Work order dispatching method, system, electronic device and computer readable medium

Abstract

The present application relates to the technical field of dispatching work orders, and provides a work order dispatching method, system, electronic device and computer-readable medium. Among them, the work order dispatching method includes obtaining the work order to be dispatched, and extracting the work order keywords of the work order to be dispatched; according to the work order keywords, matching the work order to be dispatched with multiple historical work orders to obtain the business system information corresponding to the matched historical work orders that meet the matching conditions; according to the obtained business system information corresponding to the historical work orders, confirming the business system that processes the work order to be dispatched. In the above-mentioned work order dispatching method, based on the historical work orders and the business system information corresponding to the historical work orders, multiple dimensions are estimated to be able to process the business system of the work order to be dispatched. Therefore, the above-mentioned method can improve the accuracy of the business system of the work order to be dispatched determined by the work order dispatching system, thereby reducing the error rate of the work order dispatching system, reducing the number of times the work order is returned and re-dispatched, and improving the actual work efficiency of the work order dispatching system.

Description

Technical Field

[0001] The present application relates to the technical field of work order dispatching, and in particular to a work order dispatching method, system, electronic device and computer-readable medium. Background Art

[0002] At present, banking staff may encounter technical and regulatory questions during actual business operations, as shown in Figure 1(a). Then, banking staff will create a work order through the IT service management (ITSM) system and report the work order to the work order distribution system, where the work order carries the system, problem overview, problem details, and interface screenshots filled in by the banking staff, as shown in Figure 1(b) and Figure 1(c). Subsequently, the work order distribution system distributes the work order to the corresponding business system according to the system to which it belongs, and then the relevant personnel in the business system operation and maintenance team answer the work order.

[0003] However, due to the large number of banking business systems, a huge number of work orders are circulated through the ITSM system every day, and many of the questions corresponding to the work orders are relatively similar, resulting in a low accuracy rate in the work order dispatching system. Some work orders even need to be returned and re-dispatched multiple times before they can be transferred to the correct business system, resulting in low actual work efficiency of the work order dispatching system. Summary of the invention

[0004] The purpose of this application is to solve the technical problem that the work order dispatching system has a low accuracy rate in dispatching work orders, and some work orders even need to be returned and re-dispatched multiple times before they can be transferred to the correct business system, resulting in low actual work efficiency of the work order dispatching system. Based on this, this application provides a work order dispatching method, system, electronic device and computer-readable medium.

[0005] The first aspect of the present application provides a work order dispatching method, which includes obtaining the work order to be dispatched, and extracting the work order keywords of the work order to be dispatched; matching the work order to be dispatched with multiple historical work orders based on the work order keywords; and obtaining business system information corresponding to the matched historical work orders that meet the matching conditions; and confirming the business system that processes the work order to be dispatched based on the obtained business system information corresponding to the historical work orders.

[0006] In a possible implementation of the first aspect, the work order to be dispatched is matched with multiple historical work orders based on the work order keywords, including: obtaining historical keywords of each of the multiple historical work orders respectively; judging whether each historical work order meets the matching conditions based on the work order keywords and the historical keywords of each historical work order.

[0007] In another possible implementation of the first aspect, the matching condition is that the work order keyword is the same as at least one historical keyword of each historical work order; or the matching degree between the work order keyword and at least one historical keyword of each historical work order exceeds a matching threshold.

[0008] In another possible implementation of the first aspect, based on the obtained business system information corresponding to the historical work orders, the business system that handles the work order to be dispatched is confirmed, including: determining the matching level between the work order to be dispatched and each historical work order based on the work order keywords and the historical keywords of each historical work order; determining the proportion of historical work orders belonging to each business system in the matched historical work orders that meet the matching conditions at each matching level based on the business system information corresponding to each historical work order and the matching level with the work order to be dispatched; using the work order proportion as the first probability that the work order to be dispatched belongs to each business system at each matching level; converting the matching level into the weight of the first probability according to a preset conversion logic; obtaining the second probability that the work order to be dispatched belongs to each business system according to the first probability and the weight of the first probability; and confirming the business system that handles the work order to be dispatched according to the second probability.

[0009] In another possible implementation of the first aspect, the matching level between the work order to be dispatched and each historical work order is positively correlated with the number of historical keywords that meet the matching condition in the historical keywords of each historical work order.

[0010] In another possible implementation of the first aspect, based on the business system information corresponding to each historical work order and the matching level with the work orders to be dispatched, the proportion of historical work orders belonging to each business system at each matching level among the historical work orders that meet the matching conditions is determined, including: counting a first number of historical work orders that meet the matching conditions; separately counting a second number of historical work orders belonging to each business system at each matching level; based on the first number and the second number, obtaining the quantity ratio of historical work orders belonging to each business system at each matching level among the historical work orders that meet the matching conditions, and using the quantity ratio as the proportion of work orders to be dispatched belonging to each business system at each matching level.

[0011] In another possible implementation of the first aspect, the business system for processing the work order to be dispatched is confirmed based on the second probability, including: using the business system with the highest probability of the second probability in the business system as the first business system, and using the first business system as the business system for processing the work order to be dispatched; when the work order to be dispatched does not belong to the first business system, using the business system with the second highest probability of the second probability in the business system as the second business system, and updating the business system for processing the work order to be dispatched to the second business system.

[0012] In another possible implementation manner of the first aspect, the preset conversion logic includes: based on the matching level of each first probability, assigning a weight to each first probability according to a weight factor.

[0013] In another possible implementation of the first aspect, the weight factor is obtained based on linear model training, and the linear model generation method includes: obtaining sample levels and sample weight factors corresponding to multiple sample work orders, inputting the sample level and sample weight factor corresponding to each sample work order into the initial model, obtaining a sample business system corresponding to the sample level and the sample weight factor, and obtaining the target business system corresponding to the sample work order; when the sample business system is the same as the target business system, the initial model is used as the linear model.

[0014] The second aspect of the present application provides a work order dispatching system, which includes an extraction module for obtaining the work order to be dispatched and extracting the work order keywords of the work order to be dispatched; a matching module for matching the work order to be dispatched with multiple historical work orders based on the work order keywords; an acquisition module for obtaining business system information corresponding to the matched historical work orders that meet the matching conditions; and a confirmation module for confirming the business system that processes the work order to be dispatched based on the obtained business system information corresponding to the historical work orders.

