronic devices and other devices. Optionally, the work order information of each work order to be dispatched can be obtained through a work order dispatching system, wherein the work order information can be in the form of a word vector representation.

[0032] In an optional embodiment, before obtaining the work order information of each work order to be assigned, obtain the work order description text of each work order to be assigned, and perform word segmentation processing on the work order description text of each work order to be assigned to obtain multiple words corresponding to each work order to be assigned; filter the multiple words corresponding to each work order to be assigned according to a preset word set to obtain the filtered words corresponding to each work order to be assigned; process the filtered words corresponding to each work order to be assigned according to a neural network model for generating word vectors to obtain the word vector representation corresponding to each work order to be assigned; determine the work order information of each work order to be assigned based on the word vector representation corresponding to each work order to be assigned.

[0033] Optionally, since the basic unit of Chinese semantics is a word, in order to vectorize the work order description text, the work order description text needs to be decomposed into words. Therefore, before obtaining the work order information of each work order to be assigned, the work order description text of each work order to be assigned is obtained, and the work order description text of each work order to be assigned is processed by word segmentation to obtain multiple words corresponding to each work order to be assigned, for example, by using a word segmentation tool based on the forward maximum matching principle to perform text segmentation.

[0034] Optionally, since the work order description text contains some words that are not related to the problem characteristics, such as interjections, adjectives, conjunctions, etc., these words often appear but contain almost no work order feature information, and will weaken the performance of the work order feature information and affect the classification performance, such words are collected to form an interference vocabulary set (i.e., a preset word set) to facilitate filtering of the word segmentation results. For example, multiple words corresponding to each work order to be assigned are filtered according to the interference vocabulary set to obtain the filtered words corresponding to each work order to be assigned.

[0035] Optionally, in order to enable the model to understand the text after word segmentation and filtering, word vectorization is required. For example, a neural network model for generating word vectors (such as a Word2vec model) is used to process the filtered words corresponding to each work order to be assigned, and the words are represented as dense vectors of a given fixed dimension. Each dimension of the vector contains lexical features, and a word vector representation corresponding to each work order to be assigned is obtained, so that the work order information of each work order to be assigned can be determined based on the word vector representation corresponding to each work order to be assigned. For example, the word vector representation corresponding to each work order to be assigned is used as the work order information of each work order to be assigned.

[0036] Step S102, feature extraction is performed on the work order information of each work order to be dispatched through the target model to obtain the feature vector corresponding to each work order to be dispatched, and the target work order processing system corresponding to each work order to be dispatched is determined based on the feature vector corresponding to each work order to be dispatched, wherein the target model is obtained by training a convolutional neural network for text classification based on a training sample set.

[0037] Optionally, a convolutional neural network for text classification is trained based on a training sample set. For example, a Text Convolutional Neural Network (Text Convolutional Neural Network) model is trained using preprocessed work order data to obtain an intelligent work order dispatching model (i.e., a target model). Feature extraction is performed on the work order information of each work order to be dispatched through the intelligent work order dispatching model to obtain a feature vector corresponding to each work order to be dispatched. Then, the target work order processing system corresponding to each work order to be dispatched is determined based on the feature vector corresponding to each work order to be dispatched, thereby realizing a system that automatically predicts the work order attribution based on the work order problem information.

[0038] Optionally, the features of Chinese sentences are mainly reflected in the local correlation between words and contexts. The word vectors are extracted by using TextCNN. For example, the decomposed word vector is represented as an L×D matrix, where L represents rows and D represents columns. In order to obtain different feature vectors from one-dimensional convolution, a convolution kernel with a variable width and a size of N×D is used for feature extraction. Each convolution kernel outputs L-N+1 feature vectors, where N represents rows. The maximum value of the feature vectors extracted by each group of convolution kernels can be cascaded to obtain the final feature vector of the group of convolution kernels, and the maximum value of the final feature vector of each group of convolution kernels can be cascaded to obtain the final feature vector.

[0039] Step S103: dispatch each work order to be dispatched to a target work order processing system corresponding to each work order to be dispatched.

