Work order clustering method, device, electronic device, and storage medium

Abstract

The present application provides a work order clustering method, device, electronic device, and storage medium. The method includes: obtaining a system fault work order and a system component entity; arranging and combining the keywords in the system fault work order and the system component entity to obtain a number of keyword groups; filtering a limited number of words in the system fault work order; matching the filtered system fault work order with a number of the keyword groups to obtain a fault work order group corresponding to the filtered system fault work order; clustering a number of filtered system fault work orders in the fault work order group. The work order clustering method provided in the embodiment of the present application can pre-group the system fault work orders, and then accurately cluster a number of filtered system fault work orders in the fault work order group, which can effectively improve the accuracy of processing system fault work orders, so as to improve the accuracy of work order clustering.

Description

Technical Field

[0001] The present application relates to the field of natural language processing technology, and in particular to a work order clustering method, device, electronic device, and non-transitory computer-readable storage medium. Background Art

[0002] Currently, there are relatively few intelligent methods for processing system fault tickets, but the demand for intelligent processing of system fault tickets is increasing day by day. Existing technical solutions are mostly based on the tf-idf (Term Frequency-Inverse Document Frequency) algorithm and cosine similarity calculation to directly cluster text data such as work orders. However, this technical solution involves multiple statistical methods, which are computationally complex and only applicable to situations with relatively small amounts of data. Otherwise, the accuracy of work order clustering cannot be guaranteed. As operators' IT systems grow, the data of system fault tickets will accumulate to a very large amount. If the data is still clustered directly using existing technical solutions, the clustering process will become quite complicated and cause a large amount of resource consumption, resulting in the inability to guarantee the real-time and accuracy of work order processing. Summary of the Invention

[0003] The embodiment of the present application provides a work order clustering method to solve the technical problem of low accuracy of existing work order clustering methods.

[0004] In a first aspect, an embodiment of the present application provides a work order clustering method, comprising:

[0005] Get system fault tickets and system component entities;

[0006] Arrange and combine the keywords in the system fault work order and the system component entities to obtain a plurality of keyword groups;

[0007] Filtering a limited number of words from the system fault work order;

[0008] Matching the filtered system fault work orders with the plurality of keyword groups to obtain a fault work order group corresponding to the filtered system fault work orders;

[0009] Clustering is performed on a number of filtered system fault work orders in the fault work order group.

[0010] In one embodiment, the keywords in the system fault work order and the system component entities are arranged and combined to obtain several keyword groups, including:

[0011] Performing text segmentation on the system fault work order and marking words in the system fault work order whose part of speech is adjective;

[0012] The words with the adjective part of speech in the system fault work order and the system component entity are arranged and combined to obtain a plurality of keyword groups.

[0013] In one embodiment, filtering out a limited number of words from the system fault work order includes:

[0014] Obtaining the weight of each word in the system fault work order;

[0015] Based on the weight of each word in the system fault work order, obtaining the probability of each word in the system fault work order being filtered out;

[0016] According to the probability of each word in the system fault work order being filtered out, a limited number of words in the system fault work order are filtered out.

[0017] In one embodiment, the weight of each word in the system fault work order is obtained as follows:

[0018] The weight of each word in the system fault work order is obtained according to the weight calculation formula, and the weight calculation formula is:

[0019]

[0020] Among them, w i represents the weight of the i-th word in the system fault work order, min(w) represents the weight of the word with the smallest weight in the system fault work order; max(w) represents the weight of the word with the largest weight in the system fault work order.

[0021] In one embodiment, the probability of each word in the system fault work order being filtered out is obtained based on the weight of each word in the system fault work order, specifically:

[0022] Based on the weight of each word in the system fault work order, the probability of each word in the system fault work order being filtered out is obtained according to the probability calculation formula, and the probability calculation formula is:

[0023]

[0024] Among them, p i represents the probability of the i-th word being filtered out in the system fault work order, w i represents the weight of the i-th word in the system fault work order, and w represents the weight of all words in the system fault work order.

