Picture supervision method and device and computer equipment
By obtaining and screening standard traffic penalty information and impact factor measurement indicators for sleeping tickets, and using principal component analysis and factor load matrix for supervision, the problem of inaccurate supervision of sleeping tickets is solved, the accuracy and efficiency of supervision is improved, and complexity is reduced. sexual and overfitting risks.
Patent Information
- Application Number
- CN202411029843.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-30
- Publication Date
- 2024-12-06
AI Technical Summary
The existing technology has problems of inaccurate supervision when supervising sleeping tickets, which leads to road traffic safety risks and is prone to inaccurate supervision results caused by overfitting.
By obtaining standard traffic penalty information for the sleepy ticket of driving users, obtaining impact factor measurement indicators based on preset reliability assessment conditions, using principal component analysis and factor load matrix to screen important ticket impact factors, conduct supervision, and avoiding neural network model optimization process to reduce the risk of overfitting.
It improves the accuracy and timeliness of the supervision of sleepy tickets, reduces the complexity and difficulty of supervision, improves the supervision speed and efficiency of batch sleepy tickets, and ensures the reliability of supervision results.
Smart Images

Figure CN119090119A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of traffic technology, and in particular to a ticket supervision method, device and computer equipment. Background Art
[0002] With the continuous increase of vehicles, traffic violations of vehicles are becoming more and more serious. When traffic violations occur, the traffic control department usually issues a traffic penalty ticket to punish the driver and correct the traffic violation.
[0003] In actual applications, there are often traffic penalty tickets that drivers have not confirmed or have not processed after confirmation. Such traffic penalty tickets are called sleeping tickets. Among them, the existence of sleeping tickets may cause great hidden dangers to road traffic safety. Therefore, it is particularly important to supervise sleeping tickets.
[0004] However, in the related art, there is a problem of inaccurate supervision when supervising dormant tickets. Summary of the invention
[0005] Based on this, it is necessary to provide a ticket supervision method, device and computer equipment to address the above technical problems, which can improve the accuracy of supervising dormant tickets when supervising dormant tickets.
[0006] In a first aspect, an embodiment of the present application provides a ticket supervision method, the method comprising:
[0007] Get the standard traffic penalty information of each driver's sleeping ticket;
[0008] When each standard traffic penalty information meets the preset reliability evaluation conditions, the impact factor measurement index corresponding to each standard traffic penalty information is obtained; the impact factor measurement index represents the importance of each standard traffic penalty information when described by multiple different ticket impact factors;
[0009] According to each standard traffic penalty information and each influencing factor measurement index, at least one dormant ticket corresponding to each ticket influencing factor is supervised.
[0010] The technical solution in the embodiment of the present application obtains standard traffic penalty information of dormant tickets of each driving user, and when each standard traffic penalty information meets the preset reliability evaluation conditions, obtains the impact factor measurement index corresponding to each standard traffic penalty information, and supervises at least one dormant ticket corresponding to each ticket impact factor based on each standard traffic penalty information and each impact factor measurement index, wherein the impact factor measurement index represents the importance of each standard traffic penalty information when described by multiple different ticket impact factors; the above method can perform supervision on each dormant ticket based on each standard traffic penalty information only when it is determined that each standard traffic penalty information meets the preset reliability evaluation conditions, so that the supervision of dormant tickets can be guaranteed. The reliability of the results can be improved, and the accuracy of the supervision of dormant tickets can be improved; at the same time, the above method can supervise at least one dormant ticket corresponding to each ticket influencing factor, which can not only improve the supervision speed and efficiency of batch dormant tickets, but also improve the timeliness of the supervision of each dormant ticket; in addition, the above method does not require the participation of a neural network model when supervising dormant tickets, which can avoid the optimization process of the neural network model and the problem of inaccurate supervision results caused by overfitting of the neural network model optimization process, thereby improving the timeliness and accuracy of the supervision of dormant tickets, and can also reduce the complexity of the supervision process of each dormant ticket, reduce the difficulty of batch dormant ticket supervision, and greatly improve the speed and efficiency of batch dormant ticket supervision.
[0011] In one embodiment, obtaining the impact factor measurement index corresponding to each standard traffic penalty information includes:
[0012] Conduct principal component analysis based on each standard traffic penalty information to determine the initial ticket influencing factors corresponding to each standard traffic penalty information;
[0013] According to the factor loading matrix and the rotational interpretation information of each initial ticket influencing factor, multiple ticket influencing factors are screened from each initial ticket influencing factor;
[0014] According to each rotated interpretation information and each standard traffic penalty information, the influence factor measurement index of each ticket influencing factor is obtained.
[0015] The technical solution in the embodiment of the present application performs principal component analysis based on each standard traffic penalty information, determines the initial ticket influencing factor corresponding to each standard traffic penalty information, screens multiple ticket influencing factors from each initial ticket influencing factor based on the factor loading matrix of each initial ticket influencing factor and the rotated interpretation information, and obtains the influence factor measurement index of each ticket influencing factor based on each rotated interpretation information and each standard traffic penalty information; the above method can obtain the influence factor measurement index of each ticket influencing factor through each standard traffic penalty information that meets the reliability evaluation conditions, thereby making the influence factor measurement index of each ticket influencing factor of the obtained dormant tickets more accurate, and the process does not require human participation, and can also improve the speed and efficiency of obtaining the influence factor measurement index of each ticket influencing factor.
[0016] In one embodiment, based on the factor loading matrix and the rotation interpretation information of each initial ticket influencing factor, multiple ticket influencing factors are screened from each initial ticket influencing factor, including:
[0017] According to each rotated factor loading matrix, multiple target ticket variables are obtained from multiple preset ticket variables;
[0018] Based on each rotated interpretation information, multiple candidate ticket influencing factors are selected from each initial ticket influencing factor;
[0019] According to the factor loading matrix corresponding to each candidate ticket influencing factor and the corresponding rotational interpretation information, the factor score of each target ticket variable under each candidate ticket influencing factor is determined;
[0020] Based on the scores of each factor, each ticket influencing factor is determined from the candidate ticket influencing factors.
[0021] The technical solution in the embodiment of the present application obtains multiple target ticket variables from multiple preset ticket variables based on each rotated factor loading matrix, screens multiple candidate ticket influencing factors from each initial ticket influencing factor based on each rotated explanatory information, determines the factor score of each target ticket variable under each candidate ticket influencing factor based on the factor loading matrix corresponding to each candidate ticket influencing factor and the corresponding rotated explanatory information, and determines each ticket influencing factor from each candidate ticket influencing factor based on each factor score; the above method can obtain ticket influencing factors with a high correlation with each standard traffic penalty information from the initial ticket influencing factors through double-layer screening, so as to provide reference information for subsequently improving the accuracy of clustering results.