[0040] Optionally, after predicting the system to which the work order belongs based on the intelligent work order dispatching model, each work order to be dispatched can be dispatched to the target work order processing system corresponding to each work order to be dispatched, thereby reducing the error rate of the first assignment of the work order and effectively reducing the number of misassignments.

[0041] Based on the scheme defined by the above steps S101 to S103, it can be known that in an embodiment of the present invention, a target model obtained by training a convolutional neural network for text classification based on a training sample set is used to determine the target work order processing system corresponding to each work order to be dispatched. First, the work order information of each work order to be dispatched is obtained, wherein the work order information is used to describe the problem information corresponding to the work order to be dispatched; the work order information of each work order to be dispatched is subjected to feature extraction through the target model to obtain a feature vector corresponding to each work order to be dispatched, and the target work order processing system corresponding to each work order to be dispatched is determined based on the feature vector corresponding to each work order to be dispatched, wherein the target model is obtained by training a convolutional neural network for text classification based on a training sample set; each work order to be dispatched is dispatched to the target work order processing system corresponding to each work order to be dispatched.

[0042] It is easy to notice that in the above process, the target model can be used to automatically predict the system to which the work order belongs in combination with the work order problem information, thereby reducing the error rate of the first assignment of the work order and effectively reducing the number of misassignments, thereby reducing the number of circulation times and improving the accuracy of work order dispatch, thereby reducing the waiting time for work order problems and improving user experience.

[0043] It can be seen that the technical solution of the present invention has achieved the purpose of improving user experience, thereby realizing the technical effect of improving the accuracy of work order assignment, and further solving the technical problem of low accuracy of work order assignment in the prior art based on manually filled in relevant information.

[0044] In an optional embodiment, the target work order processing system corresponding to each work order to be dispatched is determined based on the feature vector corresponding to each work order to be dispatched, including: determining the target business category corresponding to each work order to be dispatched based on the feature vector corresponding to each work order to be dispatched; determining the target work order processing system corresponding to each work order to be dispatched based on the target business category corresponding to each work order to be dispatched.

[0045] Optionally, in the process of determining the target work order processing system corresponding to each work order to be dispatched according to the feature vector corresponding to each work order to be dispatched, firstly, the target business category corresponding to each work order to be dispatched is determined according to the feature vector corresponding to each work order to be dispatched.

[0046] In an optional embodiment, in the process of determining the target business category corresponding to each work order to be dispatched based on the feature vector corresponding to each work order to be dispatched, the feature vector corresponding to each work order to be dispatched is input into a target classifier, and then the feature vector corresponding to each work order to be dispatched is classified by the target classifier to obtain the target business category corresponding to each work order to be dispatched.

[0047] Optionally, the target classifier can be a classifier based on multinomial distribution and Softmax function modeling. By inputting the feature vectors of all texts into the classifier, the category corresponding to each text can be output. For example, the classifier outputs that the business category of work order 1 to be dispatched is Class A business, and the business category of work order 2 to be dispatched is Class B business.

[0048] Optionally, the target work order processing system corresponding to each work order to be assigned may be determined according to the target business category corresponding to each work order to be assigned. For example, the work order processing system corresponding to the A-type business is the first system, and the work order processing system corresponding to the B-type business is the second system.

[0049] It should be noted that the intelligent work order dispatching model can automatically predict the system to which the work order belongs based on the work order problem information, thereby reducing the error rate of the first assignment of work orders and effectively reducing the number of misassignments, thereby reducing the number of circulation times and improving the accuracy of work order dispatch, thereby reducing the waiting time for work order issues and improving user experience.

[0050] In an optional embodiment, a target model is generated by the following steps: obtaining historical work order processing data of multiple work order processing systems, and performing data preprocessing on the historical work order processing data to obtain a training sample set, wherein the training samples in the training sample set include a first label and a second label, the first label is used to identify the business category corresponding to the training sample, and the second label is used to identify the work order processing system corresponding to the training sample; training the initial model based on the training sample set to obtain the target model.

