Work order processing method, device and electronic equipment

Abstract

This application provides a method, apparatus, and electronic device for processing work orders, including: acquiring N work orders to be processed, where N is a natural number not less than 2; traversing the N work orders and extracting tags and keywords contained in the N work orders based on a pre-constructed work order tag library and a work order keyword library, wherein the work order tag library includes tags of different work orders and the work order keyword library includes keywords of different work orders; calculating the similarity between any two work orders among the N work orders based on the tags and keywords, wherein each work order corresponds to N similarity scores; and classifying each work order based on the N similarity scores corresponding to each work order to obtain the classification results of the N work orders.

Description

Technical Field

[0001] This application relates to the field of data processing technology, and in particular to a method, apparatus and electronic device for processing work orders. Background Technology

[0002] A work order is a simple maintenance or manufacturing plan consisting of a header and at least one work line. Currently, work orders have been paperless and data centralized.

[0003] In some scenarios, work order processing is still mainly done manually. For example, the classification and processing of work orders basically rely on manual work. Due to the large number and variety of work orders, the processing efficiency and the accuracy of classification results are very low when using manual processing. Summary of the Invention

[0004] The purpose of this application is to provide a method, apparatus, and electronic device for processing work orders, so as to solve the problems of work order processing efficiency and classification accuracy.

[0005] To solve the above-mentioned technical problems, the embodiments of this application are implemented as follows:

[0006] Firstly, embodiments of this application provide a method for processing work orders, including:

[0007] Obtain N work orders to be processed, where N is a natural number not less than 2; traverse the N work orders, and extract the tags and keywords contained in the N work orders based on a pre-built work order tag library and a work order keyword library, respectively. The work order tag library includes tags for different work orders, and the work order keyword library includes keywords for different work orders; calculate the similarity between any two work orders among the N work orders based on the tags and keywords, where each work order corresponds to N similarity scores; classify each work order based on the N similarity scores corresponding to each work order to obtain the classification results of the N work orders.

[0008] Secondly, embodiments of this application provide a work order processing apparatus, comprising: an acquisition module, configured to acquire N work orders to be processed, wherein N is a natural number not less than 2; an extraction module, configured to traverse the N work orders and extract tags and keywords contained in the N work orders based on a pre-constructed work order tag library and a work order keyword library, wherein the work order tag library includes tags of different work orders and the work order keyword library includes keywords of different work orders; a calculation module, configured to calculate the similarity between any two work orders among the N work orders based on the tags and keywords, wherein each work order corresponds to N similarities; and a processing module, configured to classify each work order based on the N similarities corresponding to each work order to obtain a classification result of the N work orders.

[0009] Thirdly, embodiments of this application provide an electronic device, including a processor, a communication interface, a memory, and a communication bus; wherein the processor, the communication interface, and the memory communicate with each other via the communication bus; the memory is used to store computer programs; and the processor is used to execute the programs stored in the memory to implement the work order processing method steps as described in the first aspect.

[0010] Fourthly, embodiments of this application provide a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the work order processing method steps as described in the first aspect.

[0011] As can be seen from the technical solutions provided in the embodiments of this application above, the electronic device acquires N work orders to be processed, traverses the N work orders, and extracts the tags and keywords contained in the N work orders based on the pre-built work order tag library and work order keyword library. The work order tag library includes tags of different work orders, and the work order keyword library includes keywords of different work orders. The similarity between any two work orders in the N work orders is calculated based on the tags and keywords. The work orders are classified according to the N similarity scores corresponding to each work order, and the classification results of the N work orders are obtained. After the electronic device acquires the N work orders to be processed, it can automatically classify the N work orders without human intervention, which improves the processing efficiency of work orders and the accuracy of classification results. Attached Figure Description

[0012] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0013] Figure 1 A schematic diagram of a first method for processing work orders provided in an embodiment of this application;

[0014] Figure 2 A second flowchart illustrating the work order processing method provided in this application embodiment;

[0015] Figure 3 A schematic diagram of a third method for processing work orders provided in an embodiment of this application;

[0016] Figure 4 A schematic diagram of the module composition of the work order processing device provided in the embodiments of this application;

[0017] Figure 5This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Detailed Implementation

[0018] This application provides a method, apparatus, and electronic device for processing work orders, which solves the problems of work order processing efficiency and classification accuracy.