[0022] In one embodiment, before obtaining a plurality of target ticket variables from a plurality of preset ticket variables according to each rotated factor loading matrix, the method further comprises:
[0023] Perform the maximum variance rotation on each standard traffic penalty information to obtain the factor loading matrix of each initial ticket influencing factor under multiple preset ticket variables;
[0024] The factor loading matrix corresponding to each initial ticket influencing factor is factor rotated to obtain the rotated factor loading matrix of each initial ticket influencing factor.
[0025] The technical solution in the embodiment of the present application performs maximum variance rotation on each standard traffic penalty information to obtain the factor loading matrix of each initial ticket influencing factor under multiple preset ticket variables, and performs factor rotation on the factor loading matrix corresponding to each initial ticket influencing factor to obtain the rotated factor loading matrix of each initial ticket influencing factor; the above method does not require the participation of complex algorithms, thereby reducing the complexity of obtaining the rotated factor loading matrix and improving the speed and efficiency of obtaining the rotated factor loading matrix.
[0026] In one embodiment, according to each rotated factor loading matrix, a plurality of target ticket variables are obtained from a plurality of preset ticket variables, including:
[0027] According to the rotated factor loading matrix of each initial ticket influencing factor, the common degree of each ticket variable explained by the common factor is determined; the common factor represents the factors that exist in common in each standard traffic penalty information;
[0028] The ticket variables corresponding to multiple commonalities greater than a preset threshold among the commonalities are determined as target ticket variables.
[0029] The technical solution in the embodiment of the present application determines the commonality of each ticket variable explained by the common factor based on the rotated factor loading matrix of each initial ticket influencing factor, and determines the ticket variables corresponding to multiple commonalities greater than a preset threshold among the commonalities as the target ticket variables; the processing process of the above method is relatively simple and does not require the participation of complex algorithms, thereby improving the speed and efficiency of obtaining the target ticket variables, and the process does not require manual participation, which can reduce manually generated errors, thereby improving the correctness of the obtained target ticket variables, so as to provide reliable reference information for further accelerating the speed of clustering processing.
[0030] In one embodiment, the factor score of each target ticket variable under each candidate ticket influencing factor is determined according to the factor loading matrix corresponding to each candidate ticket influencing factor and the corresponding rotated interpretation information, including:
[0031] According to the rotated factor loading matrix corresponding to each candidate ticket influencing factor, determine the component score coefficient matrix of each candidate ticket influencing factor under each target ticket variable;
[0032] According to the component score coefficient matrix, the factor score of each target ticket variable under each candidate ticket influencing factor is obtained.
[0033] The technical solution in the embodiment of the present application determines the component score coefficient matrix of each candidate ticket influencing factor under each target ticket variable according to the rotated factor loading matrix corresponding to each candidate ticket influencing factor, and obtains the factor score of each target ticket variable under each candidate ticket influencing factor according to the component score coefficient matrix; the above method does not require complex algorithm processing, thereby reducing the complexity of obtaining factor scores, improving the speed and efficiency of obtaining factor scores, and at the same time, it can also improve the accuracy of the obtained factor scores. On this basis, it is possible to screen out ticket influencing factors with a high correlation with each standard traffic penalty information from multiple candidate ticket influencing factors based on the factor score of each target ticket variable under each candidate ticket influencing factor, thereby providing reference information for subsequently improving the accuracy of clustering results.
[0034] In one embodiment, each impact factor measurement indicator includes the contribution degree of each standard traffic penalty information to the construction of each ticket impact factor and / or the correlation described by each ticket impact factor; according to each rotated interpretation information and each standard traffic penalty information, the impact factor measurement indicator of each ticket impact factor is obtained, including:
[0035] Performing linear combination processing on each standard traffic penalty information to obtain the contribution degree of each ticket influencing factor; and / or,
[0036] For any ticket influencing factor, the comprehensive score coefficient of the ticket influencing factor is determined according to the contribution degree of the ticket influencing factor and the corresponding rotation interpretation information;
[0037] According to the comprehensive score coefficient of the ticket influencing factors, the correlation of the ticket influencing factors is determined.
[0038] The technical solution in the embodiment of the present application performs linear combination processing on each standard traffic penalty information to obtain the contribution degree of each ticket influencing factor. For any ticket influencing factor, the comprehensive score coefficient of the ticket influencing factor is determined according to the contribution degree of the ticket influencing factor and the corresponding rotated interpretation information, and the correlation of the ticket influencing factor is determined according to the comprehensive score coefficient of the ticket influencing factor. The above method does not require the participation of complex algorithms, can improve the speed and efficiency of obtaining the correlation degree of each ticket influencing factor, and on this basis can provide reference information for subsequently improving the accuracy of clustering results.
[0039] In one embodiment, the impact factor measurement index includes a contribution degree and / or a correlation degree; and according to each standard traffic penalty information and each impact factor measurement index, at least one dormant ticket corresponding to each ticket impact factor is supervised, including:
[0040] Inputting each standard traffic penalty information, each contribution degree and / or each correlation degree into the dormant ticket cluster analysis model to determine at least one standard traffic penalty information under each ticket influencing factor;
[0041] Classify and supervise the dormant tickets corresponding to different standard traffic penalty information under each ticket influencing factor.
[0042] The technical solution in the embodiment of the present application inputs each standard traffic penalty information, as well as each contribution degree and / or each correlation degree into the dormant ticket clustering analysis model, determines at least one standard traffic penalty information under each ticket influence factor, and classifies and supervises the dormant tickets corresponding to different standard traffic penalty information under each ticket influence factor; the above method can assist in clustering each standard traffic penalty information by constructing the contribution degree of each ticket influence factor through each standard traffic penalty information and / or the correlation described by each ticket influence factor, which can improve the clustering speed, thereby speeding up the classification supervision of each dormant ticket and improving the timeliness of the supervision of each dormant ticket.
[0043] In one embodiment, the method further comprises:
[0044] The reliability test of factor analysis was carried out on each standard traffic penalty information, and the test results were obtained;
[0045] If the test result is that the reliability passes, it is determined that each standard traffic penalty information meets the reliability evaluation conditions.
[0046] The technical solution in the embodiment of the present application performs a reliability test of factor analysis on each standard traffic penalty information to obtain a test result. If the test result is that the reliability passes, it is determined that each standard traffic penalty information meets the reliability assessment conditions. The above method can perform a reliability test on each standard traffic penalty information. Only when the reliability passes, subsequent processing will be performed based on each standard traffic penalty information, thereby improving the accuracy of the dormant ticket supervision results.
[0047] In a second aspect, an embodiment of the present application provides a ticket supervision device, which includes:
[0048] An information acquisition module, used to obtain standard traffic penalty information of each driving user's sleeping ticket;
[0049] The analysis module is used to obtain the impact factor measurement index corresponding to each standard traffic penalty information when each standard traffic penalty information meets the preset reliability evaluation conditions; the impact factor measurement index represents the importance of each standard traffic penalty information when described by multiple different ticket impact factors;
[0050] The supervision module is used to supervise at least one dormant ticket corresponding to each ticket impact factor according to each standard traffic penalty information and each impact factor measurement index.
