Deep learning-based work order processing method and related device

Abstract

The invention discloses a deep learning-based work order processing method and a related device, and belongs to the technical field of work order processing. The method comprises the following steps: establishing a work order processing model based on a deep learning algorithm through historical work order data; inputting a to-be-processed work order into the pre-trained work order processing model based on the deep learning algorithm, extracting keywords in the to-be-processed work order, and obtaining processing suggestions and related information of the to-be-processed work order according to the keywords; and finally, generating a work order hotspot map based on the related information of the to-be-processed work order. And processing the work order through the processing suggestion of the to-be-processed work order and the work order hotspot map. According to the invention, work orders can be automatically classified, and related work order types can be rapidly determined; on the basis of an intelligent recommendation mode, similar solutions can be quickly recommended, an optimal solution can be provided, time consumption and complexity of customer service work order processing are saved, work order processing efficiency is effectively improved, and user experience is greatly improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of work order processing, and relates to a work order processing method based on deep learning and related devices. Background Art

[0002] At present, the work order information of the marketing system lacks effective data sorting and intelligent analysis. When receiving a work order, the front-line team members can only rely on their own experience to handle it, and the level of automation and intelligence is low. At the same time, due to the decentralized management of work order processing, similar faults in the same area may be handled by different operation and maintenance personnel, and the processing information is not interoperable. Historical work order processing measures are not effectively referenced and utilized, resulting in a waste of manpower and energy, and indirectly increasing the burden on the front-line team.

[0003] Existing work order statistical analysis methods can only rely on manual operations, including work order classification, work order distribution, problem diagnosis, etc. These operations require a lot of time and manpower and are inefficient. Secondly, manual operations inevitably lead to inconsistent work order processing quality due to negligence or errors, especially when the workload is large or the team is small. This situation often occurs. Furthermore, relying on manual processing cannot guarantee timely response and processing when emergency work orders occur, which seriously affects business processes and user experience. Summary of the invention

[0004] The purpose of the present invention is to provide a work order processing method and related devices based on deep learning, so as to solve the technical problems in the prior art that work order processing relies on manpower, resulting in untimely processing and low efficiency.

[0005] In order to achieve the above object, the present invention adopts the following technical solutions: In a first aspect, the present invention provides a work order processing method based on deep learning, comprising the following steps: Get pending work orders; Input the work order to be processed into the pre-trained work order processing model based on the deep learning algorithm, extract keywords from the work order to be processed, and obtain processing suggestions and related information for the work order to be processed based on the keywords; Generate a work order hotspot map based on the relevant information of the pending work orders; process the work orders based on the processing suggestions of the pending work orders and the work order hotspot map; Among them, the work order processing model based on deep learning algorithm is established through historical work order data.

[0006] Furthermore, the historical work order data includes the acceptance content, acceptance time, classification information, address information, processing results and power supply station of the work order.

[0007] Furthermore, the training process of the work order processing model based on the deep learning algorithm includes the following steps: Obtain preprocessed historical work order data as the first training sample set; Inputting the first training sample set into a work order processing model based on a deep learning algorithm, extracting and learning keywords from the content in the first training sample set through the deep learning algorithm, and obtaining a second training sample set; The second training sample set is summarized and classified to form processing suggestions for the work orders, and a trained work order processing model based on the deep learning algorithm is obtained.

[0008] Furthermore, the step of inputting the first training sample set into the work order processing model based on the deep learning algorithm, extracting keywords and learning the content in the first training sample set by the deep learning algorithm, and obtaining the second training sample set specifically includes: Keywords are extracted and learned from the data in the first training sample set to obtain the work order acceptance content, classification information, address information and processing results. The specific expression is: W={w1,w2,…wn} Wherein, W represents the accepted content; w1 is the first characteristic information of the accepted content; w2 is the second characteristic information of the accepted content; wn is the nth characteristic information of the accepted content; X = {x1, x2, …xn} In the formula, X represents the classification information; x1 is the first feature information of the classification information; x2 is the second feature information of the classification information; xn is the nth feature information of the classification information; Y = {y1, y2, …yn} In the formula, Y represents the address information; y1 is the first characteristic information of the address information; y2 is the second characteristic information of the address information; yn is the nth characteristic information of the address information; Z={z1,z2,…zn} In the formula, Z represents the processing result; z1 is the first feature information of the processing result; z2 is the second feature information of the processing result; zn is the nth feature information of the processing result; The work order acceptance content, classification information, address information and processing results are combined to form a second training sample set.