[0019] To enable those skilled in the art to better understand the technical solutions in this application, the technical solutions in the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of this application.

[0020] For example, such as Figure 1 As shown in the figure, this application embodiment provides a method for processing work orders. The execution subject of this method can be a server, wherein the server can be an independent server or a server cluster composed of multiple servers, and the server can be a server capable of processing work orders. The work order processing method specifically includes the following steps:

[0021] In step S101, N work orders to be processed are obtained. Where N is a natural number not less than 2.

[0022] Specifically, the system can retrieve automatically entered work orders. Specifically, users can call the customer service system, which prioritizes using an intelligent chatbot to answer questions, filtering out simple, repetitive work orders with standard solutions. Only complex issues are transferred to human operators. After answering, operators can choose a real-time speech-to-text service. Enabling this service allows for simultaneous speaking and recording. Automatic entry ends when the user hangs up. Once manually reviewed and approved, the work order is submitted, avoiding manual input and saving manpower and resources. The N work orders to be processed can be from the same user or from different users; this embodiment does not limit the scope of the application.

[0023] In this context, a work order, as a data processing object to be processed, is a request form submitted by staff for a user to handle or inquire about a service. The work order contains data regarding the service the user needs to handle, user information, and the user's questions. The service data includes, but is not limited to, service type data (e.g., mobile service, family service, group service, value-added service, etc.), specific service data (e.g., value-added package services within value-added services), or channel data. User information includes, but is not limited to, user name, ID number, contact information, user's location, and services already handled by the user. The user's questions refer to inquiries about the service, including but not limited to data regarding service fees, after-sales service, service touchpoints, basic services, marketing, network quality, and product quality. It is worth noting that, depending on actual needs, work orders may also contain other types of data, which are not limited in this embodiment.

[0024] In step S102, N work orders are traversed, and the tags and keywords contained in the N work orders are extracted based on the pre-built work order tag library and work order keyword library. The work order tag library includes tags for different work orders, and the work order keyword library includes keywords for different work orders.

[0025] Specifically, the work order tag library and work order keyword library are built upon a large amount of basic work order data. This basic work order data includes, but is not limited to, work order data details, product dimension tables, work order categories, and issue keywords. The work order data details include, but are not limited to, the work order identity document (ID), work order content, city name, and work order category. The product dimension table contains the names of all services that are available for subscription and have a subscription history. The work order category is a standardized system for work orders, representing the level of the work order (which can be divided into 1 to 7 levels of tags). Tags can be divided into business tags and issue tags. Business tags can be further divided into levels 1 to 4. Level 1 business tags can include mobile services, family services, group services, value-added services, and group maintenance. Level 2 business tags can include service touchpoints, basic services, business marketing, network quality, product quality, and group maintenance. Level 3 business tags can include specific services or specific channels. Level 4 business tags can include provincial maintenance. Issue tags can be divided into levels 5 and 6. Level 5 issue tags can include marketing and promotion, fee inquiries, processing standards, business rules, after-sales service, function usage, and group maintenance. Level 6 issue tags can include issue subdivisions and group maintenance. Level 7 issue tags can be divided into provincial maintenance. Issue keywords refer to the issue positioning keywords extracted from the work order content.

[0026] After determining the basic data mentioned above, a work order tag library and a work order keyword library are built. The work order tag library and work order keyword library support batch import, export, and manual addition, deletion, modification and query for easy maintenance in the future.

[0027] This involves creating a work order tag library based on work order classification standards and data structures. Specifically, the library is created according to the types of work order category tags, and work order categories are imported for use in work order tag analysis. During initialization, a weight value is assigned to the work order tag library by default. All levels within the current work order tag library and all work order tags within those levels share the same weight value.

[0028] Secondly, a business keyword library is created based on the product dimension table and data structure. The product dimension table data is imported and used when analyzing business keywords in work orders. When initializing the business keyword library, a weight value is assigned to the business keywords in the business keyword library by default. The keywords in the business keyword library share the same weight value.