[0051] In a third aspect, an embodiment of the present application further provides a computer device, comprising a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of the method of any one of the embodiments of the first aspect when executing the computer program.
[0052] In a fourth aspect, an embodiment of the present application further provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the steps of the method of any embodiment in the above-mentioned first aspect are implemented.
[0053] In a fifth aspect, an embodiment of the present application further provides a computer program product, the computer program product comprising a computer program, which, when executed by a processor, implements the steps of the method of any embodiment of the first aspect above.
[0054] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below. BRIEF DESCRIPTION OF THE DRAWINGS
[0055] Figure 1 A diagram showing an application environment of a ticket supervision method in an embodiment;
[0056] Figure 2 A flowchart of a ticket supervision method in one embodiment;
[0057] Figure 3 A schematic diagram of the structure of a ring buffer in one embodiment;
[0058] Figure 4 A flowchart of a ticket supervision method in another embodiment;
[0059] Figure 5 A scree diagram is shown in one embodiment;
[0060] Figure 6 A flowchart of a ticket supervision method in another embodiment;
[0061] Figure 7 A flowchart of a ticket supervision method in another embodiment;
[0062] Figure 8 A flowchart of a ticket supervision method in another embodiment;
[0063] Fig. 9 A flowchart of a ticket supervision method in another embodiment;
[0064] Fig.10 A flowchart of a ticket supervision method in another embodiment;
[0065] Fig.11 A flowchart of a ticket supervision method in another embodiment;
[0066] Fig.12 A structural block diagram of a ticket supervision device in one embodiment;
[0067] Fig.13 FIG. 4 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION
[0068] 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.
[0069] In the field of traffic, drivers often violate traffic rules and commit traffic violations while driving their vehicles. Correspondingly, the traffic control department will issue traffic penalty tickets to punish the driver and correct the traffic violation. In practical applications, there are often tickets that the driver has not confirmed or has not fulfilled the penalty after confirmation. This type of penalty ticket is called a sleeping ticket. Among them, the existence of sleeping tickets may cause great hidden dangers to road traffic safety. Therefore, it is particularly important to supervise sleeping tickets. In the related art, it is mainly to obtain a sleeping ticket supervision model, and then train the sleeping ticket supervision model, and then use the trained sleeping ticket supervision model to supervise batches of sleeping tickets. However, in the related art, when supervising sleeping tickets, the sleeping ticket supervision model is prone to overfitting, so there will be problems with inaccurate supervision. Based on this, the embodiment of the present application provides a ticket supervision method, which can improve the accuracy of supervising sleeping tickets when supervising sleeping tickets.
[0070] The ticket supervision method provided in the embodiment of the present application can be applied to Figure 1The ticket supervision system shown may include a computer device and multiple information management devices, each of which is used to manage the original traffic penalty information of multiple traffic penalty tickets. Optionally, the computer device and each information management device may be connected by communication, and the communication method may be Wi-Fi, mobile network or Bluetooth connection, etc. Optionally, the above-mentioned computer device may be a personal computer, desktop computer, laptop computer, tablet computer, smart phone, smart bracelet or other electronic device with data processing function; the above-mentioned information management device may be an electronic device with information storage function, such as a database, data center, electronic device, etc. This embodiment does not limit the specific form of the computer device and each information management device. Among them, Figure 1 The ticket supervision system is illustrated by taking the computer device and each information management device as a personal computer as an example, and the ticket supervision system includes three information management devices (i.e., information management device 1, information management device 2, and information management device 3). The specific process of the ticket supervision method is introduced below by taking the computer device as the execution subject of the ticket supervision method.
[0071] like Figure 2 As shown, it is a flow chart of the ticket supervision method provided in the embodiment of the present application, and the method can be implemented by the following steps:
[0072] S100: Obtain standard traffic penalty information of each driving user's sleeping ticket.
[0073] In actual applications, if any driving user fails to confirm or process the traffic penalty ticket generated during his driving, the traffic penalty ticket can be defined as a dormant ticket; the vehicle driven by the driving user can be a motor vehicle (such as a private car, truck, bus, taxi, etc.) or a non-motor vehicle (such as an electric car, motorcycle, etc.). Optionally, the area where the traffic violation behavior corresponding to the dormant ticket of each driving user can be generated can be a town, village, county, etc., which is not limited in this embodiment of the application.
[0074] Specifically, the computer device can obtain the standard traffic penalty information of each driving user's dormant ticket in real time from local, cloud, disk, memory and other locations; it can also be the standard traffic penalty information of each driving user's dormant ticket directly received from various information management systems.
[0075] Here, the standard traffic penalty information of the sleeping ticket may be information obtained by processing the original traffic penalty information of the sleeping ticket collected by each information management system.
[0076] In an embodiment of the present application, the computer device can first obtain the original traffic penalty information of all driving users in a specific area, and filter out the original traffic penalty information of unpaid fines from the original traffic penalty information as candidate traffic penalty information for traffic tickets (i.e., original traffic penalty information such as the plaintiff information in the application for administrative reconsideration, administrative litigation judgment, administrative litigation, and non-litigation administrative execution ruling). Then, all the screened candidate traffic penalty information is correlated and collided to obtain the candidate traffic penalty information corresponding to the dormant tickets for which the fines have not been paid, administrative reconsideration has not been applied for, administrative litigation has not been filed, and compulsory execution has not been applied for after the expiration of the legal period.
[0077] Optionally, the above-mentioned original traffic penalty information may include traffic violation information, traffic violation ticket information, loss of insurance and review information, traffic violation points information, owner additional penalty information, traffic violation fine information, traffic violation collection collection information, etc.
[0078] Furthermore, the candidate traffic penalty information corresponding to each dormant ticket may be processed by at least one of dirty data cleaning, missing value filling, outlier removal, data format normalization, and standardization to reduce the dimension of the candidate traffic penalty information and reduce the difficulty and complexity of subsequent processing.
[0079] S200: When each standard traffic penalty information meets the preset reliability evaluation condition, obtain the impact factor measurement index corresponding to each standard traffic penalty information, wherein the impact factor measurement index represents the importance of each standard traffic penalty information when described by multiple different ticket impact factors.
[0080] In the embodiment of the present application, when the computer device determines that each standard traffic penalty information meets the preset reliability evaluation conditions, it is effective or meaningful to supervise the dormant tickets by executing the subsequent process through each standard traffic penalty information. Optionally, each ticket impact factor can be a plurality of impact factors of dormant tickets customized according to each standard traffic penalty information, or can be determined by corresponding processing according to each standard traffic penalty information.