[0009] Furthermore, the step of summarizing and classifying the second training sample set to form processing suggestions for the work orders and obtaining a trained work order processing model based on a deep learning algorithm specifically includes: The accepted content, classification information and processing results in the second training sample set are matched with the equation Z=g(W), X=h(W); the address information is matched with the power supply station and the key customer respectively, power supply station=i(Y), key customer=j(Y); the g, h, i and j are all algorithms summarized by deep learning.

[0010] Furthermore, the step of inputting the work order to be processed into the pre-trained work order processing model based on the deep learning algorithm, extracting keywords from the work order to be processed, and obtaining processing suggestions and related information for the work order to be processed according to the keywords specifically includes: Input the pending work orders into the pre-trained work order processing model based on deep learning algorithm, extract keywords from the pending work orders, and obtain the acceptance content, acceptance time, classification information and address information; Match the classification information and the accepted content to generate processing suggestions for pending work orders; Generate a processing time limit based on the acceptance time, and remind the work orders that are about to time out based on the processing time limit; Identify and repair addresses with errors in address information and match them with power supply stations; The final output is the processing suggestions, processing time limit, address information and power supply station of the pending work order.

[0011] Furthermore, the step of generating a work order hotspot map based on the relevant information of the work orders to be processed specifically includes: inputting the relevant information of the work orders to be processed into a platform built by folium visualization technology to form a work order hotspot map.

[0012] In a second aspect, the present invention provides a work order processing system based on deep learning, comprising: Data acquisition module, used to obtain pending work orders; A result generation module is used to input the work order to be processed into the pre-trained work order processing model based on the deep learning algorithm, extract keywords from the work order to be processed, and obtain processing suggestions and related information of the work order to be processed based on the keywords; The work order processing module is used to generate a work order hotspot map based on the relevant information of the work orders to be processed; and process the work orders through the processing suggestions of the work orders to be processed and the work order hotspot map.

[0013] In a third aspect, the present invention provides a computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the above method when executing the computer program.

[0014] In a fourth aspect, the present invention provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the above method are implemented.

[0015] Compared with the prior art, the present invention has the following beneficial effects: The present invention discloses a work order processing method and related devices based on deep learning. Through deep learning technology, historical work order data is learned, and a work order processing model based on deep learning algorithm is constructed. The work order can be automatically classified and the relevant work order type can be quickly determined. Based on the intelligent recommendation method, similar solutions can be quickly recommended and the optimal solution can be provided, which saves the time and complexity of customer service processing work orders, effectively improves the work order processing efficiency, and greatly improves the user experience. The work order processing cost is reduced. Through intelligent recommendation and automatic classification and processing of work orders, the demand for customer service personnel and technical maintenance personnel can be reduced. At the same time, the burden and time cost of front-line maintenance personnel can be reduced by mining hotspot maps, taking corresponding measures for hot issues, and summarizing experience. The present invention can quickly solve most work orders in a short time, can find service process bottlenecks and problems, and optimize related processes, further improving the company's operating efficiency and economic benefits.

[0016] Furthermore, after the work order is handled, the present invention forms a hot spot map of all work orders through folium visualization technology, which can visually view the hot spots where various work orders occur, and provide key reminders to customers who have duplicate work orders, and provide common solutions to hot issues, thereby reducing the burden and time cost of front-line maintenance personnel, improving the accuracy of customer service personnel's responses, and thus improving service quality and customer satisfaction. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for use in the embodiments are briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without creative work.

[0018] Figure 1 is a flow chart of the method of the present invention; Figure 2 is a schematic diagram of the system of the present invention; Figure 3 This is a schematic diagram of the principle of the deep learning algorithm of an embodiment of the present invention; Figure 4 This is a schematic diagram of the work order processing flow of the model according to the embodiment of the present invention; Figure 5 A first work order hotspot map generated for an embodiment of the present invention; Figure 6 A second work order hotspot map generated for an embodiment of the present invention; Figure 7 This is a schematic diagram of exporting key customer information according to an embodiment of the present invention; Figure 8A third work order hotspot map generated for an embodiment of the present invention; Fig. 9 It is a schematic diagram of the computer device structure of the present invention. DETAILED DESCRIPTION

[0019] The present invention will be described in detail below with reference to the accompanying drawings and in combination with embodiments. It should be noted that the embodiments and features in the embodiments of the present application can be combined with each other without conflict.