[0029] Finally, an issue keyword library is created based on the issue locators extracted from the work order content. Issue keyword data is imported for use when analyzing work orders. A weight value is assigned to the issue keyword library by default during initialization, and all keywords in the issue keyword library share the same weight value. The business keyword library and the issue keyword library are both part of the work order keyword library. The work order keyword library includes business keywords and issue keywords from different work orders.

[0030] Furthermore, staff can adjust the weights of the work order tag library, business keyword library, and problem keyword library according to the actual situation, based on their granularity and priority. The smaller the granularity, the higher the priority, and the higher the weight can be set.

[0031] Iterate through all work orders, and extract the work order tags, business keywords, and question keywords from the work order tag library, business keyword library, and question keyword library for each work order. Work order tags need to be accurate to the tag level, i.e., marking the level of the tags contained in the N extracted work orders. Keywords need to have their position within the work order content marked, i.e., marking the position of each keyword within the corresponding work order's content. Different positions of keywords within the work order content result in different semantic meanings. The purpose of marking the position of keywords in the work order content is to analyze semantic dependencies through the mining of hypernyms and hyponyms, obtain the syntactic structure, and then combine the extracted work order tags, business keywords, and question keywords to generate a work order title for each work order.

[0032] Furthermore, text similarity semantic models, such as the RBT3 semantic model, can be used to convert the text in the work orders (such as tags, business keywords, and problem keywords) into text vectors, which facilitates the subsequent calculation of the similarity between work orders.

[0033] The RBT3 semantic model uses a multi-layer Transformer network structure. Through the Attention mechanism, the distance between two words at any position is converted into 1, effectively solving the thorny long-term dependency problem in NLP. The RBT3 semantic model is loaded and encapsulated into a server through the bert-as-service service framework. The client can directly send a POST request to the server to obtain the text vector of the work order after text conversion, which is used for subsequent work order similarity calculation.

[0034] In step S103, the similarity between any two work orders out of N work orders is calculated based on tags and keywords. Each work order corresponds to N similarity scores.

[0035] Specifically, we can first calculate the tag similarity, business keyword similarity, and question keyword similarity between any two work orders. Then, based on the business keyword similarity and question keyword similarity, we can calculate the work order keyword similarity between any two work orders. Finally, we can calculate the overall similarity between any two work orders based on their work order keyword similarity and tag similarity. Each work order is compared to itself and to all other work orders. A work order has a similarity of 1 compared to itself; therefore, each work order has N similarity scores.

[0036] In step S104, each work order is classified according to the N similarities corresponding to each work order, and the classification results of N work orders are obtained.

[0037] Specifically, after calculating the similarity between any two work orders, each work order corresponds to N similarity scores. Based on the N similarity scores for each work order, an N-by-N similarity matrix is ​​constructed. Then, the N work orders are clustered using a work order homology clustering method, thereby grouping work orders with high similarity into the same category, and obtaining the classification results for the N work orders.

[0038] The technical solution disclosed in this application involves acquiring N work orders to be processed via an electronic device, traversing the N work orders, and extracting tags and keywords contained in the N work orders based on a pre-built work order tag library and a work order keyword library. The work order tag library includes tags for different work orders, and the work order keyword library includes keywords for different work orders. The similarity between any two work orders among the N work orders is calculated based on the tags and keywords. Each work order is classified based on the N similarity scores corresponding to each work order, resulting in the classification results for the N work orders. This allows the electronic device to automatically classify the N work orders after acquiring them, without requiring human intervention, thus improving the processing efficiency and accuracy of the classification results.

[0039] For example, such as Figure 2 As shown in the figure, this application embodiment provides a method for processing work orders. The execution subject of this method can be a server, wherein the server can be an independent server or a server cluster composed of multiple servers, and the server can be a server capable of processing work orders. The work order processing method specifically includes the following steps:

[0040] In step S201, N work orders to be processed are obtained, where N is a natural number not less than 2.