[0025] In one embodiment, the filtered system fault work tickets are matched with the plurality of keyword groups to obtain a fault work ticket group corresponding to the filtered system fault work tickets, including:

[0026] Associating the filtered system fault work order with the first matched keyword group according to the first matched keyword group;

[0027] Vectorizing the keyword groups associated with the filtered system fault work orders;

[0028] Clustering is performed on a number of vectorized keyword groups to obtain a number of fault ticket groups, where the fault ticket group to which the keyword group belongs is a fault ticket group corresponding to the filtered system fault tickets associated with the keyword group.

[0029] In one embodiment, clustering the filtered system fault tickets in the fault ticket group includes:

[0030] Using the word vector model, we can obtain the vectors of the filtered system fault work orders based on the filtered system fault work orders.

[0031] Clustering is performed on vectors of several filtered system fault work orders in the fault work order group.

[0032] In a second aspect, an embodiment of the present application provides a work order clustering device, comprising:

[0033] The acquisition module is used to obtain system fault work orders and system component entities;

[0034] a permutation and combination module, configured to permutate and combine the keywords in the system fault work order and the system component entities to obtain a plurality of keyword groups;

[0035] A filtering module, configured to: filter out a limited number of words from the system fault work order;

[0036] A matching module is used to match the filtered system fault work orders with the plurality of keyword groups to obtain a fault work order group corresponding to the filtered system fault work orders;

[0037] The clustering module is used to cluster the filtered system fault work orders in the fault work order group.

[0038] In a third aspect, an embodiment of the present application provides an electronic device, comprising a processor and a memory storing a computer program, wherein when the processor executes the program, the work order clustering method described in the first aspect or the second aspect is implemented.

[0039] In a fourth aspect, an embodiment of the present application provides a computer program product, including a computer program, which, when executed by a processor, implements the work order clustering method described in the first aspect or the second aspect.

[0040] The work order clustering method provided in the embodiment of the present application obtains several keyword groups by arranging and combining the keywords and system component entities in the system fault work order, and then matches the filtered system fault work order with the several keyword groups to obtain a fault work order group corresponding to the filtered system fault work order. The system fault work order can be grouped in advance, and then several filtered system fault work orders in the fault work order group can be accurately clustered, which can effectively improve the accuracy of processing system fault work orders, thereby improving the accuracy of work order clustering, and filtering a limited number of words in the system fault work order can prevent overfitting in the work order clustering process, thereby ensuring the smooth progress of the clustering process. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] In order to more clearly illustrate the technical solutions in the present application or the prior art, a brief introduction will be given below to the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0042] Figure 1 This is a flow chart of the work order clustering method provided in an embodiment of the present application;

[0043] Figure 2 This is a schematic diagram of the structure of the work order clustering device provided in an embodiment of the present application;

[0044] Figure 3 It is a structural diagram of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0045] To make the objectives, technical solutions, and advantages of this application more clear, the technical solutions in this application will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of this application. Obviously, the described embodiments are only part of the embodiments of this application, not all of them. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making any creative efforts are within the scope of protection of this application.

[0046] Figure 1 A flowchart of a work order clustering method is provided for an embodiment of the present application.

[0047] Reference Figure 1 , an embodiment of the present application provides a work order clustering method, which may include:

[0048] S110. Obtain system fault work order and system component entity.

[0049] S120: Arrange and combine the keywords in the system fault work order and the system component entities to obtain a plurality of keyword groups.

[0050] S130: Filter out a limited number of words from the system fault work order.

[0051] S140: Match the filtered system fault work orders with the plurality of keyword groups to obtain a fault work order group corresponding to the filtered system fault work orders.

[0052] S150: Clustering the filtered system fault work orders in the fault work order group.

[0053] It should be noted that the execution entity of the work order clustering method provided by the present invention can be any terminal-side device, such as a work order clustering system, etc.

[0054] In step S110, the terminal side device obtains the system fault work order and the system component entity.

[0055] It should be noted that system fault work orders are usually stored in a specific storage module, and the terminal side device can obtain the system fault work order from the storage module used to store the system fault work order. In general, the system fault work order usually contains the name of the faulty system or its fault description. The terminal side device can obtain the name of the system component from the enterprise CMDB (Configuration Management Database) database as a system component entity, such as database, memory, CPU, domain name, etc. In this embodiment, the system component entity can be obtained according to any rule, or other descriptions of the system component can be extracted as the system component entity, such as database capacity, CPU usage, etc. After obtaining the system fault work order and the system component entity, the terminal side device can store them to prepare for the representation of subsequent keyword groups.