[0020] The following detailed description is an exemplary description, which is intended to provide further detailed description of the present invention. Unless otherwise specified, all technical terms used in the present invention have the same meaning as those generally understood by those skilled in the art to which the present application belongs. The terms used in the present invention are only for describing specific embodiments, and are not intended to limit exemplary embodiments according to the present invention.

[0021] Embodiment 1: See also Figure 1 , an embodiment of the present invention discloses a work order processing method based on deep learning, comprising the following steps: S1, obtain the work order to be processed; A work order processing model based on a deep learning algorithm is established through historical work order data; the historical work order data includes the acceptance content, acceptance time, classification information, address information, processing results and power supply station of the work order.

[0022] The acceptance content of the work order is a detailed description of the electricity-related problems encountered by the customer. This model uses a deep learning algorithm to extract keywords from the description, and combines the corresponding classification and processing content of the work order to form a database. At the same time, when similar content appears in the work order content in the future, it can automatically classify the work order and give processing suggestions based on the information in the historical data.

[0023] The address information of the work order. This model uses a deep learning algorithm to extract the street and community names in the address, and combines it with the corresponding power supply station information to automatically repair and identify addresses with errors, such as "Jiayuan" being mistakenly recorded as "Jiayuan". Subsequently, the address of the work order and the corresponding power supply station can be accurately located based on the address.

[0024] Regarding the acceptance time in the work order, this model can automatically give a work order handling time limit prompt and remind the work order that is about to time out after processing the historical work order using a deep learning algorithm based on the acceptance content and corresponding classification, combined with the Beijing company's time limit requirements for different types of work orders.

[0025] Furthermore, in work orders, for the same address or the same customer, the system will create an important file and mark the number and type of work orders to remind the power supply station.

[0026] S2, input the work order to be processed into the pre-trained work order processing model based on the deep learning algorithm, extract the keywords in the work order to be processed, and obtain the processing suggestions and related information of the work order to be processed according to the keywords, such as Figure 4 As shown; S201, input the work order to be processed into a pre-trained work order processing model based on a deep learning algorithm, extract keywords from the work order to be processed, and obtain acceptance content, acceptance time, classification information and address information; S202, matching the classification information with the accepted content, and generating processing suggestions for the work orders to be processed; S203, generating a processing time limit according to the acceptance time, and reminding the work order that is about to time out according to the processing time limit; S204, identifying and repairing addresses with errors in the address information, and matching the power supply station; S205, finally outputting the processing suggestions, processing time limit, address information and power supply station of the work order to be processed.

[0027] S3, generating a work order hotspot map based on the relevant information of the work order to be processed; processing the work order according to the processing suggestions of the work order to be processed and the work order hotspot map.

[0028] The relevant information of the work orders to be processed is output to the platform built by folium visualization technology to form a work order hotspot map for staff to refer to. The staff will process the work orders based on this information, and promptly update the work order details in the model and update the system in real time.

[0029] Embodiment 2: In a possible embodiment of the present invention, see Figure 3 The training process of the work order processing model based on the deep learning algorithm includes the following steps: A) Obtain preprocessed historical work order data as the first training sample set; B) inputting the first training sample set into a work order processing model based on a deep learning algorithm, extracting and learning keywords from the content in the first training sample set through the deep learning algorithm, and obtaining a second training sample set; Keywords are extracted and learned from the data in the first training sample set to obtain the work order acceptance content, classification information, address information and processing results. The specific expression is: W={w1,w2,…wn} Wherein, W represents the accepted content; w1 is the first characteristic information of the accepted content; w2 is the second characteristic information of the accepted content; wn is the nth characteristic information of the accepted content; X = {x1, x2, …xn} In the formula, X represents the classification information; x1 is the first feature information of the classification information; x2 is the second feature information of the classification information; xn is the nth feature information of the classification information; Y = {y1, y2, …yn} In the formula, Y represents the address information; y1 is the first characteristic information of the address information; y2 is the second characteristic information of the address information; yn is the nth characteristic information of the address information; Z={z1,z2,…zn} In the formula, Z represents the processing result; z1 is the first feature information of the processing result; z2 is the second feature information of the processing result; zn is the nth feature information of the processing result; The work order acceptance content, classification information, address information and processing results are combined to form a second training sample set.

[0030] C) Summarize and classify the second training sample set to form processing suggestions for the work order, and obtain a trained work order processing model based on the deep learning algorithm.