[0015] The third aspect of the present application provides an electronic device, which includes: a memory storing instructions; a processor, the processor and the memory are coupled, when the program instructions stored in the memory are executed by the processor, the electronic device executes any one of the work order dispatching methods in the first aspect above.

[0016] A fourth aspect of the present application provides a computer-readable medium, which stores instructions. When the instructions are executed on an electronic device, the electronic device executes any one of the work order dispatching methods of the first aspect.

[0017] In the above-mentioned work order dispatching method, based on historical work orders and the business system information corresponding to the historical work orders, the business systems that can process the work orders to be dispatched are estimated in multiple dimensions. Therefore, the above-mentioned method can improve the accuracy of the business systems for the work orders to be dispatched determined by the work order dispatching system, thereby reducing the error rate of the work order dispatching system in dispatching work orders, thereby reducing the number of times work orders are returned and re-dispatched, and improving the actual work efficiency of the work order dispatching system. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] FIG. 1( a ) shows a display interface diagram of a user encountering a question in a terminal 1 in some embodiments of the present application;

[0019] FIG. 1( b ) shows a display interface diagram of a user creating a new work order in terminal 1 in some embodiments of the present application;

[0020] FIG. 1( c ) shows a display interface diagram of a user filling in work order information in terminal 1 in some embodiments of the present application;

[0021] Figure 2 A diagram showing application scenarios to which the work order dispatching method in some embodiments of the present application is applicable;

[0022] Figure 3 A structural diagram of a terminal 1 and a server 2 applicable to a work order dispatching method in some embodiments of the present application is shown;

[0023] Figure 4 A schematic diagram showing a work order dispatching method using work orders in some embodiments of the present application is shown;

[0024] Figure 5 A flowchart of a work order dispatching method in some embodiments of the present application is shown;

[0025] Figure 6 A schematic diagram of the main modules in the work order dispatching system 20 in some embodiments of the present application is shown;

[0026] Figure 7 A schematic diagram of main submodules in the confirmation module 400 in some embodiments of the present application is shown;

[0027] Figure 8 A schematic diagram of main units in the first probability determination submodule 420 in some embodiments of the present application is shown. DETAILED DESCRIPTION

[0028] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0029] The work order dispatching method provided in this application can be applied to Figure 2 In the application scenario shown, the terminal 1 communicates with the server 2 through the network. The terminal 1 can be, but is not limited to, various public computers, personal computers, laptops, smart phones, tablet computers, and portable smart devices, and the server 2 can be implemented by an independent server or a server cluster composed of multiple servers.

[0030] Figure 3 The structure diagram of the terminal 1 and the server 2 applicable to the work order dispatching method in some embodiments of the present application is shown. Figure 3 As shown, the server 2 includes a work order creation system 10, a work order dispatching system 20 and a business processing system 30. The work order creation system 10 may be the ITSM system mentioned above.

[0031] Figure 4 The schematic diagram of the work order dispatching method in some embodiments of the present application is shown. Figure 3 and Figure 4 It can be seen that, using the embodiments of the present application, Figure 2 In the application scenario shown, the work order creation system 10 in the server 2 is used to create a new work order based on the work order template in the work order template library 40 and the user requirements input by the user through the terminal 10, and send the new work order as a work order to be dispatched to the work order dispatching system 20 in the server 2.

[0032] After the work order dispatching system 20 receives the work order to be dispatched, in order to improve the accuracy of the pre-evaluated business system to which the work order to be dispatched may belong, the work order dispatching system 20 needs to assist the affiliated system in determining the business system to which the work order to be dispatched may belong based on the work order information and historical work orders carried by the work order to be dispatched in addition to the affiliated system. For example, in order to facilitate comparison and establish an association between the work order to be dispatched and the historical work order, the keywords of the work order to be dispatched and the historical work order are used as the comparison benchmark below. It can be understood that the comparison benchmark for the association between the work order to be dispatched and the historical work order can also be the word vectors of the work order to be dispatched and the historical work order, and this application does not impose specific restrictions on this.

[0033] Based on this, the work order dispatching system 20 obtains the work order keywords corresponding to the work order to be dispatched according to the work order information carried by the work order to be dispatched. For example, the work order information includes the system to which the question belongs, the problem overview, the problem details (as shown in Figure 1(b)) and the interface screenshots (interface 13 shown in Figure 1(b) and interface 14 shown in Figure 1(c)) filled in by the user. Assuming that a work order information includes the text message "Click to obtain the login password via SMS authorization code. The information you entered is incorrect. (Interface 11 and interface 12 shown in Figure 1(a))", its work order keywords can be "SMS authorization code", "login password" and "incorrect".

[0034] Subsequently, in order to reduce the amount of calculation and improve work efficiency, the work order dispatching system 20 can first filter out historical work orders related to the work orders to be dispatched. For example, filter out historical work order corresponding information related to historical keywords and work order keywords from the historical work order corresponding list. For another example, filter out historical work orders related to historical keywords and work order keywords from the historical work order database. Among them, the historical work order corresponding information includes a work order number, historical keywords and business system information of a historical work order. It can be understood that since the historical work order is a completed work order, the work order number, historical keywords and business system information of the historical work order are all uniquely determined.

[0035] Then, since the matching levels of different historical work orders and work orders to be dispatched are not the same, in order to provide a comprehensive reference for the work orders to be dispatched based on the historical work orders and the business systems to which the historical work orders actually belong, the work order dispatching system 20 calculates and counts the matching levels of the work order keywords and the historical keywords corresponding to each historical work order, and counts the basic probability that the work orders to be dispatched belong to each business system under different matching levels.

[0036] Subsequently, in order to integrate the basic probabilities of the same business system under multiple matching levels to obtain the probability that a work order to be dispatched corresponds to the business system, the work order dispatching system 20 converts the matching level into the weight of the basic probability. For example, the work order dispatching system 20 obtains the weight corresponding to each matching level according to the level of the matching level. For example, the higher the matching level, the greater the weight corresponding to the matching level, that is, the greater the weight of each basic probability corresponding to the matching level. Furthermore, the work order dispatching system 20 obtains the comprehensive probability that the work order to be dispatched belongs to each business system based on the business system information of the historical work order and the matching level between the historical work order and the work order to be dispatched according to the weight and the basic probability.