[0056] In one embodiment, after obtaining a system fault work order, the terminal side device can first perform text cleaning on the system fault work order. During the text cleaning process, the terminal side device can remove special characters in the system fault work order, such as punctuation marks, space characters, etc., by using regular expressions, and replace numbers, dates, or longer non-Chinese character strings in the system fault work order with special words (such as: using number instead of numbers in the system fault work order, using date instead of dates in the system fault work order, using jobnumber instead of network numbers in the system fault work order, etc.), so that the content of the cleaned system fault work order is clearer and simpler.

[0057] In step S120, the terminal side device arranges and combines the keywords in the system fault work order and the system component entities to obtain a plurality of keyword groups.

[0058] It should be noted that the keywords in the system fault work order can be obtained in advance from the system fault work order by the terminal side device, such as adjectives or nouns in the system fault work order, and the keyword regulations can be set according to actual conditions. The keywords and system component entities in the system fault work order can be arranged and combined, for example, there are N keywords and M system component entities, then there are K=N*M keyword groups. Combining the keywords in the system fault work order and the system component entities as the representation of the keyword group can more three-dimensionally reflect the content of the system fault work order.

[0059] In one embodiment, step S120 may include:

[0060] Performing text segmentation on the system fault work order and marking words in the system fault work order whose part of speech is adjective;

[0061] The words with the adjective part of speech in the system fault work order and the system component entity are arranged and combined to obtain a plurality of keyword groups.

[0062] It should be noted that the terminal side device can perform text segmentation on the system fault work order after text cleaning. Specifically, the jieba segmentation tool (jieba segmentation tool) can be used to segment the system fault work order after text cleaning, and mark the part of speech, and focus on marking the words with adjective parts of speech. The accuracy of jieba segmentation is very high. The words with adjective parts of speech in the system fault work order are arranged and combined with the system component entities, and the keyword groups obtained can be, for example: [memory, increased], [cpu usage, abnormal], [ip1.1.1.101, not connected], etc. By arranging and combining the words with adjective parts of speech and the system component entities in the system fault work order, a keyword group that is closer to the fault content described in the system fault work order can be obtained, which is conducive to the subsequent matching of the system fault work order and the keyword group, so as to accurately group a large number of system fault work orders. After obtaining the keyword group, the terminal side device can first store the keyword group for subsequent use.

[0063] In step S130, the terminal side device will filter out a limited number of words in the system fault work order.

[0064] It's important to note that a large number of system troubleshooting tickets can easily lead to overfitting when clustering them on the terminal side. Filtering out a limited number of terms in system troubleshooting tickets can effectively prevent overfitting. Furthermore, system troubleshooting tickets may contain terms that are not useful for ticket clustering, such as ticket numbers. Filtering out these terms significantly reduces the complexity of ticket clustering and improves its accuracy.

[0065] In one embodiment, step S130 may include:

[0066] Obtaining the weight of each word in the system fault work order;

[0067] Based on the weight of each word in the system fault work order, obtaining the probability of each word in the system fault work order being filtered out;

[0068] According to the probability of each word in the system fault work order being filtered out, a limited number of words in the system fault work order are filtered out.

[0069] It should be noted that the terminal-side device can obtain the weight of each word in the system fault work order based on the correlation between each word and the fault description in the system fault work order, and then obtain the probability of each word in the system fault work order being filtered out based on the weight of each word in the system fault work order. For example, if a word has a smaller weight, then it has a higher probability of being filtered out than a word with a larger weight. In this way, the terminal-side device can sort each word in the system fault work order based on the probability of each word being filtered out. The N words with the highest probability of being filtered out are the words that need to be filtered out. In this embodiment, N can be 5.

[0070] Based on the weight of each word in the system fault work order, the probability of each word in the system fault work order being filtered out is obtained. Then, based on the probability of each word in the system fault work order being filtered out, a limited number of words in the system fault work order are filtered out. This can effectively improve the accuracy of the system fault work order, reduce the complexity of processing the system fault work order, and avoid resource consumption caused by clustering meaningless data of the system fault work order.