[0031] The accepted content, classification information and processing results in the second training sample set are matched with the equation Z=g(W), X=h(W); the address information is matched with the power supply station and the key customer respectively, power supply station=i(Y), key customer=j(Y); the g, h, i and j are all algorithms summarized by deep learning.

[0032] For example, "on-site delivery" and "handheld game console" in the work order processing content are used as two feature values ​​in the processing result information, and "not delivered", "payment of *** yuan", and "power outage" in the acceptance content are used as feature values ​​extracted from the acceptance content. For the deep learning algorithm, "on-site delivery, handheld game console" can be connected with "not delivered", "payment of *** yuan", and "power outage". The same problem can be used to continue to correct the features. Finally, the most appropriate feature association is extracted until a known work order is given, and the algorithm model can give an answer that is basically consistent with the known result. At this time, it is basically considered that the model is basically built.

[0033] Finally, when a new work order is entered, the g, h, i, and j algorithms can be automatically called to output the corresponding results, and the results can be directly input into the platform built by folium visualization technology to form a work order hotspot map and provide information prompts to front-line staff.

[0034] Embodiment 3: This embodiment first presets a deep learning algorithm model, which is a model obtained after training with historical work order data information in the past three years as training samples; the model can specifically use deep learning algorithms to quickly classify, locate, warn, and make work suggestions for work orders. Since work order requirements are relatively broad, the use of deep learning algorithms is more convenient for processing work orders, and the extraction of keywords through fuzzy judgment can deepen the connection between keywords, and finally derive work order processing suggestions.

[0035] This embodiment first establishes a work order processing model based on a deep learning algorithm and then performs model training. Model training determines the basic content of the work order, such as what type of work order it is, what type of demand it expresses, what the processing suggestions are, which power supply station it belongs to, how long the processing time is, and when the final completion deadline is. All content is fed back to the staff to assist them in quickly completing the work order processing.

[0036] The following describes the basic composition of the work order information required to import the system, i.e. the training samples: The acquisition of training samples can be completed by the Marketing 2.0 system. The basic data comes from historical work order information, which is the first training set, including work order classification. The second training sample set is obtained by extracting keywords and combining them with information such as address and processing content. The training sample set can select historical work orders, such as all work orders in the past three years, and establish training samples through phenomenon feature annotation, keywords, etc.

[0037] The main work in the model training phase is to carefully control the work orders, summarize the description features of each type of work order, and learn the work orders with the same features to summarize the corresponding classification and processing methods. At the same time, it can accurately process the work order address, automatically correct the address with input errors (such as "Jiayuan" mistakenly entered as "Jiayuan"), match the power supply station, and give a time limit reminder.

[0038] The original information of the work order is the first training set. Based on the first training set, customers with repeated work orders and high-frequency addresses can be preliminarily located, and accurate results can be obtained by combining with the second training set.

[0039] Based on the second training sample set, the processing results, classification information and acceptance content can be matched by equation Z=g(W), X=h(W), g and h are algorithms summarized by deep learning. Match the address with the power supply station and key customers. Power supply station=i(Y), repeated customer=j(Y), i and j are algorithms summarized by deep learning.

[0040] Finally, when a new work order is entered, the g, h, i, and j algorithms can be automatically called to output the corresponding results, and the results can be directly input into the platform built by folium visualization technology to form a work order hotspot map and provide information prompts to front-line staff.

[0041] Specifically, the input content is as follows: Accepted content: [Electricity not issued] The customer with account number 1109031839600 paid 100 yuan at 3:00 on November 3. The electricity fee has been credited to the account, but it has not been issued to the meter. The customer side [power outage, meter remaining 0 yuan], please verify and handle as soon as possible.

[0042] Category: Business application - Electricity service demand - Electricity quantity not issued Handling situation: Gao Ming of Majiabao Power Supply Station of Fengtai Power Supply Company used the phone number 67564410 to contact the customer at 20:50 on November 7, 2024. The person who answered the phone was the customer who reported the problem. Customer number: 1109027521800, meter type: electronic-intelligent local fee control (CPU card), meter manufacturer: Shenzhen Kelu, meter service life: 10 years, the meter abnormality is that the electricity fee has not been issued, which does not affect normal use. The reason is that the customer's on-site collection signal is unstable. The electricity fee was successfully issued at 21:15 on November 7, 2024. Liu Qiang of Majiabao Power Supply Station of Fengtai Power Supply Company used the phone number 67564410 to inform the customer at 21:17 on November 7, 2024 that the electricity fee had been issued, and the customer expressed satisfaction.