[0041] In step S202, N work orders are traversed, and the tags and keywords contained in the N work orders are extracted based on the pre-built work order tag library and work order keyword library. The work order tag library includes tags of different work orders, and the work order keyword library includes keywords of different work orders.

[0042] In step S203, the tag set of each work order in any two work orders is determined, and the first weight of each tag in the tag set is determined. The tag similarity of any two work orders is calculated based on the tags in the tag set and the first weight. The business keyword similarity and question keyword similarity of any two work orders are calculated based on the Jaccard similarity coefficient. The keyword similarity of any two work orders is calculated based on the second weight, the third weight, the business keyword similarity, and the question keyword similarity. The similarity of any two work orders is calculated based on the classification weight of the tags, the classification weight of the keywords, the tag similarity, and the keyword similarity.

[0043] Specifically, for each work order, there may be multiple tags. Based on the work order tag library, the tags contained in each work order are extracted from the work order content to form a tag set. For each tag in the tag set, the weight of each tag is determined, and then the tag similarity between any two work orders is calculated using the following formula:

[0044]

[0045] Among them, w i Let x be the first weight of the i-th label in the label set. i Let x be the similarity score of the i-th label in the label set. If the labels in the two label sets are the same, x is the similarity score. i The value is 1; if the labels in the two label sets are different, x... i The value is 0. Thus, the similarity of each tag in the tag sets of the two work orders is summed sequentially to obtain the similarity of the tags of the two work orders.

[0046] After extracting the business keywords and question keywords from any two work orders, keyword set A and keyword set B are formed for each work order. Keyword set A includes the business keyword set and question keyword set of one of the work orders, and keyword set B includes the business keyword set and question keyword set of the other work order. The similarity of business keywords and question keywords between any two work orders is calculated based on the Jaccard similarity coefficient using the following formula:

[0047]

[0048] Wherein, when keyword set A and keyword set B are business keyword sets, J(A,B) is the similarity of any two work order business keywords; when keyword set A and keyword set B are problem keyword sets, J(A,B) is the similarity of any two work order problem keywords.

[0049] After obtaining the business keyword similarity and question keyword similarity of any two work orders, the keyword similarity between any two work orders is calculated based on the second weight of the business keywords and the third weight of the question keywords in the two work orders. The keyword similarity is calculated as follows:

[0050] keywords_similarity=w1*sim buiness +w2sim question

[0051] Where w1 is the second weight, w2 is the third weight, and sim buiness For the similarity of business keywords, sim question The similarity score is the similarity of the keywords in the question, calculated based on J(A,B).

[0052] Finally, based on the tag similarity and keyword similarity calculated above, the similarity final_similarity between any two work orders is calculated, as follows:

[0053] final_similarity = w label *label_similarity+w keywords *keywords_similarity

[0054] Among them, w label The classification weight of the work order's tags, w keywords The category weight of keywords for work orders.

[0055] In step S204, each work order is classified according to the N similarities corresponding to each work order, and the classification results of N work orders are obtained.

[0056] It is worth noting that steps S201, S202 and S204 have the same or similar implementation as steps S101, S102 and S104 in the above embodiments, and they can be referred to each other. The embodiments of this application will not be described again here.

[0057] As can be seen from the technical solutions provided in the embodiments of this application above, after the electronic device acquires N work orders to be processed, it can automatically classify and process the N work orders without human intervention, thereby improving the processing efficiency and the accuracy of the classification results. Furthermore, calculating the similarity between any two work orders by using the tag similarity and keyword similarity can improve the accuracy of the calculated similarity, thus improving the accuracy of the classification results for the N work orders.

[0058] For example, such as Figure 3 As shown in the figure, this application embodiment provides a work order processing method. The execution subject of this method can be a server, wherein the server can be an independent server or a server cluster composed of multiple servers. Moreover, the server can be a server capable of identifying encrypted services. The work order processing method specifically includes the following steps:

[0059] In step S301, N work orders to be processed are obtained, where N is a natural number not less than 2.

[0060] In step S302, N work orders are traversed, and the tags and keywords contained in the N work orders are extracted based on the pre-built work order tag library and work order keyword library. The work order tag library includes tags of different work orders, and the work order keyword library includes keywords of different work orders.