[0051] Optionally, first determine the training sample set. By acquiring historical work order processing data from multiple work order processing systems and performing data preprocessing such as word segmentation, filtering, and labeling on the historical work order processing data, a training sample set can be obtained. For example, in order to ensure a sufficient number of training samples and the timeliness of newer data, the historical data and knowledge base data of the work order processing system are used as training sample sets.

[0052] Optionally, since the basic unit of Chinese semantics is a word, in order to vectorize the work order description text, the work order description text needs to be decomposed into words. Therefore, the historical work order processing data is segmented to obtain corresponding multiple words, for example, by using a segmentation tool based on the forward maximum matching principle to perform text segmentation.

[0053] Optionally, since the work order description text contains some words that are irrelevant to the problem characteristics, such as interjections, adjectives, conjunctions, etc., these words appear frequently but contain almost no work order feature information, and will weaken the performance of the work order feature information and affect the classification performance, such words are collected to form an interference vocabulary set to facilitate filtering of the word segmentation results. For example, the word segmentation results are filtered according to the interference vocabulary set to obtain filtered words.

[0054] Optionally, the initial model is trained according to the training sample set to obtain the target model (i.e., the aforementioned intelligent work order dispatching model). In an optional embodiment, the initial model includes at least: an initial classifier and a convolutional neural network for text classification. In the process of training the initial model according to the training sample set to obtain the target model, the convolutional neural network for text classification and the initial classifier are trained according to the training sample set to obtain a trained model and a target classifier. Then, the target model can be composed according to the trained model and the target classifier. For example, the intelligent work order dispatching model can be composed according to the trained TextCNN model and the classifier obtained by modeling based on multinomial distribution and Softmax function.

[0055] It should be noted that by generating an intelligent work order dispatch model, a data basis is provided for subsequent work order dispatch, so that the intelligent work order dispatch model can be used in combination with work order problem information to automatically predict the system to which the work order belongs, thereby reducing the error rate of the first assignment of work orders, effectively reducing the number of misassignments, reducing the number of transfers, and improving the accuracy of work order dispatch, thereby reducing the waiting time for work order issues and improving user experience.

[0056] In an optional embodiment, the training sample set includes at least: a training set, a validation set and a test set, wherein a convolutional neural network and an initial classifier for text classification are trained respectively according to the training sample set to obtain a trained model and a target classifier, including: training the convolutional neural network and the initial classifier for text classification respectively according to the training set to obtain a first model and a first classifier; extracting features of the validation samples in the validation set according to the first model to obtain feature vectors corresponding to the validation samples, and classifying the feature vectors corresponding to the validation samples according to the first classifier to obtain predicted business categories and predicted work order processing systems corresponding to the validation samples; optimizing the first classifier according to the first label, the second label, the predicted business category corresponding to the validation sample, the predicted work order processing system and the test set, and feedback adjusting the model parameters of the first model until the accuracy of the first classifier meets a preset threshold to obtain a trained model and a target classifier.

[0057] Optionally, firstly, the convolutional neural network and the initial classifier for text classification are trained according to the training set to obtain the first model and the first classifier, then the verification sample in the verification set can be feature extracted according to the first model to obtain the feature vector corresponding to the verification sample, and the feature vector corresponding to the verification sample can be classified according to the first classifier to obtain the predicted business category and the predicted work order processing system corresponding to the verification sample, then the first classifier can be optimized according to the first label, the second label, the predicted business category corresponding to the verification sample, the predicted work order processing system and the test set, and the model parameters of the first model can be adjusted by feedback until the accuracy of the first classifier meets the preset threshold, and the trained model and target classifier are obtained. For example, during the training process, the feature vectors of all texts are input into the classifier obtained by modeling based on multinomial distribution and Softmax function for classification judgment, and the category corresponding to each text is output, and the output predicted category is compared with the existing label, and the classification result and the labeling result are verified, and then the model parameters are continuously adjusted by feedback, and after a large number of iterative training, the model with high classification accuracy and its weight parameters are saved to obtain the trained model and target classifier.