[0081] In one embodiment, the computer device can perform arithmetic operations, comparisons, matching and / or analysis on each standard traffic penalty information, and determine whether each standard traffic penalty information meets the reliability evaluation conditions for executing subsequent processes based on the processing results. Optionally, the arithmetic operations can be implemented by at least one of addition, subtraction, logarithmic, exponential, multiplication, and division.
[0082] In another embodiment, the computer device may also pre-train an algorithm model, and then input each standard traffic penalty information into the algorithm model, and the algorithm model outputs a result of whether each standard traffic penalty information meets the reliability evaluation conditions.
[0083] In the embodiment of the present application, the computer device can perform a correlation analysis on each standard traffic penalty information to obtain an impact factor measurement index corresponding to each standard traffic penalty information. Optionally, the above correlation analysis can be a correlation analysis method, a regression analysis method, a correspondence analysis method, etc., but in the embodiment of the present application, the correlation analysis can be a factor analysis.
[0084] Optionally, the above-mentioned ticket influencing factors can be understood as the influencing factors that cause the driving user to generate a dormant ticket, such as driving without a license, speeding, running a red light, illegal parking, not wearing a seat belt, driving on the line, driving under the influence of alcohol, time, weather and other factors.
[0085] In practical applications, the impact factor measurement index of the ticket impact factor can indicate the importance of each standard traffic penalty information being able to be summarized as each ticket impact factor.
[0086] Specifically, the computer device may perform arithmetic operations, comparisons, matching and / or analysis on each standard traffic penalty information to obtain an impact factor measurement index corresponding to each standard traffic penalty information.
[0087] S300: According to each standard traffic penalty information and each impact factor measurement index, supervise at least one dormant ticket corresponding to each ticket impact factor.
[0088] In one implementation, a computer device can pre-train a supervision strategy prediction model, and then input each standard traffic penalty information and each influencing factor measurement indicator into the supervision strategy prediction model. The supervision strategy prediction model outputs a supervision strategy, and then supervises at least one dormant ticket corresponding to each ticket influencing factor according to the supervision strategy.
[0089] Optionally, the above-mentioned supervision strategy prediction model can be implemented by at least one of a convolutional neural network model, a fully connected neural network model, a long short-term memory neural network model, a recurrent neural network model, etc.
[0090] In another implementation, the computer device can analyze and process each standard traffic penalty information and each influencing factor measurement indicator, build a supervision strategy based on the analysis and processing results, and then supervise at least one dormant ticket corresponding to each ticket influencing factor according to the supervision strategy.
[0091] It should be noted here that at least one dormant ticket corresponding to each ticket impact factor may belong to the same category of dormant tickets. In practical applications, at least one dormant ticket corresponding to each ticket impact factor may be subject to batch supervision, that is, all dormant tickets may be subject to classified supervision.
[0092] The technical solution in the embodiment of the present application obtains standard traffic penalty information of dormant tickets of each driving user, and when each standard traffic penalty information meets the preset reliability evaluation conditions, obtains the impact factor measurement index corresponding to each standard traffic penalty information, and supervises at least one dormant ticket corresponding to each ticket impact factor based on each standard traffic penalty information and each impact factor measurement index, wherein the impact factor measurement index represents the importance of each standard traffic penalty information when described by multiple different ticket impact factors; the above method can perform supervision on each dormant ticket based on each standard traffic penalty information only when it is determined that each standard traffic penalty information meets the preset reliability evaluation conditions, so that the supervision of dormant tickets can be guaranteed. The reliability of the results can be improved, and the accuracy of the supervision of dormant tickets can be improved; at the same time, the above method can supervise at least one dormant ticket corresponding to each ticket influencing factor, which can not only improve the supervision speed and efficiency of batch dormant tickets, but also improve the timeliness of the supervision of each dormant ticket; in addition, the above method does not require the participation of a neural network model when supervising dormant tickets, which can avoid the optimization process of the neural network model and the problem of inaccurate supervision results caused by overfitting of the neural network model optimization process, thereby improving the timeliness and accuracy of the supervision of dormant tickets, and can also reduce the complexity of the supervision process of each dormant ticket, reduce the difficulty of batch dormant ticket supervision, and greatly improve the speed and efficiency of batch dormant ticket supervision.
[0093] The following is an explanation of the process of obtaining the impact factor measurement index corresponding to each standard traffic penalty information. Figure 3 As shown, the steps in S200 above can be implemented in the following manner:
[0094] S210 , performing principal component analysis based on each standard traffic penalty information to determine the initial ticket influencing factor corresponding to each standard traffic penalty information.
[0095] Specifically, the computer device can perform principal component analysis on each standard traffic penalty information according to the principal component analysis method to determine the initial ticket influencing factor corresponding to each standard traffic penalty information. Optionally, the principal component analysis method can be a principal component analysis method based on eigenvalue decomposition, a kernel principal component analysis method, a dynamic principal component analysis method, a sparse principal component analysis method, a principal component analysis method based on singular value decomposition, and / or a robust principal component analysis method, etc.
[0096] In addition, the computer device can also pre-train a principal component analysis model, and then input each standard traffic penalty information into the principal component analysis model, and the principal component analysis model outputs the initial ticket impact factor corresponding to each standard traffic penalty information. Optionally, the principal component analysis model can be implemented by at least one of a decision tree, a multi-layer perceptron, a convolutional neural network model, a fully connected neural network model, a long short-term memory neural network model, a recurrent neural network model, an attention mechanism model, etc.
[0097] S220. Filter multiple ticket influencing factors from each initial ticket influencing factor according to the factor loading matrix and the rotated interpretation information of each initial ticket influencing factor.
[0098] In practical applications, the computer device can obtain the factor loading matrix and the rotated explanatory information of each initial ticket influencing factor, and then analyze, compare and / or match the factor loading matrix and the rotated explanatory information of each initial ticket influencing factor, and comprehensively determine to select multiple ticket influencing factors from each initial ticket influencing factor based on the processing results of the two.
[0099] In addition, the computer device can also determine whether the factor loading matrix and the rotated interpretation information of each initial ticket influencing factor meet the screening rules, and determine at least one initial ticket influencing factor that meets the screening rules as the ticket influencing factor. Optionally, the above screening rules can include the factor loading matrix and the rotated interpretation information of each ticket influencing factor to be screened.
[0100] It should be noted that before executing the steps in S220, each standard traffic penalty information may be roughly classified to obtain multiple types of standard traffic penalty information, and then the rotated explanation information of each initial ticket influencing factor may be calculated using the variance formula according to each type of standard traffic penalty information; the rotated explanation information here may include characteristic roots, rotated variance explanation rate, and rotated cumulative total variance explanation rate. Optionally, the above-mentioned rough classification method may be at least one of feature extraction, similarity matching, division, and comparison.
[0101] S230: Obtain an influence factor measurement index of each ticket influence factor according to each rotated interpretation information and each standard traffic penalty information.