[0061] In step S303, the similarity between any two work orders among the N work orders is calculated based on the tags and keywords, where each work order corresponds to N similarity scores.

[0062] In step S304, a similarity matrix of N work orders is constructed. The value of the similarity matrix is ​​the similarity of the N work orders, and the similarity matrix is ​​a symmetric matrix. The N work orders are classified based on the homology clustering algorithm to obtain the classification result.

[0063] Specifically, the homogeneous clustering algorithm extends and merges work orders one by one according to the proximity principle. Finally, among all possible homogeneous partitioning results, it finds all homogeneous work orders that meet the homogeneous partitioning criteria to complete the clustering. After obtaining the similarity of N work orders, each work order corresponds to N similarity scores, and an N-by-N matrix is ​​constructed. The values ​​in the matrix are the similarity between work orders calculated in the above embodiment. The specific similarity matrix is ​​as follows:

[0064]

[0065] The similarity matrix is ​​a symmetric matrix, and the diagonal elements are all 1s (meaning the similarity between the work order in that row and itself is 1). S 1,2 To S N,3 The similarity between two work orders.

[0066] After obtaining the similarity matrix, each row represents the similarity between the work order in that row and all other work orders, including the work order itself. Each work order has N similarity scores. Then, based on a preset threshold, a set of work orders that are relatively close to the work orders in that row is selected, meaning work orders with high similarity scores are identified. Alternatively, when using distance as the criterion, work orders that are close to the work orders in that row are selected based on the preset threshold. This means that work orders within a circle centered on the work order in that row (the source work order) and with a radius equal to the preset threshold are considered similar work orders and can be clustered together. Or, when using similarity as the criterion, work orders with a similarity score at least as high as the preset threshold are selected based on the preset threshold and clustered together.

[0067] In one possible implementation, classifying N work orders based on a homology clustering algorithm includes: using each work order in a row of the similarity matrix as the center and a preset threshold as the radius, selecting work orders in the same row as the center work order with a similarity not less than the preset threshold as similar work orders, clustering these similar work orders into one class, and selecting the class with the most similar work orders from the resulting N clusters as a clustering result. This iterative process is repeated until the similarity of the clustered similar work orders is less than the preset threshold. The iterative process includes constructing a new similarity matrix from the similarity of the remaining un-clustered work orders, sequentially using each work order in the new similarity matrix as the center and a preset threshold as the radius, selecting work orders in the same row as the center work order with a similarity not less than the preset threshold as similar work orders, clustering these similar work orders into one class, and selecting the class with the most similar work orders from the resulting new clusters as a clustering result.

[0068] Specifically, a threshold is applied to the similarity scores in each row of the similarity matrix, resulting in N clusters centered on each work order. The number of clusters will vary depending on the cluster center. The work orders with the most similar work orders in the N clusters are sorted, and the cluster with the most work orders is selected as the first cluster. Then, the work orders that have already been clustered are removed from the similarity matrix, resulting in a new similarity matrix. For example, after completing the first cluster in an NxN similarity matrix, the work orders that have been clustered (including rows and columns) are removed. Assuming there are m work orders in the first cluster, the new similarity matrix becomes (Nm)x(Nm). A threshold is then applied to the similarity scores in each row of this new similarity matrix, resulting in Nm clusters centered on each work order in the new similarity matrix. The number of clusters will vary depending on the cluster center. Sort the number of similar work orders in the Nm clustering results, select the cluster with the most work order data as the second cluster, and so on, until the number of similar work orders in the clustering results is less than the preset threshold.

[0069] The technical solution disclosed in this application allows for the automatic classification and processing of N work orders after they are acquired by the electronic device, eliminating the need for manual intervention and improving processing efficiency and classification accuracy. Furthermore, this application utilizes work order tags and keywords to define "same-source work orders," meaning that all work orders within the same source group share similar tags, business keywords, and problem keywords, with the consistency of these three elements equal to a preset threshold distance, further enhancing the accuracy of work order classification.