[0058] Figure 2 is a flow chart of an optional intelligent work order dispatching process according to an embodiment of the present invention, such as Figure 2 As shown, first determine the training sample set, then perform text preprocessing, including text segmentation, segmentation filtering, etc., then perform model training based on the preprocessed data, perform feature vector representation of the text during the training process, then perform feature selection, and then use a classifier to classify the test sample to achieve intelligent work order dispatching. The specific implementation details can refer to the content of the aforementioned embodiment, which will not be repeated here.

[0059] In an embodiment of the present invention, a target model obtained by training a convolutional neural network for text classification based on a training sample set is used to determine a target work order processing system corresponding to each work order to be dispatched. First, the work order information of each work order to be dispatched is obtained, wherein the work order information is used to describe the problem information corresponding to the work order to be dispatched; the work order information of each work order to be dispatched is feature extracted through the target model to obtain a feature vector corresponding to each work order to be dispatched, and the target work order processing system corresponding to each work order to be dispatched is determined based on the feature vector corresponding to each work order to be dispatched, wherein the target model is obtained by training a convolutional neural network for text classification based on a training sample set; each work order to be dispatched is dispatched to the target work order processing system corresponding to each work order to be dispatched.

[0060] It is easy to notice that in the above process, the target model can be used to automatically predict the system to which the work order belongs in combination with the work order problem information, thereby reducing the error rate of the first assignment of the work order and effectively reducing the number of misassignments, thereby reducing the number of circulation times and improving the accuracy of work order dispatch, thereby reducing the waiting time for work order problems and improving user experience.

[0061] It can be seen that the technical solution of the present invention has achieved the purpose of improving user experience, thereby realizing the technical effect of improving the accuracy of work order assignment, and further solving the technical problem of low accuracy of work order assignment in the prior art based on manually filled in relevant information.

[0062] Example 2

[0063] According to an embodiment of the present invention, an embodiment of a work order dispatching device is provided, wherein: Figure 3 is a schematic diagram of an optional work order dispatching device according to an embodiment of the present invention, such as Figure 3 As shown, the device includes: an acquisition module 301, used to obtain the work order information of each work order to be dispatched, wherein the work order information is used to describe the problem information corresponding to the work order to be dispatched; a determination module 302, used to extract features of the work order information of each work order to be dispatched through a target model, obtain a feature vector corresponding to each work order to be dispatched, and determine the target work order processing system corresponding to each work order to be dispatched based on the feature vector corresponding to each work order to be dispatched, wherein the target model is obtained by training a convolutional neural network for text classification based on a training sample set; a processing module 303, used to dispatch each work order to be dispatched to the target work order processing system corresponding to each work order to be dispatched.

[0064] In the work order dispatching device provided by the embodiment of the present invention, the work order information of each work order to be dispatched is obtained by the acquisition module 301, wherein the work order information is used to describe the problem information corresponding to the work order to be dispatched; the determination module 302 extracts the features of the work order information of each work order to be dispatched through the target model, obtains the feature vector corresponding to each work order to be dispatched, and determines the target work order processing system corresponding to each work order to be dispatched according to the feature vector corresponding to each work order to be dispatched, wherein the target model is obtained by training the convolutional neural network for text classification based on the training sample set; the processing module 303 dispatches each work order to be dispatched to the target work order processing system corresponding to each work order to be dispatched. The target model can be used to automatically predict the system to which the work order belongs in combination with the work order problem information, thereby reducing the error rate of the first assignment of the work order, effectively reducing the number of misassignments, thereby reducing the number of circulations, and improving the accuracy of work order dispatching, thereby reducing the waiting time for work order problems and improving user experience. The purpose of improving user experience is achieved, thereby realizing the technical effect of improving the accuracy of work order dispatching, and further solving the technical problem of low accuracy of work order dispatching in the prior art based on manually filled relevant information.

[0065] It should be noted that the above-mentioned acquisition module 301, determination module 302 and processing module 303 correspond to steps S101 to S103 in the above-mentioned embodiment, and the examples and application scenarios implemented by the three modules and the corresponding steps are the same, but are not limited to the contents disclosed in the above-mentioned embodiment 1.