[0037] Finally, the work order dispatching system 20 determines the business system information in the business processing system 30 corresponding to the work order to be dispatched according to the comprehensive probability, and then dispatches the work order to be dispatched to the determined business system. The business system responds to the work order to be dispatched and returns the response result to the work order creation system.

[0038] For example, for the work order mentioned above that includes the text message "SMS authorization code, click to obtain login password. The information you entered is incorrect.", its corresponding business system can be an account and password management system.

[0039] Among them, the basic probability refers to the proportion of historical work orders corresponding to each business system in the total historical work orders screened out at each matching level, that is, the probability that the work orders to be dispatched correspond to each business system at each matching level. The probability weight refers to the proportion of each basic probability based on the matching level in the actual probability. The comprehensive probability refers to the probability that the work orders to be dispatched belong to each business system, combining the influence of the matching level and the basic probability. The business system refers to the system used to maintain various businesses, in which the operation and maintenance personnel corresponding to each business system can answer questions in the work orders and send the answers to the users.

[0040] In the above work order dispatching method, since the comprehensive probability that the work order to be dispatched belongs to each business system is obtained according to the basic probability that the work order to be dispatched belongs to each business system under different matching levels, the probability of the business system to which the work order to be dispatched belongs can be comprehensively evaluated from multiple dimensions, so that the comprehensive probability is closer to the actual probability that the work order to be dispatched belongs to each business system, that is, the reliability of the comprehensive probability is improved. Based on this, the above work order dispatching method can improve the accuracy of the business system to which the work order to be dispatched is determined by the work order dispatching system, thereby reducing the error rate of the work order dispatching system, thereby reducing the number of times the work order is returned and re-dispatched, and improving the actual work efficiency of the work order dispatching system.

[0041] This application discloses a work order dispatching method. Figure 5 This is a flowchart of the work order distribution method in this application. Figure 5 Describe in detail the work order distribution scheme in this application. Figure 5 As shown, the work order dispatching method in this application specifically includes the following steps:

[0042] Step S510: Server 2 receives the work order to be dispatched, wherein the work order to be dispatched carries work order information.

[0043] Work order information refers to the characteristics, phenomena or details corresponding to the work order to be dispatched. Work order information may include the information reflected in the screenshot of the error interface uploaded by the business personnel when the work order is created, for example, the transaction sequence number, error information and error code in the screenshot of the interface. Work order information may also include text information entered by the business personnel when the work order is created. For example, relevant descriptions such as "system", "problem overview", "problem details" filled in by the business personnel. It can be understood that the present application does not impose specific restrictions on the specific form of the work order information. Any information that can reflect the user's request can be the work order information of the person in this application, and they are all within the scope of protection of this application, and will not be described one by one here.

[0044] Step S520: Server 2 extracts the work order keywords corresponding to the work order to be dispatched based on the work order information and the keyword database.

[0045] Among them, the keyword represents the relevant words that can reflect the actual business required by the user, and the work order keyword refers to the relevant words that can reflect the actual business required by the user in the work order to be dispatched. It can be understood that the number of work order keywords corresponding to the work order to be dispatched is several, such as one, two, three, etc., and this application does not make specific restrictions.

[0046] In some implementations, when the work order information is text information, the server 2 directly extracts the work order keywords from the text information. In other implementations, when the work order information is non-text information, for example, when the work order information is a screenshot of the error reporting interface, the server 2 first obtains the text information based on the screenshot of the error reporting interface, and then extracts the work order keywords from the obtained text information.

[0047] The following is a brief description of the scheme for extracting work order keywords from text information. Server 2 performs word segmentation on the text information to obtain a word segmentation result, and the word segmentation result includes at least one initial vocabulary. Specifically, server 2 uses text mining technology to parse the text information to obtain a word segmentation result corresponding to the text information. For example, server 2 uses text mining technology to parse the text information, which may be to clean the text information, segment the text information, and extract feature information from the segmented text information to obtain a word segmentation result, and the extracted feature information is the initial vocabulary in the word segmentation result. Among them, extracting feature information from the segmented text information may be extracting vocabulary, extracting terms, extracting information based on templates, converting concepts based on semantic dictionaries, extracting grammatical features based on shallow syntactic analysis, extracting semantic features based on shallow semantic analysis, etc. The present application does not impose specific restrictions on the specific method of extracting feature information from the segmented text information.

[0048] Then, the server 2 matches each initial word in the word segmentation result with the standard keyword in the keyword database, and determines whether the initial word is a work order keyword of the work order to be dispatched according to the matching result.

[0049] In some implementations, the matching result may be whether the initial word is identical to one of the standard keywords. When the matching result is that the initial word is identical to one of the standard keywords in the keyword database, the initial word is a work order keyword for the work order to be dispatched. When the matching result is that the initial word is not identical to any of the standard keywords, the initial word is not a work order keyword for the work order to be dispatched.

[0050] In some other implementations, the matching result may be whether the maximum matching rate between the initial vocabulary and any standard keyword in the keyword database exceeds a preset first matching threshold. Among them, the matching rate represents the similarity between the initial vocabulary and the standard keyword, and the maximum matching rate represents the maximum value of the matching rates of the initial vocabulary and the standard keywords. When the maximum matching rate exceeds the preset first matching threshold, the initial vocabulary is a work order keyword corresponding to the work order to be dispatched. Among them, the first matching threshold may be 100%, 90%, and so on. It can be understood that the matching rate can be determined based on the similarity between the word vector corresponding to the initial vocabulary and the word vector corresponding to the standard keyword. The present application does not make a specific method for obtaining the matching rate between the initial vocabulary and the standard keyword. Any comparison method that can obtain the similarity between the initial vocabulary and the standard keyword is within the protection scope of the present application, and the present application does not make specific limitations.

[0051] Based on this, each initial word in the word segmentation result is matched with the standard keyword in the keyword database until all initial words in the word segmentation result are matched with the standard keywords in the keyword database, that is, all work order keywords corresponding to the work orders to be dispatched are obtained.

[0052] For example, the text information corresponding to Figure 1(a) is "SMS authorization code click to obtain login password. The information you entered is incorrect." Server 2 uses text mining technology to parse the text information, and after cleaning the information data, obtains "SMS authorization code login password input information is incorrect". After segmenting the text information, it obtains "SMS authorization code / login password / input / of / information / incorrect", extracts the feature information in the text information, and obtains the word segmentation results of "SMS authorization code", "login password", "input", "information" and "incorrect". "SMS authorization code", "login password", "input", "information" and "incorrect" are matched with the standard keywords in the keyword database respectively, and finally the work order keywords corresponding to the work order to be dispatched are obtained as "SMS authorization code", "login password" and "incorrect".