[0070] Corresponding to the work order processing method provided in the above embodiments, based on the same technical concept, this application also provides a work order processing apparatus. Figure 4 This application provides a schematic diagram of the module composition of a work order processing device, which is used to execute... Figures 1 to 3 The described work order processing method, such as Figure 4 The work order processing device 400 includes: an acquisition module 401, used to acquire N work orders to be processed, where N is a natural number not less than 2; an extraction module 402, used to traverse the N work orders and extract the tags and keywords contained in the N work orders based on a pre-built work order tag library and a work order keyword library, where the work order tag library includes tags of different work orders and the work order keyword library includes keywords of different work orders; a calculation module 403, used to calculate the similarity between any two work orders among the N work orders based on the tags and keywords, where each work order corresponds to N similarities; and a processing module 404, used to classify each work order based on the N similarities corresponding to each work order to obtain the classification results of the N work orders.

[0071] As can be seen from the technical solutions provided in the above embodiments of this application, after the electronic device obtains N work orders to be processed, the electronic device can automatically classify and process the N work orders without human intervention, thereby improving the processing efficiency of work orders and the accuracy of classification results.

[0072] In one possible implementation, the tags have a first weight corresponding to the tag level, the keywords include business keywords and question keywords, the business keywords have a preset second weight, and the question keywords have a preset third weight. The calculation module 403 is also used to determine the tag set of each work order in any two work orders, and the first weight of each tag in the tag set; calculate the tag similarity of any two work orders based on the tags in the tag set and the first weight; calculate the business keyword similarity and question keyword similarity of any two work orders based on the Jaccard similarity coefficient; calculate the keyword similarity of any two work orders based on the second weight, the third weight, the business keyword similarity, and the question keyword similarity; and calculate the similarity of any two work orders based on the classification weight of the work order's tags, the classification weight of the work order's keywords, the tag similarity, and the keyword similarity.

[0073] In one possible implementation, the processing module 404 is further used to construct a similarity matrix of N work orders. The value of the similarity matrix is ​​the similarity of the N work orders. The similarity matrix is ​​a symmetric matrix, and each row of the similarity matrix corresponds to all the similarities of a work order. The N work orders are classified based on the homology clustering algorithm to obtain the classification result.

[0074] In one possible implementation, the processing module 404 is further configured to, using each work order in each row of the similarity matrix as the center and a preset threshold as the radius, select work orders in the similarity matrix whose similarity to the work order in the row containing the center work order as similar work orders, and cluster the similar work orders in the row containing the center work order into one class. From the N clusters obtained, the class with the most similar work orders is selected as a clustering result. This process is repeated until the similarity of the clustered similar work orders is less than the preset threshold. The process includes: constructing a new similarity matrix from the similarity of the remaining un-clustered work orders; sequentially using each work order in the new similarity matrix as the center and a preset threshold as the radius, selecting work orders in the new similarity matrix whose similarity to the work order in the row containing the center work order as similar work orders, and clustering the similar work orders in the row containing the center work order into one class. From the new clusters obtained, the class with the most similar work orders is selected as a clustering result.

[0075] In one possible implementation, the processing device 400 further includes a tagging module, used to tag the level of the tags contained in the extracted N work orders, and to tag the position of the keywords in the work order content of the corresponding work order.

[0076] The work order processing device provided in this application embodiment can implement the various processes in the embodiments corresponding to the above work order processing method. To avoid repetition, it will not be described again here.

[0077] It should be noted that the work order processing device and the work order processing method provided in this application embodiment are based on the same application concept. Therefore, the specific implementation of this embodiment can refer to the implementation of the aforementioned work order processing method, and the repeated parts will not be described again.

[0078] Corresponding to the work order processing method provided in the above embodiments, based on the same technical concept, this application also provides an electronic device for executing the above-described work order processing method. Figure 5 To illustrate the structure of an electronic device according to various embodiments of this application, as shown in the following diagrams... Figure 5 As shown. Electronic devices can vary considerably due to differences in configuration or performance, and may include one or more processors 501 and memory 502. Memory 502 may store one or more application programs or data. Memory 502 may be temporary or persistent storage. The application programs stored in memory 502 may include one or more modules (not shown), and each module may include a series of computer-executable instructions for the electronic device.