[0102] Furthermore, for any ticket influencing factor, the computer device may perform arithmetic operations on the rotated interpretation information of the ticket influencing factor and each standard traffic penalty information to obtain an influence factor measurement index of the ticket influencing factor.
[0103] In addition, for any ticket influencing factor, the computer device can also pre-train an algorithm model, and then input the rotated interpretation information of the ticket influencing factor and various standard traffic penalty information into the algorithm model, and the algorithm model outputs the influence factor measurement index of the ticket influencing factor.
[0104] The technical solution in the embodiment of the present application performs principal component analysis based on each standard traffic penalty information, determines the initial ticket influencing factor corresponding to each standard traffic penalty information, screens multiple ticket influencing factors from each initial ticket influencing factor based on the factor loading matrix of each initial ticket influencing factor and the rotated interpretation information, and obtains the influence factor measurement index of each ticket influencing factor based on each rotated interpretation information and each standard traffic penalty information; the above method can obtain the influence factor measurement index of each ticket influencing factor through each standard traffic penalty information that meets the reliability evaluation conditions, thereby making the influence factor measurement index of each ticket influencing factor of the obtained dormant tickets more accurate, and the process does not require human participation, and can also improve the speed and efficiency of obtaining the influence factor measurement index of each ticket influencing factor.
[0105] The following is a description of the process of selecting multiple ticket influencing factors from each initial ticket influencing factor based on the factor loading matrix and the rotated interpretation information of each initial ticket influencing factor. Figure 4 As shown, the steps in the above S220 may include:
[0106] S221. According to each rotated factor loading matrix, obtain multiple target ticket variables from multiple preset ticket variables.
[0107] In practical applications, the computer device can compare, perform arithmetic operations, analyze, and other processing on each rotated factor loading matrix, and obtain multiple target ticket variables from multiple preset ticket variables according to the processing results. Optionally, the preset ticket variables can be reference variables for screening dormant tickets from multiple traffic penalty tickets.
[0108] At the same time, the computer device can also pre-train an algorithm model, and then input each rotated factor loading matrix into the algorithm model, and the algorithm model outputs multiple target ticket variables obtained from multiple preset ticket variables.
[0109] S222: Based on each rotated interpretation information, select a plurality of candidate ticket influencing factors from each initial ticket influencing factor.
[0110] Furthermore, the computer device can compare, perform arithmetic operations, analyze, and other processing on each rotated interpretation information, and screen a plurality of candidate ticket influencing factors from each initial ticket influencing factor based on the processing results.
[0111] In addition, the computer device can pre-train an algorithm model, and then input each rotated interpretation information into the algorithm model, and the algorithm model outputs a plurality of candidate ticket influencing factors selected from each initial ticket influencing factor.
[0112] At the same time, the computer device can also obtain the characteristic roots in the rotated interpretation information of each initial ticket influence factor, and then compare the characteristic values, and determine the initial ticket influence factors corresponding to at least two characteristic roots with smaller size differences as candidate ticket influence factors.
[0113] In an embodiment of the present application, the computer device can also generate a scree plot based on the acquired characteristic roots of each initial ticket influence factor, and then determine to screen multiple candidate ticket influence factors from each initial ticket influence factor based on the smooth line segments on the broken lines in the scree plot; specifically, each initial ticket influence factor on the smooth line segment can be determined as each candidate ticket influence factor.
[0114] For example, there are five initial ticket influencing factors. The scree plot generated based on the initial ticket influencing factors and the characteristic roots of the initial ticket influencing factors is as follows: Figure 5 As shown, Figure 5 The horizontal axis represents the identification of the initial ticket influencing factor, and the vertical axis represents the characteristic root of the initial ticket influencing factor. Figure 5 If the broken line segment among initial ticket influence factor 2, initial ticket influence factor 3 and initial ticket influence factor 4 is a stable line segment, initial ticket influence factor 2, initial ticket influence factor 3 and initial ticket influence factor 4 can be determined as candidate ticket influence factors.
[0115] In the embodiment of the present application, the execution order of the above steps S221 and S222 can be interchanged without limitation.
[0116] S223. Determine the factor score of each target ticket variable under each candidate ticket influencing factor according to the factor loading matrix corresponding to each candidate ticket influencing factor and the corresponding rotated explanatory information.
[0117] Furthermore, arithmetic operations can be performed based on the factor loading matrix corresponding to each candidate ticket influencing factor and the corresponding rotated explanation information to obtain the factor score of each target ticket variable under each candidate ticket influencing factor. Optionally, the factor score under the candidate ticket influencing factor can represent the significance of each standard traffic penalty information under the candidate ticket influencing factor.
[0118] S224. Determine each ticket impact factor from each candidate ticket impact factor based on the score of each factor.
[0119] Furthermore, the scores of each factor may be sorted and / or compared, and at least one of the candidate ticket influencing factors may be selected as the ticket influencing factor.
[0120] The technical solution in the embodiment of the present application obtains multiple target ticket variables from multiple preset ticket variables based on each rotated factor loading matrix, screens multiple candidate ticket influencing factors from each initial ticket influencing factor based on each rotated explanatory information, determines the factor score of each target ticket variable under each candidate ticket influencing factor based on the factor loading matrix corresponding to each candidate ticket influencing factor and the corresponding rotated explanatory information, and determines each ticket influencing factor from each candidate ticket influencing factor based on each factor score; the above method can obtain ticket influencing factors with a high correlation with each standard traffic penalty information from the initial ticket influencing factors through double-layer screening, so as to provide reference information for subsequently improving the accuracy of clustering results.
[0121] In one embodiment, before executing the step in S221 above, Figure 6 As shown, the above method can also be implemented by the following steps:
[0122] S225. Perform maximum variance rotation on each standard traffic penalty information to obtain a factor loading matrix of each initial ticket influencing factor under multiple preset ticket variables.
[0123] In practical applications, the computer device can use the maximum variance method to perform maximum variance rotation on each standard traffic penalty information to obtain the factor loading coefficient of each initial ticket influencing factor under each preset ticket variable, and then combine the factor loading coefficients of each initial ticket influencing factor under each preset ticket variable to generate a factor loading matrix. Optionally, the maximum variance method can be a maximum variance normalization method, a maximum variance filtering method, and / or a maximum variance projection, etc.
[0124] Among them, the factor loading matrix of the above-mentioned initial ticket influencing factors under multiple preset ticket variables may include the factor loading coefficient of each initial ticket influencing factor under each preset ticket variable.
[0125] S226. Perform factor rotation on the factor loading matrix corresponding to each initial ticket influencing factor to obtain a rotated factor loading matrix of each initial ticket influencing factor.
[0126] The computer device may use a factor rotation method to perform factor rotation on the factor loading matrix corresponding to each initial ticket influencing factor. Optionally, the factor rotation method may be a varimax rotation method, a quarticmax rotation method, a direct oblique rotation method, a Promax rotation method, or the like.