[0066] Optionally, the determination module includes: a first determination module, used to determine the target business category corresponding to each work order to be dispatched based on the feature vector corresponding to each work order to be dispatched; a second determination module, used to determine the target work order processing system corresponding to each work order to be dispatched based on the target business category corresponding to each work order to be dispatched.

[0067] Optionally, the first determination module includes: an input module, used to input the feature vector corresponding to each work order to be dispatched into a target classifier; and a classification module, used to classify the feature vector corresponding to each work order to be dispatched through the target classifier to obtain a target business category corresponding to each work order to be dispatched.

[0068] Optionally, the work order dispatching device also includes: a first acquisition module, which is used to obtain the work order description text of each work order to be dispatched before obtaining the work order information of each work order to be dispatched, and perform word segmentation processing on the work order description text of each work order to be dispatched to obtain multiple words corresponding to each work order to be dispatched; a first processing module, which is used to filter the multiple words corresponding to each work order to be dispatched according to a preset word set to obtain the filtered words corresponding to each work order to be dispatched; a second processing module, which is used to process the filtered words corresponding to each work order to be dispatched according to a neural network model for generating word vectors to obtain the word vector representation corresponding to each work order to be dispatched; and a third determination module, which is used to determine the work order information of each work order to be dispatched based on the word vector representation corresponding to each work order to be dispatched.

[0069] Optionally, the work order dispatching device also includes the following modules, which are used to generate a target model through the following steps: a second acquisition module, which is used to acquire historical work order processing data of multiple work order processing systems, and perform data preprocessing on the historical work order processing data to obtain a training sample set, wherein the training samples in the training sample set include a first label and a second label, the first label is used to identify the business category corresponding to the training sample, and the second label is used to identify the work order processing system corresponding to the training sample; a training module, which is used to train the initial model based on the training sample set to obtain a target model.

[0070] Optionally, the training module includes: a first training module, used to train the convolutional neural network and the initial classifier for text classification according to the training sample set, so as to obtain a trained model and a target classifier; and a generation module, used to form a target model based on the trained model and the target classifier.

[0071] Optionally, the first training module includes: a second training module, used to train the convolutional neural network and the initial classifier for text classification according to the training set, respectively, to obtain a first model and a first classifier; a third training module, used to extract features of the verification samples in the verification set according to the first model, to obtain feature vectors corresponding to the verification samples, and to classify the feature vectors corresponding to the verification samples according to the first classifier, to obtain predicted business categories and predicted work order processing systems corresponding to the verification samples; a fourth training module, used to optimize the first classifier based on the first label, the second label, the predicted business category corresponding to the verification sample, the predicted work order processing system, and the test set, and to feedback and adjust the model parameters of the first model until the accuracy of the first classifier meets the preset threshold, to obtain a trained model and a target classifier.

[0072] Example 3

[0073] According to another aspect of an embodiment of the present invention, a computer program product is provided, including a computer program, and when the computer program is executed by a processor, the work order dispatching method described above is implemented.

[0074] Example 4

[0075] According to another aspect of an embodiment of the present invention, a computer-readable storage medium is provided, in which a computer program is stored, wherein the computer program is configured to execute the above-mentioned work order dispatching method when running.

[0076] Example 5

[0077] According to another aspect of an embodiment of the present invention, there is also provided an electronic device, wherein: Figure 4 is a schematic diagram of an optional electronic device according to an embodiment of the present invention, such as Figure 4 As shown, the electronic device includes one or more processors; a memory for storing one or more programs, when the one or more programs are executed by the one or more processors, the one or more processors are used to run the programs, wherein the programs are set to execute the above-mentioned work order dispatching method when running. When the processor executes the program, the following steps are implemented: obtaining the work order information of each work order to be dispatched, wherein the work order information is used to describe the problem information corresponding to the work order to be dispatched; extracting features from the work order information of each work order to be dispatched through a target model to obtain a feature vector corresponding to each work order to be dispatched, and determining the target work order processing system corresponding to each work order to be dispatched based on the feature vector corresponding to each work order to be dispatched, wherein the target model is obtained by training a convolutional neural network for text classification based on a training sample set; and dispatching each work order to be dispatched to the target work order processing system corresponding to each work order to be dispatched.