[0079] Furthermore, the processor 501 may be configured to communicate with the memory 502 and execute a series of computer-executable instructions stored in the memory 502 on the electronic device. The electronic device may also include one or more power supplies 503, one or more wired or wireless network interfaces 504, one or more input / output interfaces 505, and one or more keyboards 506.

[0080] Specifically, in this embodiment, the electronic device includes a processor, a communication interface, a memory, and a communication bus; wherein, the processor, the communication interface, and the memory communicate with each other through the bus; the memory is used to store computer programs; the processor is used to execute the programs stored in the memory to implement the various steps in the above method embodiments and has the beneficial effects of the above method embodiments. To avoid repetition, the embodiments of this application will not be described again here.

[0081] This embodiment also provides a computer-readable storage medium storing a computer program. When the computer program is executed by a processor, it implements the steps in the above method embodiments and has the beneficial effects of the above method embodiments. To avoid repetition, the embodiments of this application will not be described again here.

[0082] Those skilled in the art will understand that embodiments of this application can be provided as methods, apparatus, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied 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.

[0083] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0084] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work 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 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0085] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0086] In a typical configuration, an electronic device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0087] Memory may include non-persistent storage in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM. Memory is an example of computer-readable media.

[0088] Computer-readable media include both permanent and non-permanent, removable and non-removable media that can store information by any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic magnetic disk storage or other magnetic storage devices, or any other non-transferable medium that can be used to store information accessible by a computing device. As defined herein, computer-readable media does not include transient computer-readable media, such as modulated data signals and carrier waves.

[0089] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.

[0090] Those skilled in the art will understand that embodiments of this application can be provided as methods, apparatus, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied 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.

[0091] The above are merely embodiments of this application and are not intended to limit the scope of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of the claims of this application.

Claims

1. A method for processing work orders, characterized in that, Applied to electronic devices, the processing method includes: Obtain N work orders to be processed, where N is a natural number not less than 2; Iterate through N work orders, and extract the tags and keywords contained in the N work orders based on the pre-built work order tag library and work order keyword library respectively. The work order tag library includes tags of different work orders, and the work order keyword library includes keywords of different work orders. The similarity between any two work orders among the N work orders is calculated based on the tags and keywords, where each work order corresponds to N similarity scores. The work orders are classified according to the N similarity scores corresponding to each work order, resulting in the classification results of the N work orders. The keywords include business keywords and problem keywords. The step of calculating the similarity between any two work orders among the N work orders based on the tags and the keywords includes: Calculate the similarity of tags, business keywords, and question keywords between any two work orders; Calculate the keyword similarity between any two work orders based on the similarity of their business keywords and the similarity of their question keywords; Calculate the similarity between any two work orders based on their keyword similarity and tag similarity. The step of classifying each work order based on N similarities to obtain N classification results for each work order includes: Construct a similarity matrix for N work orders, where the value of the similarity matrix is ​​the similarity of the N work orders. The similarity matrix is ​​a symmetric matrix, and each row of the similarity matrix corresponds to all the similarities of a work order. Using the work orders in each row of the similarity matrix as the center and a preset threshold as the radius, select work orders in the similarity matrix whose similarity in the row containing the work order as the center is not less than the preset threshold as similar work orders, and cluster the similar work orders in the row containing the work order as the center into one class. Select the class with the most similar work orders from the N clusters as a clustering result, and perform the loop process until the similarity of the clustered similar work orders is less than the preset threshold. The cyclic process includes: constructing a new similarity matrix from the similarity of the remaining unclustered work orders; taking each work order in the new similarity matrix as the center and the preset threshold as the radius, selecting work orders in the new similarity matrix whose similarity in the row of the work order with the center is not less than the preset threshold as similar work orders; clustering the similar work orders in the row of the work order with the center into one class; and selecting the class with the most similar work orders from the new clusters as a clustering result.