[0043] Address: Room x, Block x, Jiaomen North Road, Jiaomen Dongli West Community Committee, Majiabao Sub-district Office, Fengtai District, Beijing Power supply station: Majiabao The work order is subjected to keyword extraction (here only a brief description of the algorithm is given, and the specific algorithm learning process is more complicated). The content extraction can extract keywords such as payment of ** yuan, deposit, not issued, etc. The type extraction is business application-electricity service demand-power not issued for matching, and the processing results are extracted from keywords such as unstable acquisition signal and successful issuance. The address extraction includes keywords such as Majiabao Street, Jiaomen Dongli Community, and No. x Courtyard, Jiaomen North Road.

[0044] After keyword extraction, the keywords are matched accordingly. As a preliminary learning result, new work orders with similar content are matched to see if the keywords can be further simplified or supplemented to form a new formula algorithm. Finally, after processing all work orders, an algorithm formula that can cover all features is obtained.

[0045] After the model training is completed, a new work order is input, and the trained algorithm formula can be called to decompose the work order, and all the results of this work order are output to the platform built by folium visualization technology for display.

[0046] Embodiment 4: In this embodiment, the annual work order forms of Huaxiang Work Order Office from 2021 to 2023 are imported into the work order processing model based on the deep learning algorithm as the first training sample set. After the model's internal algorithm, the second training sample set is automatically generated. After historical work order training, the model is displayed in the platform built by folium visualization technology as follows: Figure 5 You can select different types of work orders to handle on the interface, such as Figure 6 shown; in addition, see Figure 7 , you can export the important customers involved in the work order, that is, the customers with more frequent work orders. Importing new work orders into the trained model, you can get the following Figure 8 The model in this embodiment automatically gives the time of arrival and handling of this work order, and gives the work order address, power supply station and handling suggestions in detail. It is convenient for staff to check and effectively improves the efficiency of work order handling.

[0047] The present invention uses deep learning technology to learn the historical work order data of the past three years, and constructs a fast and efficient deep learning classification model, which can automatically classify work orders and quickly determine the relevant work order types; then, based on the determined separation, the algorithm is used to summarize the historical disposal methods and provide them to the work order acceptance personnel, saving time-consuming and repetitive manual work. The work order information processed by the above process is visualized using folium visualization technology to form a work order hotspot map. The area and frequency of different types of work orders can be displayed on the map, and key reminders can be given to customers with repeated failures, forming a good work reminder for the staff, effectively improving the efficiency of work order disposal. Improve customer service experience.

[0048] See also Figure 2 An embodiment of the present invention provides a work order processing system based on deep learning, including a data acquisition module, a result generation module and a work order processing module.

[0049] Among them, the data acquisition module is used to obtain the work orders to be processed; the result generation module is used to input the work orders to be processed into the pre-trained work order processing model based on the deep learning algorithm, extract the keywords in the work orders to be processed, and obtain the processing suggestions and related information of the work orders to be processed based on the keywords; the work order processing module is used to generate a work order hotspot map based on the relevant information of the work orders to be processed; and the work orders are processed through the processing suggestions of the work orders to be processed and the work order hotspot map.

[0050] In one embodiment of the present invention, see Fig. 9 , a computer device is provided, the computer device includes a processor and a memory, the memory is used to store a computer program, the computer program includes program instructions, and the processor is used to execute the program instructions stored in the computer storage medium. The processor may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc. It is the computing core and control core of the terminal, which is suitable for implementing one or more instructions, and is specifically suitable for loading and executing one or more instructions in a computer storage medium to implement a corresponding method flow or a corresponding function; the processor described in the embodiment of the present invention can be used for the operation of a work order processing method based on deep learning.

[0051] The present invention also provides a storage medium, specifically a computer-readable storage medium (Memory), which is a memory device in a computer device for storing programs and data. It is understandable that the computer-readable storage medium here can include both built-in storage media in a computer device and, of course, extended storage media supported by the computer device. The computer-readable storage medium provides a storage space, which stores the operating system of the terminal. In addition, one or more instructions suitable for being loaded and executed by a processor are also stored in the storage space, and these instructions can be one or more computer programs (including program codes). It should be noted that the computer-readable storage medium here can be a high-speed RAM memory or a non-volatile memory (non-volatile memory), such as at least one disk memory. The processor can load and execute one or more instructions stored in the computer-readable storage medium to implement the corresponding steps of the work order processing method based on deep learning in the above embodiment.