[0127] In the embodiment of the present application, the rotated factor loading matrix of the above-mentioned initial ticket influencing factors may include the rotated factor loading coefficient of each initial ticket influencing factor.
[0128] The technical solution in the embodiment of the present application performs maximum variance rotation on each standard traffic penalty information to obtain the factor loading matrix of each initial ticket influencing factor under multiple preset ticket variables, and performs factor rotation on the factor loading matrix corresponding to each initial ticket influencing factor to obtain the rotated factor loading matrix of each initial ticket influencing factor; the above method does not require the participation of complex algorithms, thereby reducing the complexity of obtaining the rotated factor loading matrix and improving the speed and efficiency of obtaining the rotated factor loading matrix.
[0129] The following is an explanation of the process of obtaining multiple target ticket variables from multiple preset ticket variables based on each rotated factor loading matrix. Figure 7 As shown, the steps in S221 above can be implemented in the following manner:
[0130] S2211. Determine the commonality of each ticket variable explained by the common factor based on the rotated factor loading matrix of each initial ticket influencing factor, wherein the common factor represents the factor that exists in common in each standard traffic penalty information.
[0131] In practical applications, the rotated factor loading matrix may include the rotated factor loading coefficients of different initial ticket influencing factors. Among them, the commonality explained by the common factor for any ticket variable can be obtained by calculating the sum of squares of the rotated factor loading coefficients of different initial ticket influencing factors under the ticket variable. For example, if the commonality of the i-th target ticket variable is h_i, the commonality h_i can be equal to the sum of squares of the i-th row elements in the rotated factor loading matrix of each initial ticket influencing factor.
[0132] At the same time, taking two initial ticket influencing factors (initial ticket influencing factor 1 and initial ticket influencing factor 2) and five ticket variables 9 (including ticket collection frequency, fine amount, cumulative deduction points, number of complaints and ticket processing completion) as examples, the rotated factor loading coefficients of different initial ticket influencing factors under different ticket variables and the commonality explained by the common factors of each ticket variable can be shown in Table 1.
[0133] Table 1
[0134]
[0135] S2212. Determine the ticket variables corresponding to multiple commonalities greater than a preset threshold among the commonalities as target ticket variables.
[0136] In the embodiment of the present application, the computer device can determine whether each commonality is greater than a preset threshold, and determine the ticket variable corresponding to the commonality greater than the preset threshold as the target ticket variable. Optionally, the preset threshold can be user-defined or determined based on historical experience; in the embodiment of the present application, the preset threshold can be equal to 0.4.
[0137] Continuing with the previous example, the commonality of 0.381 corresponding to the completion of ticket processing is less than 0.4. Therefore, the completion of ticket processing can be removed, and the remaining ticket collection frequency, fine amount, accumulated deduction points and number of complaints can be used as target ticket variables.
[0138] The technical solution in the embodiment of the present application determines the commonality of each ticket variable explained by the common factor based on the rotated factor loading matrix of each initial ticket influencing factor, and determines the ticket variables corresponding to multiple commonalities greater than a preset threshold among the commonalities as the target ticket variables; the processing process of the above method is relatively simple and does not require the participation of complex algorithms, thereby improving the speed and efficiency of obtaining the target ticket variables, and the process does not require manual participation, which can reduce manually generated errors, thereby improving the correctness of the obtained target ticket variables, so as to provide reliable reference information for further accelerating the speed of clustering processing.
[0139] The following is an explanation of the process of determining the factor score of each target ticket variable under each candidate ticket influencing factor based on the factor loading matrix corresponding to each candidate ticket influencing factor and the corresponding rotated interpretation information. Figure 8 As shown, the steps in S223 above can be implemented in the following manner:
[0140] S2231. According to the rotated factor loading matrix corresponding to each candidate ticket influencing factor, determine the component score coefficient matrix of each candidate ticket influencing factor under each target ticket variable.
[0141] Specifically, the computer device can pre-train an algorithm model, and then input the rotated factor loading matrix corresponding to each candidate ticket influencing factor into the algorithm model, and the algorithm model outputs the component score coefficient matrix of each candidate ticket influencing factor under each target ticket variable.
[0142] In the embodiment of the present application, the computer device can solve the inverse matrix according to the rotation factor loading matrix corresponding to each candidate ticket influencing factor to obtain the component score coefficient matrix of each candidate ticket influencing factor under each target ticket variable. The above-mentioned component score coefficient matrix may include the component score coefficient of each candidate ticket influencing factor under each target ticket variable.
[0143] For example, if the rotation factor loading matrix corresponding to each candidate ticket influencing factor is A, then solving the inverse matrix according to A yields (A T A) -1 ·A T , that is, the component score coefficient matrix of each candidate ticket influencing factor under each target ticket variable; where A T represents the transposed matrix of A. Table 2 below is a component score coefficient matrix obtained in one scenario.
[0144] Table 2
[0145]
[0146] S2232. Obtain the factor score of each target ticket variable under each candidate ticket influencing factor according to the component score coefficient matrix.
[0147] Furthermore, arithmetic operations and / or analysis processing may be performed based on the component score coefficient matrix to obtain the factor scores of each target ticket variable under each candidate ticket influencing factor.
[0148] In an embodiment of the present application, for any candidate ticket influence factor, the factor score of each target ticket variable under the candidate ticket influence factor can be equal to the sum of the products of the component score coefficient of the candidate ticket influence factor under each target ticket variable and the specific value of the target ticket variable.
[0149] Continuing with the previous example, as shown in Table 2, for candidate ticket influence factor 1, the factor score of each target ticket variable under candidate ticket influence factor 1 is equal to the product of the specific value of the ticket collection frequency and the component score coefficient of candidate ticket influence factor 1 under the ticket collection frequency -0.106, the specific value of the fine amount and the component score coefficient of candidate ticket influence factor 1 under the fine amount 0.017, the specific value of the cumulative deductions and the component score coefficient of candidate ticket influence factor 1 under the cumulative deductions 0.449, and the specific value of the number of complaints and the component score coefficient of candidate ticket influence factor 1 under the number of complaints 0.505.
[0150] At the same time, for candidate ticket influence factor 2, the factor score of each target ticket variable under candidate ticket influence factor 2 is equal to the product of the specific value of the ticket collection frequency and the component score coefficient of candidate ticket influence factor 2 under the ticket collection frequency of 0.722, the specific value of the fine amount and the component score coefficient of candidate ticket influence factor 2 under the fine amount of 0.596, the specific value of the cumulative deduction points and the component score coefficient of candidate ticket influence factor 2 under the cumulative deduction points of 0.112, and the specific value of the number of complaints and the component score coefficient of candidate ticket influence factor 2 under the number of complaints of -0.094.