2. The processing method according to claim 1, characterized in that, The tags are assigned a first weight corresponding to their levels, the business keywords are assigned a second weight, and the question keywords are assigned a third weight. The calculation of the similarity between any two work orders out of N work orders based on the tags and keywords includes: Determine the tag set for each work order in any two work orders, and the first weight for each tag in the tag set; calculate the tag similarity between any two work orders based on the tags in the tag set and the first weight. The similarity of business keywords and question keywords between any two work orders is calculated based on the Jaccard similarity coefficient. The similarity of keywords between any two work orders is calculated based on the second weight, the third weight, the similarity of business keywords, and the similarity of question keywords. The similarity between any two work orders is calculated based on the classification weight of the work order's tags, the classification weight of the work order's keywords, the tag similarity, and the keyword similarity.

3. The processing method according to claim 1, characterized in that, After extracting the tags and keywords contained in N work orders based on the pre-built work order tag library and work order keyword library, the method further includes: The level of the tags contained in the N extracted work orders is marked, and the position of the keyword in the work order content of the corresponding work order is marked.

4. A work order processing device, characterized in that, The processing device includes: The acquisition module is used to acquire N work orders to be processed, where N is a natural number not less than 2; An extraction module is used to traverse N work orders and extract the tags and keywords contained in the N work orders based on a pre-built work order tag library and a work order keyword library, respectively. The work order tag library includes tags of different work orders, and the work order keyword library includes keywords of different work orders. The calculation module is used to calculate the similarity between any two work orders among the N work orders based on the tags and keywords, wherein each work order corresponds to N similarity scores; The processing module is used to classify each work order according to N similarities corresponding to each work order, and obtain the classification results of N work orders; The keywords include business keywords and problem keywords. The calculation module is specifically used to calculate the tag similarity, business keyword similarity, and problem keyword similarity of any two work orders. Calculate the keyword similarity between any two work orders based on the similarity of their business keywords and the similarity of their question keywords; Finally, the similarity between any two work orders is calculated based on the similarity of their work order keywords and tags. The processing module is further configured to: Construct a similarity matrix for N work orders, where the value of the similarity matrix is ​​the similarity of the N work orders. The similarity matrix is ​​a symmetric matrix, and each row of the similarity matrix corresponds to all the similarities of a work order. Using the work orders in each row of the similarity matrix as the center and a preset threshold as the radius, select work orders in the similarity matrix whose similarity in the row containing the center work order is not less than the preset threshold as similar work orders. Cluster the similar work orders in the row containing the center work order into one class. Select the class with the most similar work orders from the N clusters as a clustering result. Perform a loop process until the similarity of the clustered similar work orders is less than the preset threshold. The cyclic process includes: constructing a new similarity matrix from the similarity of the remaining unclustered work orders; taking each work order in the new similarity matrix as the center and the preset threshold as the radius, selecting work orders in the new similarity matrix whose similarity in the row of the work order with the center is not less than the preset threshold as similar work orders; clustering the similar work orders in the row of the work order with the center into one class; and selecting the class with the most similar work orders from the new clusters as a clustering result.

5. The processing apparatus according to claim 4, characterized in that, The tag has a first weight corresponding to the tag's level, the business keyword has a preset second weight, and the question keyword has a preset third weight. The calculation module is also used for: Determine the tag set for each work order in any two work orders, and the first weight for each tag in the tag set; calculate the tag similarity between any two work orders based on the tags in the tag set and the first weight. The similarity of business keywords and question keywords between any two work orders is calculated based on the Jaccard similarity coefficient. The keyword similarity between any two work orders is calculated based on the second weight, the third weight, the similarity of business keywords, and the similarity of question keywords. The similarity between any two work orders is calculated based on the classification weight of the work order's tags, the classification weight of the work order's keywords, the similarity of tags, and the similarity of keywords.

6. An electronic device, characterized in that, It includes a processor, a communication interface, a memory, and a communication bus; wherein the processor, the communication interface, and the memory communicate with each other through the communication bus; the memory is used to store computer programs; the processor is used to execute the programs stored in the memory to implement the work order processing method steps as described in any one of claims 1-3.

7. A computer-readable storage medium storing a computer program, wherein when executed by a processor, the computer program implements the steps of the work order processing method as described in any one of claims 1-3.

Images