[0053] It can be understood that the present application does not specifically limit the method for obtaining the keywords corresponding to the work orders to be dispatched, and other methods for obtaining the keywords corresponding to the work orders to be dispatched are within the protection scope of the present application.

[0054] In addition, the keyword database refers to a database for storing standard keywords. Standard keywords are extracted from historical work orders and are used to determine the empirical vocabulary of keywords in the work orders to be dispatched. The keyword database can be a database composed of keywords extracted from a large number of historical work orders. It can be understood that the standard keywords in the keyword database can be extracted from multiple dimensions such as users, channels, accounts, business scenarios, transaction names, error codes and error messages.

[0055] The method for obtaining standard keywords in the keyword database is roughly the same as the method for obtaining work order keywords in the work order to be dispatched, that is, server 2 uses semantic parsing technology to parse the historical text information of the historical work order, obtains the historical keywords corresponding to the historical text information, and then stores the obtained historical keywords as the keyword database. It can be understood that with the update of historical work orders, the standard keywords in the keyword database generated based on historical work orders are also constantly adjusted to adapt to the needs of users at different stages. For the work orders to be dispatched that are eventually transferred to manual dispatch, the work orders to be dispatched are summarized within a preset period, and the work order keywords are manually extracted, and then the extracted work order keywords are stored in the keyword database, wherein the keyword database can be updated manually in this implementation.

[0056] In some implementations, the historical keywords may be determined based on historical work orders. In other implementations, the historical keywords may also be determined based on historical work orders and the systems corresponding to the historical work orders.

[0057] Step S530: Server 2 determines the basic probability that the to-be-dispatched work order belongs to each business system at different matching levels based on the work order keywords and the historical corresponding list. For ease of understanding, before describing how to determine the business system corresponding to the to-be-dispatched work order based on the work order keywords corresponding to the to-be-dispatched work order and the historical corresponding list, it is also necessary to first describe the historical corresponding list and the method for generating the historical corresponding list.

[0058] The historical corresponding list can represent the historical keywords and corresponding business systems in the historical work orders. The historical corresponding list can be a historical work order corresponding list, or a historical work order cumulative list.

[0059] Among them, the historical corresponding list includes the historical work order corresponding information corresponding to each historical work order. The historical work order corresponding information represents the historical keywords and business system information corresponding to each historical work order. It can be understood that the business system information corresponding to each historical work order corresponds to the business system to which the historical work order is actually dispatched. Therefore, the business system information corresponding to any historical work order is uniquely determined, and the historical keywords corresponding to the historical work order are also uniquely determined, that is, the historical work order corresponding information of the historical work order is uniquely determined.

[0060] Specifically, server 2 obtains the work order number, historical keywords corresponding to the historical work order, and business system information carried by the historical work order. Server 2 associates and stores the work order number, historical keywords, and business system information corresponding to each historical work order as a historical work order corresponding information. For example, the historical work order corresponding information of historical work order 1 in Table 1 is "historical work order 1 - a1', a2', a3', a4' - S1". Based on this, m historical work order correspondences constitute a historical corresponding list, as shown in Table 1.

[0061] Table 1 Corresponding list of historical work orders

[0062] Historical work order number Historical Keywords Business system information Historical Work Order 1 <![CDATA[a1',a2',a3',a4']]> <![CDATA[S1 <!-- 6 -->]]> Historical Work Order 2 <![CDATA[a1',a2',a3',a4']]> <![CDATA[S1]]> Historical Work Order 3 <![CDATA[a2',a3',a5',a6',a8']]> <![CDATA[S2]]> Historical Work Order 4 <![CDATA[a1',a2',a3',a4']]> <![CDATA[S3]]> Historical Work Order 5 <![CDATA[a1',a2',a3',a5',a9',a 10 ']]> <![CDATA[S1]]> …… …… …… Historical Work Orders <![CDATA[a2',a 11 ',a 12 ',a m ']]> <![CDATA[S m ]]>

[0063] Through the historical work order corresponding information corresponding to historical work order 1 in Table 1 "Historical work order 1 - a1', a2', a3', a4'a5', a6', a7' - S1" and the historical work order corresponding information corresponding to historical work order 2 "Historical work order 2 - a1', a2', a3', a4' - S1", it can be seen that different historical work orders can have the same historical keywords and the same business system. Through the historical work order corresponding information corresponding to historical work order 1 in Table 1 "Historical work order 1 - a1', a2', a3', a4' - S1" and the historical work order corresponding information corresponding to historical work order 4 "Historical work order 4 - a1', a2', a3', a4' - S3", it can be seen that different historical work orders may correspond to different business systems even if they have the same historical keywords. To facilitate subsequent descriptions, all historical keywords corresponding to each historical work order are now regarded as a historical keyword set.

[0064] It can be understood that as historical work orders are updated, the historical work order correspondence in the historical work order correspondence list generated based on historical work orders will also be continuously adjusted to adapt to the needs of users at different stages. For example, when the work order dispatching system in server 2 successfully dispatches the work order to be dispatched, server 2 will associate the corresponding work order keywords of the work order to be dispatched with the business system to which it is finally dispatched as a historical work order corresponding information and store it in the historical work order corresponding list. For the work orders to be dispatched that are ultimately transferred to manual dispatch, the work orders to be dispatched are summarized within a preset period, and through manual analysis, the extracted work order keywords and the business system corresponding to the work order to be dispatched are associated as a historical work order corresponding information and stored in the historical work order corresponding list, wherein the historical work order corresponding list can be updated manually in this implementation.

[0065] As can be seen from the above, the number of work order keywords corresponding to the work order to be dispatched can be multiple. After obtaining the work order keyword and the corresponding list of historical work orders, the server 2 compares the work order keyword with the historical keywords in all historical keyword sets in the corresponding list of historical work orders. When the work order keyword successfully matches at least one historical keyword in the historical keyword set, the server 2 filters out the historical work order corresponding to the historical keyword set, and the historical work order corresponding information corresponding to the filtered historical work order.