[0151] The technical solution in the embodiment of the present application determines the component score coefficient matrix of each candidate ticket influencing factor under each target ticket variable according to the rotated factor loading matrix corresponding to each candidate ticket influencing factor, and obtains the factor score of each target ticket variable under each candidate ticket influencing factor according to the component score coefficient matrix; the above method does not require complex algorithm processing, thereby reducing the complexity of obtaining factor scores, improving the speed and efficiency of obtaining factor scores, and at the same time, it can also improve the accuracy of the obtained factor scores. On this basis, it is possible to screen out ticket influencing factors with a high correlation with each standard traffic penalty information from multiple candidate ticket influencing factors based on the factor score of each target ticket variable under each candidate ticket influencing factor, thereby providing reference information for subsequently improving the accuracy of clustering results.
[0152] The following is an explanation of the process of obtaining the impact factor measurement index of each ticket impact factor based on each rotated interpretation information and each standard traffic penalty information. In one embodiment, each impact factor measurement index includes the contribution degree of each standard traffic penalty information to the construction of each ticket impact factor and / or the correlation described by each ticket impact factor; Fig. 9 As shown, the steps in the above S231 may include:
[0153] S2311. Perform linear combination processing on each standard traffic penalty information to obtain the contribution degree of each ticket influencing factor.
[0154] Specifically, the computer device can obtain various types of standard traffic penalty information corresponding to each ticket influencing factor from each standard traffic penalty information, and perform at least one of least squares processing, weighted average processing, Kalman filtering processing, and principal component analysis processing on the various types of standard traffic penalty information corresponding to each ticket influencing factor to obtain the linear combination coefficient of each ticket influencing factor, that is, the contribution degree of each standard traffic penalty information to the construction of each ticket influencing factor.
[0155] In an embodiment of the present application, for the standard traffic penalty information corresponding to any ticket influence factor, a correlation coefficient matrix or a covariance matrix can be calculated for the standard traffic penalty information corresponding to the ticket influence factor, and then the correlation coefficient matrix or the covariance matrix can be eigenvalue decomposition performed to obtain the eigenvalue of the ticket influence factor, and then the contribution degree of the ticket influence factor can be obtained by dividing the correlation coefficient matrix or the covariance matrix by the square root of the eigenvalue.
[0156] S2312. For any ticket influencing factor, determine the comprehensive score coefficient of the ticket influencing factor according to the contribution degree of the ticket influencing factor and the corresponding rotated interpretation information.
[0157] In an embodiment of the present application, the rotational explanation information of the ticket influence factor may include the rotational variance explanation rate and the rotational cumulative variance explanation rate of the ticket influence factor.
[0158] Among them, for any ticket influencing factor, the computer device can perform arithmetic operations on the contribution degree of the ticket influencing factor, the corresponding variance explanation rate after rotation, and the cumulative variance explanation rate after rotation to obtain the comprehensive score coefficient of the ticket influencing factor.
[0159] In an embodiment of the present application, the computer device may multiply the contribution degree of the ticket influence factor by the variance explanation rate of the ticket influence factor after rotation, and then divide it by the cumulative variance explanation rate of the ticket influence factor after rotation to obtain the comprehensive score coefficient of the ticket influence factor.
[0160] S2313. Determine the correlation of the ticket influencing factors based on the comprehensive score coefficients of the ticket influencing factors.
[0161] Furthermore, the comprehensive score coefficient of the ticket impact factor can be converted, analyzed, mapped, etc. to obtain the correlation between each standard traffic penalty information described by the ticket impact factor. In the embodiment of the present application, the correlation between each standard traffic penalty information described by the ticket impact factor can be called the weight of each standard traffic penalty information described by the ticket impact factor.
[0162] In the embodiment of the present application, the computer device can normalize the comprehensive score coefficient of the ticket impact factor to obtain the correlation degree of the ticket impact factor. Fig. 9 The impact factor measurement indicators include contribution degree and correlation degree as an example.
[0163] The technical solution in the embodiment of the present application performs linear combination processing on each standard traffic penalty information to obtain the contribution degree of each ticket influencing factor. For any ticket influencing factor, the comprehensive score coefficient of the ticket influencing factor is determined according to the contribution degree of the ticket influencing factor and the corresponding rotated interpretation information, and the correlation of the ticket influencing factor is determined according to the comprehensive score coefficient of the ticket influencing factor. The above method does not require the participation of complex algorithms, can improve the speed and efficiency of obtaining the correlation degree of each ticket influencing factor, and on this basis can provide reference information for subsequently improving the accuracy of clustering results.
[0164] In one embodiment, the above-mentioned impact factor measurement indicators include contribution degree and / or correlation degree; Fig.10 As shown, the step of supervising at least one dormant ticket corresponding to each ticket impact factor according to each standard traffic penalty information and each impact factor measurement index in the above S300 may include:
[0165] S310: Input each standard traffic penalty information, each contribution degree and / or each correlation degree into a dormant ticket cluster analysis model to determine at least one standard traffic penalty information under each ticket influencing factor.
[0166] It should be noted here that the above-mentioned dormant ticket clustering analysis model can be a model built based on a clustering algorithm, and the clustering algorithm can be K-Means algorithm, hierarchical clustering, density clustering, mean shift clustering, spectral clustering or Gaussian mixture model clustering, etc.
[0167] In practical applications, the computer device can input each standard traffic penalty information, each contribution degree and / or each correlation degree into the dormant ticket cluster analysis model to determine at least one standard traffic penalty information under each ticket influencing factor. The dormant ticket cluster analysis model can classify all standard traffic penalty information into at least one category, and each category can include at least one standard traffic penalty information.
[0168] S320. Classify and supervise dormant tickets corresponding to different standard traffic penalty information under each ticket influencing factor.
[0169] In the embodiment of the present application, the sleeping tickets corresponding to the standard traffic penalty information under each category in the clustering results can be regarded as the same type of sleeping tickets, and then batch supervision of each type of sleeping tickets can be performed separately, which can improve the speed and efficiency of batch sleeping tickets, thereby improving the timeliness of batch sleeping tickets. It should be noted here that each ticket impact factor corresponds to a category.
[0170] In one embodiment, the method of classifying and supervising the dormant tickets corresponding to the different standard traffic penalty information under each ticket impact factor can be that the computer device sends reminder information or reminder signals to the user terminals of the driving users to whom the dormant tickets corresponding to the different standard traffic penalty information under each category belong, so as to inform the driving users to promptly handle the dormant tickets corresponding to them. Optionally, the reminder information can be output in the form of text messages, voice, video, email, screen pop-up windows, etc.; the reminder signal can be output in the form of vibration, flashing indicator light, beeping sound, etc.
[0171] In addition, before executing the steps in S320 above, the above method can also test the clustering results. When the clustering effect of the clustering results meets the requirements, continue to execute the steps in S320 based on the clustering results to improve the accuracy of the classification supervision results.