[0078] Optionally, when the processor executes the program, the following steps are also implemented: based on the feature vector corresponding to each work order to be dispatched, the target business category corresponding to each work order to be dispatched is determined; based on the target business category corresponding to each work order to be dispatched, the target work order processing system corresponding to each work order to be dispatched is determined.

[0079] Optionally, when the processor executes the program, the following steps are also implemented: inputting the feature vector corresponding to each work order to be dispatched into the target classifier; classifying the feature vector corresponding to each work order to be dispatched through the target classifier to obtain the target business category corresponding to each work order to be dispatched.

[0080] Optionally, the processor also implements the following steps when executing the program: before obtaining the work order information of each work order to be assigned, obtain the work order description text of each work order to be assigned, and perform word segmentation on the work order description text of each work order to be assigned to obtain multiple words corresponding to each work order to be assigned; filter the multiple words corresponding to each work order to be assigned according to a preset word set to obtain the filtered words corresponding to each work order to be assigned; process the filtered words corresponding to each work order to be assigned according to a neural network model for generating word vectors to obtain the word vector representation corresponding to each work order to be assigned; determine the work order information of each work order to be assigned based on the word vector representation corresponding to each work order to be assigned.

[0081] Optionally, when the processor executes the program, it also implements the following steps: obtaining historical work order processing data of multiple work order processing systems, and performing data preprocessing on the historical work order processing data to obtain a training sample set, wherein the training samples in the training sample set include a first label and a second label, the first label is used to identify the business category corresponding to the training sample, and the second label is used to identify the work order processing system corresponding to the training sample; training the initial model based on the training sample set to obtain a target model.

[0082] Optionally, when the processor executes the program, the following steps are also implemented: based on the training sample set, the convolutional neural network and the initial classifier for text classification are trained to obtain a trained model and a target classifier; and the target model is composed based on the trained model and the target classifier.

[0083] Optionally, when the processor executes the program, the following steps are also implemented: according to the training set, the convolutional neural network and the initial classifier for text classification are trained respectively to obtain a first model and a first classifier; according to the first model, feature extraction is performed on the verification sample in the verification set to obtain a feature vector corresponding to the verification sample, and the feature vector corresponding to the verification sample is classified according to the first classifier to obtain a predicted business category and a predicted work order processing system corresponding to the verification sample; according to the first label, the second label, the predicted business category corresponding to the verification sample, the predicted work order processing system and the test set, the first classifier is optimized, and the model parameters of the first model are adjusted by feedback until the accuracy of the first classifier meets the preset threshold, so as to obtain a trained model and a target classifier.

[0084] The devices in this article can be servers, PCs, PADs, mobile phones, etc.

[0085] The serial numbers of the above embodiments of the present invention are only for description and do not represent the advantages or disadvantages of the embodiments.

[0086] In the above embodiments of the present invention, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0087] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. Among them, the device embodiments described above are only schematic. For example, the division of the units can be a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of units or modules, which can be electrical or other forms.

[0088] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple units. Some or all of the units may be selected according to actual needs to achieve the purpose of the present embodiment.

[0089] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0090] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for a computer device (which can be a personal computer, a server or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), mobile hard disk, magnetic disk or optical disk and other media that can store program codes.

[0091] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principle of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A work order dispatching method, characterized in that: include: Obtaining work order information of each work order to be assigned, wherein the work order information is used to describe the problem information corresponding to the work order to be assigned; Extracting features from the work order information of each work order to be assigned through a target model to obtain a feature vector corresponding to each work order to be assigned, and determining a target work order processing system corresponding to each work order to be assigned based on the feature vector corresponding to each work order to be assigned, wherein the target model is obtained by training a convolutional neural network for text classification based on a training sample set; Each of the work orders to be dispatched is dispatched to a target work order processing system corresponding to each of the work orders to be dispatched.