[0066] In some implementations, the server 2 obtains the matching keywords between the work order keyword and the historical keyword set in each historical work order, and counts the number of matching keywords. The matching level between the work order to be dispatched and each historical work order is positively correlated with the number of historical keywords that meet the matching conditions in the historical keywords of each historical work order. Among them, the matching keywords refer to the keywords that meet the keyword matching conditions between the work order keyword and the historical keyword set. In some implementations, the server 2 uses the number of matching keywords as the matching level between the work order to be dispatched and the historical work order. The larger the value of the matching level, the higher the level of the matching level. The keyword matching condition can be that the work order keyword is the same as the historical keyword, and the keyword matching condition can also be that the matching degree between the work order keyword and each historical keyword exceeds a preset second matching threshold.

[0067] Assuming that the work order keywords include a1, a2, a3, a4, a5, a6, a7, and a8, where a1 matches a1', a2 matches a2', a3 matches a3', a4 matches a4', a5 matches a5', a6 matches a6', a7 matches a7', and a8 matches a8', the matching keywords and matching levels of the work order to be dispatched and each historical work order are shown in Table 2.

[0068] Table 2 Matching keyword-matching level list

[0069] Historical work order number Matching keywords Business system information Matchmaking Level Historical work order 1 <![CDATA[a1',a2',a3',a4']]> <![CDATA[S1]]> 4 Historical Work Order 2 <![CDATA[a1',a2',a3',a4']]> <![CDATA[S1]]> 4 Historical Work Order 3 <![CDATA[a2',a3',a5',a6',a8']]> <![CDATA[S2]]> 5 Historical Work Order 4 <![CDATA[a1',a2',a3',a4']]> <![CDATA[S3]]> 4 Historical Work Order 5 <![CDATA[a1',a2',a3',a5']]> <![CDATA[S1]]> 4 …… …… …… Historical Work Orders <![CDATA[a2']]> <![CDATA[S m ]]> 1

[0070] In some other implementations, server 2 matches each keyword in the work order keyword with the keywords in the historical keyword set in all historical work orders, counts the number of times the work order keyword matches each historical work order, and uses the number of matches as the matching level between the work order to be dispatched and the historical work order.

[0071] After completing the matching of all work order keywords with all historical keyword sets and the screening of the corresponding information of historical work orders, Server 2 counts the number of historical work orders corresponding to different business systems at different matching levels based on the corresponding information of all historical work orders obtained, as shown in Table 3. As shown in Table 3, there are N1+N4 historical work orders with a matching level of 4 and a business system information of S1 in the historical work orders. It can be understood that the matching level corresponds to the number of matching keywords, and the number of matching keywords only limits the number of keywords that match the work order keywords in the historical keywords, and does not specifically limit which keywords are matched. That is, the matching keywords of two historical work orders and the work order to be dispatched at the same matching level may be different, but the number of matching keywords of two historical work orders and the work order to be dispatched at the same matching level must be the same.

[0072] Table 3. Cumulative number of work orders corresponding to different matching levels and different business systems

[0073] Serial Number Matchmaking Level Business system information Cumulative number of work orders 1 4 <![CDATA[S1]]> <![CDATA[N1+N4]]> 2 5 <![CDATA[S2]]> <![CDATA[N2]]> 3 4 <![CDATA[S3]]> <![CDATA[N3]]> …… …… …… …… m 2 <![CDATA[S m ]]> <![CDATA[N m ]]>

[0074] Based on Table 3 above, the proportion of historical work orders with different matching levels and different business system information in the total filtered historical work orders is calculated, and these work order proportions are used as the basic probability that the work orders to be dispatched belong to each business system at different matching levels. Among them, the basic probability is the probability that the work orders to be dispatched correspond to each business system at different matching levels, which is directly determined based on the historical work orders, as shown in Table 4.

[0075] Among them, it can be understood that N 总 =N1+N2+N3+N4+……+N n , that is, x1%+x2%+x3%+x4%+x5%+x6%=1 in Table 4.

[0076] Table 4 Basic probability table corresponding to different matching levels and different business system information

[0077]

[0078] In some other implementations, when the historical keywords of two historical work orders are the same and the systems corresponding to the two work orders are the same (such as historical work order 1 and historical work order 2 in Table 1), the two historical work orders can also be stored cumulatively, or the two historical work orders can be stored in association.

[0079] For example, combining Table 1 and Table 2, it can be seen that the keywords corresponding to historical work order 1 and historical work order 2 are a1', a2', a3' and a4', and the systems corresponding to historical work order 1 and historical work order 2 are both S1 systems. Therefore, the historical work order corresponding information corresponding to historical work order 1 and the historical work order corresponding information corresponding to historical work order 2 can be merged to obtain the merged historical work order corresponding information, see the cumulative information of historical work orders with serial number 1 in Table 5.

[0080] However, although the keywords corresponding to historical work order 4 in Table 1 are a1', a2', a3', and a4', since the system corresponding to historical work order 4 is the S3 system, historical work order 4 cannot be accumulated into the historical work order accumulation information corresponding to historical work order 1 and historical work order 2, see the historical work order accumulation information with sequence number 3 in Table 3. Based on this, the corresponding information of m work orders in Table 1 is accumulated to form a historical work order accumulation list, as shown in Table 5. According to Table 4, each historical work order accumulation information in the historical work order accumulation list represents the accumulated number of work orders with the same matching keywords and the same corresponding system.

[0081] Table 5 Historical work order cumulative list

[0082]

[0083] After obtaining the work order keyword and the historical work order accumulation list, the server 2 compares the work order keyword with each historical keyword set in the historical work order accumulation list. When the work order keyword successfully matches at least one keyword in the historical keyword set, the historical work order accumulation information corresponding to the historical keyword set is filtered out.

[0084] After completing the matching of all work order keywords with the historical keyword set in all historical work order cumulative information and filtering the historical work order cumulative information, server 2 counts the number of different keywords and the number of work orders corresponding to different business systems based on the filtered historical work order cumulative information. The specific method is similar to the process of obtaining Table 2 from Table 1, and will not be repeated here.

[0085] Step S540: Server 2 determines the comprehensive probability that the work order to be dispatched belongs to each business system according to the matching level and the basic probability corresponding to each system category.

[0086] The comprehensive probability refers to the probability that the work orders to be dispatched correspond to each business system by comprehensively considering the basic probabilities under all matching levels. In some embodiments, after obtaining the basic probabilities of the work orders to be dispatched corresponding to different business systems under different matching levels, the server 2 obtains the comprehensive probability of the business system information corresponding to the work orders to be dispatched based on the basic probabilities corresponding to different matching levels and different degrees.