[0071] In step S140, the terminal side device matches the filtered system fault work order with the plurality of keyword groups to obtain a fault work order group corresponding to the filtered system fault work order.

[0072] It should be noted that some words still exist in the filtered system fault tickets. The filtered system fault tickets are matched with several keyword groups. The keyword group that is first matched for each filtered system fault ticket is used as the matching result for the filtered system fault ticket. All system fault tickets that match the same keyword group can be grouped into one fault ticket group. Alternatively, all system fault tickets that match the same keyword group can be associated with the keyword group first, and then several keyword groups can be clustered. The class to which the keyword group belongs is the fault ticket group corresponding to the system fault tickets associated with the keyword group. Pre-grouping the filtered system fault tickets can effectively improve the accuracy of ticket clustering.

[0073] In one embodiment, step S140 may include:

[0074] Associating the filtered system fault work order with the first matched keyword group according to the first matched keyword group;

[0075] Vectorizing the keyword groups associated with the filtered system fault work orders;

[0076] Clustering is performed on a number of vectorized keyword groups to obtain a number of fault ticket groups, where the fault ticket group to which the keyword group belongs is a fault ticket group corresponding to the filtered system fault tickets associated with the keyword group.

[0077] It should be noted that a filtered system fault ticket may match multiple keyword groups, then the keyword group that the filtered system fault ticket matches for the first time will be used as the matching result of the filtered system fault ticket. Multiple system fault tickets may all match the same keyword group, then these system fault tickets can be associated with the keyword group, and then the vectors of several keyword groups are clustered, and the class where the keyword group belongs is the fault ticket group corresponding to the system fault ticket associated with the keyword group. For example, the keyword group is [memory, increase], and the filtered system fault ticket also has the two words memory and increase, and the weights of these two words are higher. The first keyword group that matches is [memory, increase], then the filtered system fault ticket can be associated with the keyword group [memory, increase]. By matching each filtered system fault ticket in this way, the pre-grouping of the filtered system fault tickets can be completed. On the other hand, the filtered system fault work tickets may not match any keyword groups. In this case, the terminal-side device may group the system fault work tickets that do not match any keyword groups into one fault work ticket group.

[0078] It should be noted that the terminal-side device can vectorize the keyword groups associated with the filtered system fault work orders through the word2vec dictionary in the prior art, and then cluster the vectors of the keyword groups using the prior art K-means algorithm (a clustering algorithm). Specifically, the clusters can be set to C categories. In this embodiment, C can be set to 10, and the filtered system fault work orders will be divided into 11 fault work order groups (the 10 clustered fault work order groups and the fault work order groups corresponding to the system fault work orders that do not match all keyword groups). Then the embodiment of the present application pre-groups all system fault work orders, greatly improving the clustering precision and clustering accuracy of the system fault work orders.

[0079] It should be noted that, in the actual deployment process, step S140 can be performed in parallel by computers, which can greatly reduce the time for subsequent clustering and improve efficiency on ultra-large-scale data volumes.

[0080] In step S150, the terminal side device clusters the filtered system fault work orders in the fault work order group.

[0081] It should be noted that the terminal-side device can first calculate the vector of each word in the filtered system fault work order through the vector generation model, and then take the average value of the vector of each word in the filtered system fault work order as the vector of the filtered system fault work order, and then cluster the vectors of the filtered system fault work orders in each fault work order group, which can effectively enhance vectorization and improve clustering accuracy.

[0082] In one embodiment, step S150 may include:

[0083] Using the word vector model, we can obtain the vectors of the filtered system fault work orders based on the filtered system fault work orders.

[0084] Clustering is performed on vectors of several filtered system fault work orders in the fault work order group.

[0085] It should be noted that the word vector model can be any model used for vectorization in the prior art, such as the word2vec model in the prior art. The terminal side device can obtain the vector of the filtered system fault work order based on the remaining words of the filtered system fault work order through the word2vec model, and then use the hierarchical clustering algorithm in the prior art to cluster the vectors of the filtered system fault work order of each fault work order group. The terminal side device can control the number of clustering categories by setting the similarity threshold of the hierarchical clustering algorithm. Assuming that there are m categories in the Nth bucket (fault work order group), the final system fault work order is clustered into A class.