2. The method according to claim 1, characterized in that Determining a target work order processing system corresponding to each work order to be assigned according to the feature vector corresponding to each work order to be assigned includes: Determine the target business category corresponding to each work order to be assigned according to the feature vector corresponding to each work order to be assigned; According to the target business category corresponding to each work order to be dispatched, a target work order processing system corresponding to each work order to be dispatched is determined.

3. The method according to claim 2, characterized in that Determining the target business category corresponding to each work order to be assigned according to the feature vector corresponding to each work order to be assigned includes: Inputting the feature vector corresponding to each work order to be assigned into the target classifier; The target classifier is used to classify the feature vector corresponding to each work order to be dispatched, so as to obtain the target business category corresponding to each work order to be dispatched.

4. The method according to claim 1, characterized in that Before obtaining the work order information of each work order to be assigned, the method further includes: Acquire the work order description text of each work order to be assigned, and perform word segmentation processing on the work order description text of each work order to be assigned to obtain multiple words corresponding to each work order to be assigned; Filtering the multiple words corresponding to each of the work orders to be assigned according to a preset word set to obtain the filtered words corresponding to each of the work orders to be assigned; Processing the filtered words corresponding to each of the work orders to be assigned according to the neural network model for generating word vectors to obtain a word vector representation corresponding to each of the work orders to be assigned; The work order information of each work order to be assigned is determined according to the word vector representation corresponding to each work order to be assigned.

5. The method according to claim 1, characterized in that The target model is generated by the following steps: Acquire historical work order processing data of multiple work order processing systems, and perform data preprocessing on the historical work order processing data to obtain the training sample set, wherein the training samples in the training sample set include a first label and a second label, the first label is used to identify the business category corresponding to the training sample, and the second label is used to identify the work order processing system corresponding to the training sample; The initial model is trained according to the training sample set to obtain the target model.

6. The method according to claim 5, characterized in that The initial model at least includes: an initial classifier and the convolutional neural network for text classification, wherein the initial model is trained according to the training sample set to obtain the target model, including: According to the training sample set, the convolutional neural network for text classification and the initial classifier are trained to obtain a trained model and a target classifier; The target model is composed according to the trained model and the target classifier.

7. The method according to claim 6, characterized in that The training sample set at least includes: a training set, a validation set and a test set, wherein the convolutional neural network for text classification and the initial classifier are trained according to the training sample set to obtain a trained model and a target classifier, including: According to the training set, the convolutional neural network for text classification and the initial classifier are trained respectively to obtain a first model and a first classifier; Extracting features of the verification samples in the verification set according to the first model to obtain feature vectors corresponding to the verification samples, and classifying the feature vectors corresponding to the verification samples according to the first classifier to obtain predicted business categories and predicted work order processing systems corresponding to the verification samples; The first classifier is optimized based on the first label, the second label, the predicted business category corresponding to the verification sample, the predicted work order processing system and the test set, and the model parameters of the first model are adjusted by feedback until the accuracy of the first classifier meets the preset threshold, thereby obtaining the trained model and the target classifier.

8. A work order dispatching device, characterized in that: include: An acquisition module, used to acquire work order information of each work order to be assigned, wherein the work order information is used to describe the problem information corresponding to the work order to be assigned; A determination module, configured to extract features from the work order information of each work order to be assigned through a target model, obtain a feature vector corresponding to each work order to be assigned, and determine a target work order processing system corresponding to each work order to be assigned based on the feature vector corresponding to each work order to be assigned, wherein the target model is obtained by training a convolutional neural network for text classification based on a training sample set; The processing module is used to dispatch each of the work orders to be dispatched to a target work order processing system corresponding to each of the work orders to be dispatched.

9. A computer program product, characterized in that The invention comprises a computer program, which implements the work order dispatching method described in any one of claims 1 to 7 when being executed by a processor.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, wherein the computer program is configured to execute the work order dispatching method described in any one of claims 1 to 7 when running.

11. An electronic device, characterized in that: The electronic device includes one or more processors; A memory for storing one or more programs, which, when executed by the one or more processors, enables the one or more processors to run the programs, wherein the programs are configured to execute the work order dispatching method described in any one of claims 1 to 7 when run.