[0087] In some embodiments, the server 2 converts the matching level into a probability weight of the basic probability according to a preset conversion logic, and then performs a weighted sum based on the probability weight and the basic probability to obtain a comprehensive probability that the work order to be dispatched corresponds to each business system information. For details, please refer to the following formula:

[0088]

[0089] Among them, P i Indicates that the work order to be dispatched corresponds to S i The comprehensive probability of the system, n represents n and S i The basic probabilities related to the system are x i1 、x i2 、x i3 ,……x i(n-1) and x in , at the same time, with x i1 The corresponding probability weight is k i1 , and x i2 The corresponding probability weight is k i2 , and x i3 The corresponding probability weight is k i3 , ..., and x i(n-1) The corresponding probability weight is k i(n-1) , and x in The corresponding probability weight is k in .

[0090] In some implementations, the matching level includes multiple groups of matching levels, and the server 2 determines the probability weight corresponding to each basic probability based on the matching level corresponding to each basic probability based on a preset conversion logic. The conversion logic is a data conversion rule or processing rule pre-stored in the server 2, and the conversion process of different categories of matching levels may require the use of different conversion logics. The conversion logic may be that the server 2 assigns a probability weight to each basic probability based on the matching level of each basic probability according to a weight factor.

[0091] For example, it is assumed that there are k levels of matching levels between the work orders to be dispatched and the historical work orders, and the weight factor is y%.

[0092] Among them, in terms of multiple basic probabilities corresponding to each business system information:

[0093] The probability weight of the basic probability with the highest matching level, that is, the weight of the basic probability with the largest number of matching keywords, k1=y%;

[0094] The probability weight of the basic probability with the second highest matching level, that is, the weight of the basic probability with the second largest number of matching keywords k2 = (1-k1)*y%;

[0095] The probability weight of the third highest basic probability of matching level, that is, the weight of the basic probability of matching the third largest number of keywords k3 = (1-k1-k2)*y%;

[0096] The probability weight of the fourth highest basic probability of matching level, that is, the weight of the basic probability of matching the fourth largest number of keywords k4 = (1-k1-k2-k3)*y%;

[0097] …

[0098] The probability weight of the base probability with the k-1th highest matching level, that is, the weight k of the probability of matching the (k-1)th largest number of keywords k-1 =(1-k1-k2-k3-……-k k-2 )*y%;

[0099] The probability weight of the base probability with the kth highest matching level, that is, the weight k of the probability with the kth largest number of matching keywords k =1-k1-k2-k3-……-k k-2 -k k-1 .

[0100] It is not difficult to find that the above k1, k2, k3, ... k k-2 , k k-1 It is a geometric progression with a proportional coefficient of 1-k1. That is, the proportional coefficient of the geometric progression is the sum of all weights except the weight corresponding to the highest matching level.

[0101] It can be understood that when k=1, y=100%. When k=2, the weight of the basic probability of matching a large number of keywords is y%, and the weight of the basic probability of matching a large number of keywords is 1-y%.

[0102] In some implementations, the weight factor y% may be obtained based on model training. For example, y% is obtained through linear model training. The linear model refers to a model that is pre-stored in the server 2 and can be calculated through matching level, basic probability, and weight factor y% to obtain business system information corresponding to the work order. The linear model may be at least one of a logistic regression model, a least squares model, a linear regression model, and a decision tree obtained by training with sample data.

[0103] In a specific implementation, the generation method of the linear model includes: server 2 obtains multiple sample data, which may be positive sample data and negative sample data, performs machine learning on the sample features corresponding to the multiple sample data and the sample results corresponding to the sample data to obtain an initial model, and optimizes the initial model to obtain a linear model. The sample features may include a sample level, sample probability, and sample weight factor corresponding to a sample work order, wherein the sample probability is the basic probability corresponding to the sample work order. The sample results may include the target business system corresponding to the sample work order. Server 2 calculates the sample business system corresponding to the sample work order through the linear model according to the sample features, and continuously linearly regresses the model until the sample business system is consistent with the target business system, and uses the obtained model as the linear model.

[0104] In some other implementations, the matching level includes multiple matching levels, and the weight corresponding to each basic probability is determined according to the number of matching keywords corresponding to each matching level.

[0105] In the above work order dispatching method, the number of matched keywords represents the matching level between the work order to be dispatched and the historical work order. The more matching keywords there are, the higher the matching level between the work order to be dispatched and the historical work order. In other words, the more similar the work order to be dispatched is to the historical work order, the greater the probability that it belongs to the business system corresponding to the historical work order. The higher the matching level, the more dependent the work order to be dispatched is on the basic probability of each business under the matching level.

[0106] Step S550: Server 2 determines the system category corresponding to the dispatched work order according to the comprehensive probability that the work order to be dispatched belongs to each system category, and dispatches the work order to be dispatched to the business system corresponding to the determined system category.

[0107] The comprehensive probability is the probability that the work order to be dispatched belongs to each business system obtained by server 2 according to the basic probability and the corresponding weight. Server 2 obtains the comprehensive probability that the work order to be dispatched corresponds to all business system information, as shown in Table 6.

[0108] Table 6

[0109]

[0110] Server 2 uses the first business system corresponding to the highest value in the comprehensive probability as the business system most likely to process the work order to be dispatched, and dispatches the work order to be dispatched to the first business system, so that the first business system further dispatches the work order to be dispatched to the operation and maintenance personnel of the corresponding system.

[0111] In some implementations, when the first business system corresponding to the highest value in the comprehensive probability is not a business system that can handle the work order to be dispatched, the first business system returns the work order to be dispatched to the work order dispatching system 20 in server 2. Then, server 2 uses the second business system corresponding to the second highest value in the comprehensive probability as the business system that is most likely to handle the work order to be dispatched, and dispatches the work order to be dispatched to the second business system. Based on this, server 2 dispatches the work order to the corresponding business system in order according to the size of the comprehensive probability value until it is dispatched to a business system that can handle the work order to be dispatched. When all the business systems corresponding to the comprehensive probabilities cannot handle the work order to be dispatched, it is transferred to manual processing, which will not be described in detail here.