[0052] It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0053] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0054] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0055] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0056] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the relevant field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents. Any modification or equivalent replacement that does not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. A work order processing method based on deep learning, characterized in that: The following steps are involved: Get pending work orders; Input the work order to be processed into the pre-trained work order processing model based on the deep learning algorithm, extract keywords from the work order to be processed, and obtain processing suggestions and related information for the work order to be processed based on the keywords; Generate a work order hotspot map based on the relevant information of the pending work orders; process the work orders based on the processing suggestions of the pending work orders and the work order hotspot map; Among them, the work order processing model based on deep learning algorithm is established through historical work order data.

2. A work order processing method based on deep learning according to claim 1, characterized in that: The historical work order data includes the acceptance content, acceptance time, classification information, address information, processing results and power supply station of the work order.

3. The work order processing method based on deep learning according to claim 1, characterized in that: The training process of the work order processing model based on the deep learning algorithm includes the following steps: Obtain preprocessed historical work order data as the first training sample set; Inputting the first training sample set into a work order processing model based on a deep learning algorithm, extracting and learning keywords from the content in the first training sample set through the deep learning algorithm, and obtaining a second training sample set; The second training sample set is summarized and classified to form processing suggestions for the work orders, and a trained work order processing model based on the deep learning algorithm is obtained.

4. A work order processing method based on deep learning according to claim 3, characterized in that: The step of inputting the first training sample set into the work order processing model based on the deep learning algorithm, extracting keywords and learning the content in the first training sample set by the deep learning algorithm, and obtaining the second training sample set specifically includes: Keywords are extracted and learned from the data in the first training sample set to obtain the work order acceptance content, classification information, address information and processing results. The specific expression is: W={w1,w2,…wn} Wherein, W represents the accepted content; w1 is the first characteristic information of the accepted content; w2 is the second characteristic information of the accepted content; wn is the nth characteristic information of the accepted content; X = {x1, x2, …xn} In the formula, X represents the classification information; x1 is the first feature information of the classification information; x2 is the second feature information of the classification information; xn is the nth feature information of the classification information; Y = {y1, y2, …yn} In the formula, Y represents the address information; y1 is the first characteristic information of the address information; y2 is the second characteristic information of the address information; yn is the nth characteristic information of the address information; Z={z1,z2,…zn} In the formula, Z represents the processing result; z1 is the first feature information of the processing result; z2 is the second feature information of the processing result; zn is the nth feature information of the processing result; The work order acceptance content, classification information, address information and processing results are combined to form a second training sample set.

5. A work order processing method based on deep learning according to claim 4, characterized in that: The step of summarizing and classifying the second training sample set to form processing suggestions for the work orders and obtaining a trained work order processing model based on a deep learning algorithm specifically includes: The accepted content, classification information and processing results in the second training sample set are matched with the equation Z=g(W), X=h(W); the address information is matched with the power supply station and the key customer respectively, power supply station=i(Y), key customer=j(Y); the g, h, i and j are all algorithms summarized by deep learning.

6. A work order processing method based on deep learning according to claim 1, characterized in that: The step of inputting the work order to be processed into the pre-trained work order processing model based on the deep learning algorithm, extracting keywords from the work order to be processed, and obtaining processing suggestions and related information for the work order to be processed according to the keywords specifically includes: Input the pending work orders into the pre-trained work order processing model based on deep learning algorithm, extract keywords from the pending work orders, and obtain the acceptance content, acceptance time, classification information and address information; Match the classification information and the accepted content to generate processing suggestions for pending work orders; Generate a processing time limit based on the acceptance time, and remind the work orders that are about to time out based on the processing time limit; Identify and repair addresses with errors in address information and match them with power supply stations; The final output is the processing suggestions, processing time limit, address information and power supply station of the pending work order.

7. A work order processing method based on deep learning according to claim 1, characterized in that: The step of generating a work order hotspot map based on the relevant information of the work order to be processed specifically includes: inputting the relevant information of the work order to be processed into a platform built by folium visualization technology to form a work order hotspot map.

8. A work order processing system based on deep learning, characterized in that: include: Data acquisition module, used to obtain pending work orders; A result generation module is used to input the work order to be processed into the pre-trained work order processing model based on the deep learning algorithm, extract keywords from the work order to be processed, and obtain processing suggestions and related information of the work order to be processed based on the keywords; The work order processing module is used to generate a work order hotspot map based on the relevant information of the work orders to be processed; and process the work orders through the processing suggestions of the work orders to be processed and the work order hotspot map.

9. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images