[0086] The work order clustering method provided in the embodiment of the present application obtains several keyword groups by arranging and combining the keywords and system component entities in the system fault work order, and then matches the filtered system fault work order with the several keyword groups to obtain a fault work order group corresponding to the filtered system fault work order. The system fault work order can be grouped in advance, and then several filtered system fault work orders in the fault work order group can be accurately clustered, which can effectively improve the accuracy of processing system fault work orders, thereby improving the accuracy of work order clustering, and filtering a limited number of words in the system fault work order can prevent overfitting in the work order clustering process, thereby ensuring the smooth progress of the clustering process.

[0087] In one embodiment, obtaining the weight of each word in the system fault work order is specifically as follows:

[0088] The weight of each word in the system fault work order is obtained according to the weight calculation formula, and the weight calculation formula is:

[0089]

[0090] Among them, w i represents the weight of the i-th word in the system fault work order, min(w) represents the weight of the word with the smallest weight in the system fault work order; max(w) represents the weight of the word with the largest weight in the system fault work order.

[0091] It should be noted that the terminal side device can first obtain the tfidf weight of each word in the system fault work order (the tfidf weight can be obtained based on the tf-idf algorithm of the existing technology), and then standardize the tfidf weight of each word in the system fault work order through the weight calculation formula, specifically converting it into a value between 0 and 1, so that it becomes the weight of each word in the system fault work order.

[0092] Furthermore, based on the weight of each word in the system fault work order, the probability of each word in the system fault work order being filtered out is obtained, specifically:

[0093] Based on the weight of each word in the system fault work order, the probability of each word in the system fault work order being filtered out is obtained according to the probability calculation formula, and the probability calculation formula is:

[0094]

[0095] Among them, p i represents the probability of the i-th word being filtered out in the system fault work order, w i represents the weight of the i-th word in the system fault work order, and w represents the weight of all words in the system fault work order.

[0096] It should be noted that the terminal-side device first obtains the weight of each word in the system fault work order through a weight calculation formula, and then based on the weight of each word in the system fault work order, obtains the probability of each word in the system fault work order being filtered out according to the probability calculation formula. This can accurately filter out the effective number of words in the system fault work order (filtering is also called dropout), making the words in the filtered system fault work order more representative, which is conducive to improving the subsequent work order clustering efficiency and clustering accuracy.

[0097] The work order clustering device provided in an embodiment of the present application is described below. The work order clustering device described below and the work order clustering method described above can be referenced to each other.

[0098] Figure 2 A schematic structural diagram of a work order clustering device is provided for an embodiment of the present application.

[0099] Reference Figure 2 , an embodiment of the present application provides a work order clustering device, which may include:

[0100] The acquisition module 210 is used to: acquire the system fault work order and the system component entity;

[0101] The permutation and combination module 220 is used to permutate and combine the keywords in the system fault work order and the system component entity to obtain a plurality of keyword groups;

[0102] The filtering module 230 is used to: filter out a limited number of words from the system fault work order;

[0103] The matching module 240 is configured to match the filtered system fault work orders with the plurality of keyword groups to obtain a fault work order group corresponding to the filtered system fault work orders;

[0104] The clustering module 250 is configured to cluster the filtered system fault work orders in the fault work order group.

[0105] In one embodiment, the permutation and combination module 220 includes:

[0106] A word segmentation submodule is used to: perform text segmentation on the system fault work order and mark out words in the system fault work order whose part of speech is adjective;

[0107] The permutation and combination submodule is used to permutate and combine the words with adjective parts of speech in the system fault work order and the system component entities to obtain a plurality of keyword groups.

[0108] In one embodiment, the filtering module 230 includes:

[0109] The weight obtaining submodule is used to obtain the weight of each word in the system fault work order;

[0110] A probability obtaining submodule, configured to obtain a probability of each word in the system fault work order being filtered out based on the weight of each word in the system fault work order;

[0111] The filtering submodule is used to filter out a limited number of words in the system fault work order according to the probability of each word being filtered out in the system fault work order.