[0112] In some embodiments of the present application, the present application also provides a work order dispatching system 20. Figure 6 As shown, it is a schematic diagram of the main modules in a work order dispatching system 20 provided by an embodiment of the present invention. The system has the function of implementing the work order dispatching method in the above method embodiment. The function can be implemented by hardware or by hardware executing corresponding software. Figure 6 As shown, the work order dispatching system 20 may include an extraction module 100, a matching module 200, an acquisition module 300 and a confirmation module 400. The extraction module 100 is used to obtain the work order to be dispatched and extract the work order keywords of the work order to be dispatched; the matching module 200 is used to match the work order to be dispatched with multiple historical work orders according to the work order keywords; the acquisition module 300 is used to obtain the business system information corresponding to the matched historical work orders that meet the matching conditions; the confirmation module 400 is used to confirm the business system that processes the work order to be dispatched according to the obtained business system information corresponding to the historical work orders.

[0113] In some implementations, such as Figure 6 As shown, the work order dispatching system 20 may further include a dispatching module 500. The dispatching module 500 is used to send the work order to be dispatched to the corresponding business system according to the business system for dispatching the work order confirmed by the confirmation module 400.

[0114] In some implementations, the matching module 200 may include a historical keyword acquisition submodule (not shown) and a historical work order judgment submodule (not shown). The historical keyword acquisition submodule is used to respectively obtain the historical keywords of each historical work order in the multiple historical work orders. The historical work order judgment submodule is used to judge whether each historical work order meets the matching condition according to the work order keyword and the historical keywords of each historical work order. The historical work order judgment submodule judges whether each historical work order meets the matching condition through the matching condition judgment module (not shown).

[0115] In some implementations, the matching condition judgment module includes a first matching submodule (not shown) and / or a second matching submodule (not shown). The matching logic of the first matching submodule is that the work order keyword is the same as at least one historical keyword of each historical work order. The matching logic of the second matching submodule is that the matching degree between the work order keyword and at least one historical keyword of each historical work order exceeds a matching threshold.

[0116] Figure 7 Schematic diagram of the main submodules in the confirmation module 400 in some embodiments of the present application is shown. Figure 7 As shown, the confirmation module 400 includes a matching level determination submodule 410, a first probability determination submodule 420, a weight conversion submodule 430, a second probability acquisition module 440 and a business system confirmation submodule 450. Among them, the matching level determination submodule 410 is used to determine the matching level of the work order to be dispatched and each historical work order according to the work order keyword and the historical keyword of each historical work order. The first probability determination submodule 420 is used to determine the proportion of historical work orders belonging to each business system in the matched historical work orders that meet the matching conditions under each matching level according to the business system information corresponding to each historical work order and the matching level with the work order to be dispatched, and use the work order proportion as the first probability that the work order to be dispatched belongs to each business system under each matching level. The weight conversion submodule 430 is used to convert the matching level into the weight of the first probability according to the preset conversion logic. The second probability acquisition module 440 is used to obtain the second probability that the work order to be dispatched belongs to each business system according to the first probability and the weight of the first probability. The business system confirmation submodule 450 is used to confirm the business system that processes the work order to be dispatched according to the second probability.

[0117] Figure 8 4 shows a schematic diagram of the main units in the first probability determination submodule 420 in some embodiments of the present application. In some implementations, such as Figure 8 As shown, the first probability determination submodule 420 includes a first quantity counting unit 421, a second quantity counting unit 422 and a first probability determination unit 423. The first quantity counting unit 421 is used to count the first quantity of historical work orders that meet the matching conditions obtained by matching. The second quantity counting unit 422 is used to count the second quantity of historical work orders belonging to each business system at each matching level. The first probability acquisition unit 423 is used to obtain the quantity ratio of the historical work orders belonging to each business system at each matching level in the historical work orders that meet the matching conditions obtained by matching according to the first quantity and the second quantity, and use the quantity ratio as the first probability that the work orders to be dispatched belong to each business system at each matching level.

[0118] In some implementations, the business system confirmation submodule 450 includes a business system confirmation unit (not shown) and a business system update unit (not shown). The business system confirmation unit is used to use the business system with the highest probability of the second probability in the business system as the first business system, and use the first business system as the business system for processing the work order to be dispatched. The business system update unit is used to use the business system with the second highest probability of the second probability in the business system as the second business system when the work order to be dispatched does not belong to the first business system, and update the business system for processing the work order to be dispatched to the second business system.

[0119] In some implementations, the linear model generation module (not shown) includes a sample level, sample probability and sample weight factor acquisition unit (not shown), a target business system acquisition unit (not shown) and a linear model confirmation unit (not shown). Among them, the sample level, sample probability and sample weight factor acquisition unit respectively acquires the sample level, sample probability and sample weight factor corresponding to multiple sample work orders. The target business system acquisition unit is used to input the sample level, sample probability and sample weight factor corresponding to each sample work order into the initial model, obtain the sample business system corresponding to the sample level, sample probability and sample weight factor, and obtain the target business system corresponding to the sample work order. The linear model confirmation unit is used to use the initial model as the linear model when the sample business system is the same as the target business system.

[0120] It can be understood that the module in the work order dispatching system in the above embodiment is only one of the embodiments of the present application, the submodule in the module is also only one of the embodiments of the present application, and the unit in the submodule is also only one of the embodiments of the present application. It can be understood that the modules, submodules and units in the above embodiment can be reorganized or split arbitrarily, and the present application does not make specific restrictions. Any other form of the work order dispatching system that can implement the work order dispatching solution in the present application is within the scope of protection of the present application, and the present application does not make specific restrictions.

[0121] References in the specification to "some embodiments" or "embodiments" mean that a particular feature, structure, or characteristic described in conjunction with the embodiments is included in at least one exemplary implementation or technique according to the present disclosure. The appearances of the phrase "in one embodiment" in various places in the specification do not necessarily all refer to the same embodiment.

[0122] The present disclosure also relates to an apparatus for performing the operations in the text. The apparatus may be constructed specifically for the required purpose or it may include a general-purpose computer selectively activated or reconfigured by a computer program stored in the computer. Such a computer program may be stored in a computer-readable medium, such as, but not limited to, any type of disk, including a floppy disk, an optical disk, a CD-ROM, a magneto-optical disk, a read-only memory (ROM), a random access memory (RAM), an EPROM, an EEPROM, a magnetic or optical card, an application-specific integrated circuit (ASIC), or any type of medium suitable for storing electronic instructions, and each may be coupled to a computer system bus. In addition, the computer mentioned in the specification may include a single processor or may be an architecture involving multiple processors for increased computing power.

[0123] The process and display proposed herein do not inherently relate to any specific computer or other device. Various general purpose systems may also be used together with the program according to the teachings herein, or it may prove convenient to construct more special purpose devices to perform one or more method steps. The structure for various these systems is discussed in the following description. In addition, any specific programming language sufficient to implement the technology and embodiments of the present disclosure may be used. Various programming languages ​​may be used to implement the present disclosure, as discussed herein.

[0124] Additionally, the language used in this specification has been primarily selected for readability and instructional purposes and may not have been selected to delineate or limit the disclosed subject matter.Thus, this disclosure is intended to illustrate but not to limit the scope of the concepts discussed herein.

Claims

1. A work order dispatching method, characterized in that: include: Acquire a work order to be dispatched, and extract work order keywords of the work order to be dispatched according to the work order information carried by the dispatched work order and a keyword database, wherein standard keywords in the keyword database are extracted from historical work orders, and the standard keywords are words extracted based on at least one of the following dimension analysis: user, channel, account, business scenario, transaction name, error code, and error message; According to the work order keywords, the work order to be dispatched is matched with multiple historical work orders; Obtaining business system information corresponding to the historical work orders that meet the matching conditions; Determine, according to the work order keyword and the historical keywords of each of the historical work orders, a matching level between the work order to be dispatched and each of the historical work orders, wherein the matching level between the work order to be dispatched and each of the historical work orders is positively correlated with the number of the historical keywords that meet the matching condition in the historical keywords of each of the historical work orders; According to the business system information corresponding to each of the historical work orders and the matching level with the work order to be dispatched, determining the proportion of the historical work orders belonging to each business system in the historical work orders meeting the matching conditions obtained by the matching at each matching level; The work order ratio is used as a first probability that the work order to be dispatched belongs to each business system at each matching level; Converting the matching level into a weight of a first probability according to a preset conversion logic; According to the first probability and the weight of the first probability, obtaining a second probability that the to-be-dispatched work order belongs to each of the business systems; The business system that processes the work order to be dispatched is confirmed according to the second probability.

2. The method according to claim 1, characterized in that The matching of the to-be-dispatched work order with a plurality of historical work orders according to the work order keywords includes: Respectively obtain historical keywords of each of the multiple historical work orders; It is determined whether each of the historical work orders meets the matching condition according to the work order keyword and the historical keyword of each of the historical work orders.

3. The method according to claim 1 or 2, characterized in that: The matching condition is The work order keyword is the same as at least one historical keyword of each of the historical work orders; or The matching degree between the work order keyword and at least one historical keyword of each of the historical work orders exceeds a matching threshold.

4. The method according to claim 1, characterized in that: Determining the proportion of historical work orders belonging to each business system at each matching level among the historical work orders meeting the matching conditions obtained by matching according to the business system information corresponding to each historical work order and the matching level with the work order to be dispatched, includes: Counting a first number of historical work orders obtained by the matching that meet the matching condition; Counting the second number of historical work orders belonging to each business system at each matching level respectively; Based on the first quantity and the second quantity, obtain the quantity ratio of the historical work orders belonging to each business system at each matching level among the historical work orders that meet the matching conditions obtained by the matching, and use the quantity ratio as the work order ratio of the work orders to be dispatched belonging to each business system at each matching level.

5. The method according to claim 1 or 4, characterized in that: The business system for confirming and processing the work order to be dispatched according to the second probability includes: The business system with the highest probability of the second probability among the business systems is used as a first business system, and the first business system is used as a business system for processing the work order to be dispatched; When the work order to be dispatched does not belong to the first business system, the business system with the second highest probability among the business systems is used as the second business system, and the business system processing the work order to be dispatched is updated to the second business system.

6. The method according to claim 1, characterized in that The preset conversion logic includes: Based on the matching level of each first probability, a weight is assigned to each first probability according to a weight factor.

7. The method according to claim 6, characterized in that The weight factor is obtained based on linear model training, and the linear model generation method includes: Obtain sample levels, sample probabilities, and sample weight factors corresponding to multiple sample work orders respectively; Input the sample level, sample probability and sample weight factor corresponding to each sample work order into the initial model, obtain the sample business system corresponding to the sample level, sample probability and sample weight factor, and obtain the target business system corresponding to the sample work order; When the sample business system is the same as the target business system, the initial model is used as the linear model.

8. A work order dispatching system, characterized in that: include: An extraction module, used to obtain a work order to be dispatched, and extract work order keywords of the work order to be dispatched according to the work order information carried by the dispatched work order and a keyword database, wherein standard keywords in the keyword database are extracted from historical work orders, and the standard keywords are words extracted based on at least one of the following dimensions: user, channel, account, business scenario, transaction name, error code, and error message; A matching module, used for matching the to-be-dispatched work order with a plurality of historical work orders according to the work order keywords; An acquisition module is used to acquire the business system information corresponding to the historical work orders that meet the matching conditions; A confirmation module, configured to determine a matching level between the work order to be dispatched and each of the historical work orders according to the work order keyword and the historical keywords of each of the historical work orders, wherein the matching level between the work order to be dispatched and each of the historical work orders is positively correlated with the number of the historical keywords that meet the matching condition in the historical keywords of each of the historical work orders; According to the business system information corresponding to each of the historical work orders and the matching level with the work order to be dispatched, determining the proportion of the historical work orders belonging to each business system in the historical work orders meeting the matching conditions obtained by the matching at each matching level; The work order ratio is used as a first probability that the work order to be dispatched belongs to each business system at each matching level; Converting the matching level into a weight of a first probability according to a preset conversion logic; According to the first probability and the weight of the first probability, obtaining a second probability that the to-be-dispatched work order belongs to each of the business systems; The business system that processes the work order to be dispatched is confirmed according to the second probability.

9. An electronic device, characterized in that: include: A memory storing instructions; A processor is coupled to a memory, and when the program instructions stored in the memory are executed by the processor, the electronic device executes the work order dispatching method as described in any one of claims 1 to 7.

10. A computer-readable medium having instructions stored therein, characterized in that: When the instruction is executed on the electronic device, the electronic device executes the work order dispatching method as described in any one of claims 1 to 7.

Images