[0112] It should be noted that the weight obtaining submodule is specifically used for:

[0113] The weight of each word in the system fault work order is obtained according to the weight calculation formula, and the weight calculation formula is:

[0114]

[0115] Among them, w i represents the weight of the i-th word in the system fault work order, min(w) represents the weight of the word with the smallest weight in the system fault work order; max(w) represents the weight of the word with the largest weight in the system fault work order.

[0116] It should be noted that the probability obtaining submodule is specifically used for:

[0117] Based on the weight of each word in the system fault work order, the probability of each word in the system fault work order being filtered out is obtained according to the probability calculation formula, and the probability calculation formula is:

[0118]

[0119] Among them, p i represents the probability of the i-th word being filtered out in the system fault work order, w i represents the weight of the i-th word in the system fault work order, and w represents the weight of all words in the system fault work order.

[0120] In one embodiment, the matching module 240 includes:

[0121] An associating submodule, configured to: associate the filtered system fault work order with the first matched keyword group according to the first matched keyword group of the filtered system fault work order;

[0122] A vectorization submodule, configured to: vectorize a keyword group associated with the filtered system fault work order;

[0123] The first clustering submodule is used to cluster several vectorized keyword groups to obtain several fault ticket groups, where the fault ticket groups to which the keyword groups belong are fault ticket groups corresponding to the filtered system fault tickets associated with the keyword groups.

[0124] In one embodiment, the clustering module 250 includes:

[0125] The vector acquisition submodule is used to obtain the vectors of the filtered system fault work orders based on the filtered system fault work orders using the word vector model;

[0126] The second clustering submodule is used to cluster the vectors of the filtered system fault work orders in the fault work order group.

[0127] Figure 3 An example of a physical structure diagram of an electronic device is shown below. Figure 3 As shown, the electronic device may include: a processor 810, a communication interface 820, a memory 830, and a communication bus 840, wherein the processor 810, the communication interface 820, and the memory 830 communicate with each other via the communication bus 840. The processor 810 may call a computer program in the memory 830 to execute the steps of the work order clustering method, for example, including:

[0128] Get system fault tickets and system component entities;

[0129] Arrange and combine the keywords in the system fault work order and the system component entities to obtain a plurality of keyword groups;

[0130] Filtering a limited number of words from the system fault work order;

[0131] Matching the filtered system fault work orders with the plurality of keyword groups to obtain a fault work order group corresponding to the filtered system fault work orders;

[0132] Clustering is performed on a number of filtered system fault work orders in the fault work order group.

[0133] In addition, the logic instructions in the above-mentioned memory 830 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when sold or used as an independent product. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0134] On the other hand, embodiments of the present application further provide a computer program product, comprising a computer program. The computer program may be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer may perform the steps of the work order clustering method provided in the above embodiments, for example, including:

[0135] Get system fault tickets and system component entities;

[0136] Arrange and combine the keywords in the system fault work order and the system component entities to obtain a plurality of keyword groups;

[0137] Filtering a limited number of words from the system fault work order;

[0138] Matching the filtered system fault work orders with the plurality of keyword groups to obtain a fault work order group corresponding to the filtered system fault work orders;

[0139] Clustering is performed on a number of filtered system fault work orders in the fault work order group.

[0140] On the other hand, an embodiment of the present application further provides a processor-readable storage medium, wherein the processor-readable storage medium stores a computer program, wherein the computer program is configured to cause a processor to execute the steps of the work order clustering method provided in the above embodiments, for example, including:

[0141] Get system fault tickets and system component entities;

[0142] Arrange and combine the keywords in the system fault work order and the system component entities to obtain a plurality of keyword groups;

[0143] Filtering a limited number of words from the system fault work order;

[0144] Matching the filtered system fault work orders with the plurality of keyword groups to obtain a fault work order group corresponding to the filtered system fault work orders;

[0145] Clustering is performed on a number of filtered system fault work orders in the fault work order group.

[0146] The processor-readable storage medium can be any available medium or data storage device that can be accessed by the processor, including but not limited to magnetic storage (such as floppy disks, hard disks, magnetic tapes, magneto-optical disks (MO)), optical storage (such as CDs, DVDs, BDs, HVDs, etc.), and semiconductor storage (such as ROMs, EPROMs, EEPROMs, non-volatile memories (NANDFLASH), solid-state drives (SSDs)), etc.

[0147] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, i.e., they may be located in one location or distributed across multiple network units. Some or all of the modules may be selected based on actual needs to achieve the objectives of the present embodiment. Persons of ordinary skill in the art will be able to understand and implement the present invention without inventive effort.

[0148] Through the description of the above embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus a necessary general hardware platform, or of course, by hardware. Based on this understanding, the essence of the above technical solution or the part that contributes to the existing technology can be embodied in the form of a software product. The computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a magnetic disk, an optical disk, etc., and includes a number of instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or certain parts of the embodiments.

[0149] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A work order clustering method, characterized in that: include: Get system fault tickets and system component entities; Arrange and combine the keywords in the system fault work order and the system component entities to obtain a plurality of keyword groups; Filtering a limited number of words from the system fault work order; Matching the filtered system fault work orders with the plurality of keyword groups to obtain a fault work order group corresponding to the filtered system fault work orders; Clustering is performed on a number of filtered system fault work orders in the fault work order group.

2. The work order clustering method according to claim 1, characterized in that: The keywords in the system fault work order and the system component entities are arranged and combined to obtain several keyword groups, including: Performing text segmentation on the system fault work order and marking words in the system fault work order whose part of speech is adjective; The words with the adjective part of speech in the system fault work order and the system component entity are arranged and combined to obtain a plurality of keyword groups.

3. The work order clustering method according to claim 2, characterized in that: The filtering of a limited number of words from the system fault work order includes: Obtaining the weight of each word in the system fault work order; Based on the weight of each word in the system fault work order, obtaining the probability of each word in the system fault work order being filtered out; According to the probability of each word in the system fault work order being filtered out, a limited number of words in the system fault work order are filtered out.

4. The work order clustering method according to claim 3, characterized in that: The weight of each word in the system fault work order is obtained as follows: The weight of each word in the system fault work order is obtained according to the weight calculation formula, and the weight calculation formula is: Among them, w i represents the weight of the i-th word in the system fault work order, min(w) represents the weight of the word with the smallest weight in the system fault work order; max(w) represents the weight of the word with the largest weight in the system fault work order.

5. The work order clustering method according to claim 4, characterized in that: The probability of each word in the system fault work order being filtered out is obtained based on the weight of each word in the system fault work order, specifically: Based on the weight of each word in the system fault work order, the probability of each word in the system fault work order being filtered out is obtained according to the probability calculation formula, and the probability calculation formula is: Among them, p i represents the probability of the i-th word being filtered out in the system fault work order, w i represents the weight of the i-th word in the system fault work order, and w represents the weight of all words in the system fault work order.

6. The work order clustering method according to any one of claims 1 to 5, characterized in that: The filtered system fault work orders are matched with the plurality of keyword groups to obtain a fault work order group corresponding to the filtered system fault work orders, including: Associating the filtered system fault work order with the first matched keyword group according to the first matched keyword group; Vectorizing the keyword groups associated with the filtered system fault work orders; Clustering is performed on a number of vectorized keyword groups to obtain a number of fault ticket groups, where the fault ticket group to which the keyword group belongs is a fault ticket group corresponding to the filtered system fault tickets associated with the keyword group.

7. The work order clustering method according to any one of claims 1 to 5, characterized in that: Clustering the filtered system fault work tickets in the fault work ticket group includes: Using the word vector model, we can obtain the vectors of the filtered system fault work orders based on the filtered system fault work orders. Clustering is performed on vectors of several filtered system fault work orders in the fault work order group.

8. A work order clustering device, characterized in that: include: The acquisition module is used to obtain system fault work orders and system component entities; a permutation and combination module, configured to permutate and combine the keywords in the system fault work order and the system component entities to obtain a plurality of keyword groups; A filtering module, configured to: filter out a limited number of words from the system fault work order; A matching module is used to match the filtered system fault work orders with the plurality of keyword groups to obtain a fault work order group corresponding to the filtered system fault work orders; The clustering module is used to cluster the filtered system fault work orders in the fault work order group.

9. An electronic device comprising a processor and a memory storing a computer program, characterized in that: When the processor executes the computer program, the work order clustering method according to any one of claims 1 to 7 is implemented.

10. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the work order clustering method according to any one of claims 1 to 7 is implemented.

Images