[0172] The technical solution in the embodiment of the present application inputs each standard traffic penalty information, as well as each contribution degree and / or each correlation degree into the dormant ticket clustering analysis model, determines at least one standard traffic penalty information under each ticket influence factor, and classifies and supervises the dormant tickets corresponding to different standard traffic penalty information under each ticket influence factor; the above method can assist in clustering each standard traffic penalty information by constructing the contribution degree of each ticket influence factor through each standard traffic penalty information and / or the correlation described by each ticket influence factor, which can improve the clustering speed, thereby speeding up the classification supervision of each dormant ticket and improving the timeliness of the supervision of each dormant ticket.
[0173] In practical applications, in order to accurately obtain the contribution of each standard traffic penalty information to the construction of each ticket impact factor and / or the correlation described by each ticket impact factor based on each standard traffic penalty information, the reliability of each standard traffic penalty information can be tested first to ensure that the effective contribution and / or correlation can be obtained through each standard traffic penalty information. The reliability test process is described below. In one embodiment, before executing the steps in S200 above, Fig.11 As shown, the above method may further include the following steps:
[0174] S400: Perform a reliability test of factor analysis on each standard traffic penalty information to obtain a test result.
[0175] In one embodiment, the reliability test of factor analysis on each standard traffic penalty information may be performed by calculating a KMO value for each standard traffic penalty information and using the KMO value as the test result.
[0176] In another embodiment, the reliability test of factor analysis on each standard traffic penalty information may also be performed by performing a Bartlett test on each standard traffic penalty information, and using the p value obtained by the Bartlett test as the test result.
[0177] In another embodiment, the reliability test of factor analysis for each standard traffic penalty information can also be performed by calculating the KMO value for each standard traffic penalty information, performing a sphericity test on each standard traffic penalty information, and using the KMO value and the p-value obtained by the sphericity test as the test result. The sphericity test here can be called a Bartlett test.
[0178] S500: If the test result is that the reliability passes, determine whether each standard traffic penalty information meets the reliability evaluation condition.
[0179] Further, it can be determined whether the KMO value is greater than or equal to 0.7, and / or the p-value obtained by the sphericity test is less than 0.05. When the KMO value is greater than or equal to 0.7 and / or the p-value is less than 0.05, the test result is determined to pass reliability.
[0180] The technical solution in the embodiment of the present application performs a reliability test of factor analysis on each standard traffic penalty information to obtain a test result. If the test result is that the reliability passes, it is determined that each standard traffic penalty information meets the reliability assessment conditions. The above method can perform a reliability test on each standard traffic penalty information. Only when the reliability passes, subsequent processing will be performed based on each standard traffic penalty information, thereby improving the accuracy of the dormant ticket supervision results.
[0181] In one embodiment, the present application also provides a ticket supervision method, which includes the following process:
[0182] (1) Obtain standard traffic penalty information for each driver's sleeping ticket;
[0183] (2) Conduct a reliability test on each standard traffic penalty information through factor analysis and obtain the test results;
[0184] (3) If the test result is reliability passed, it is determined that each standard traffic penalty information meets the reliability evaluation conditions;
[0185] (4) When each standard traffic penalty information meets the preset reliability evaluation conditions, a principal component analysis is performed based on each standard traffic penalty information to determine the initial ticket influencing factor corresponding to each standard traffic penalty information;
[0186] (5) Performing maximum variance rotation on each standard traffic penalty information to obtain the factor loading matrix of each initial ticket influencing factor under multiple preset ticket variables;
[0187] (6) Perform factor rotation on the factor loading matrix corresponding to each initial ticket influencing factor to obtain the rotated factor loading matrix of each initial ticket influencing factor;
[0188] (7) Based on the rotated factor loading matrix of each initial ticket influencing factor, determine the commonality of each ticket variable explained by the common factor; the common factor represents the factors that exist in all standard traffic penalty information;
[0189] (8) determining the ticket variables corresponding to multiple commonalities greater than a preset threshold among the commonalities as target ticket variables;
[0190] (9) based on each rotated interpretation information, select a plurality of candidate ticket influencing factors from each initial ticket influencing factor;
[0191] (10) Determine the component score coefficient matrix of each candidate ticket influencing factor under each target ticket variable based on the factor loading matrix corresponding to each candidate ticket influencing factor and the corresponding rotated explanatory information;
[0192] (11) According to the component score coefficient matrix, obtain the factor score of each target ticket variable under each candidate ticket influencing factor;
[0193] (12) determining each ticket influencing factor from each candidate ticket influencing factor based on the score of each factor;
[0194] (13) performing linear combination processing on each standard traffic penalty information to obtain the contribution degree of each ticket influencing factor; and / or,
[0195] (14) For any ticket influencing factor, determine the comprehensive score coefficient of the ticket influencing factor according to the contribution degree of the ticket influencing factor and the corresponding rotated interpretation information;
[0196] (15) Determine the correlation of the ticket influencing factors based on the comprehensive score coefficients of the ticket influencing factors;
[0197] (16) inputting each standard traffic penalty information, each contribution degree and / or each correlation degree into the dormant ticket cluster analysis model to determine at least one standard traffic penalty information under each ticket influencing factor;
[0198] (17) Classify and supervise dormant tickets corresponding to different standard traffic penalty information under each ticket influencing factor.
[0199] The execution process of the above (1) to (17) can be specifically referred to the description of the above embodiment. The implementation principles and technical effects are similar and will not be repeated here.
[0200] It should be understood that, although the steps in the flowcharts involved in the above embodiments are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps is not strictly limited in order, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above embodiments may include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily carried out in sequence, but can be executed in turn or alternately with other steps or at least a part of the steps or stages in other steps.
[0201] Based on the same inventive concept, the embodiment of the present application also provides a ticket supervision device for implementing the ticket supervision method involved above. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the above method, so the specific limitations in one or more ticket supervision device embodiments provided below can refer to the limitations of the ticket supervision method above, and will not be repeated here.
[0202] In one embodiment, Fig.12 This is a schematic diagram of the structure of a ticket supervision device in one embodiment of the present application. The ticket supervision device provided in the embodiment of the present application can be applied to a computer device. Fig.12 As shown, the ticket supervision device of the embodiment of the present application may include: an information acquisition module 11, an analysis module 12 and a supervision module 13, wherein:
[0203] The information acquisition module 11 is used to obtain the standard traffic penalty information of each driving user's sleeping ticket;
[0204] The analysis module 12 is used to obtain the impact factor measurement index corresponding to each standard traffic penalty information when each standard traffic penalty information meets the preset reliability evaluation conditions; the impact factor measurement index represents the importance of each standard traffic penalty information when described by multiple different ticket impact factors;
[0205] The supervision module 13 is used to supervise at least one dormant ticket corresponding to each ticket impact factor according to each standard traffic penalty information and each impact factor measurement index.
[0206] The ticket supervision device provided in the embodiment of the present application can be used to execute the technical solution in the above-mentioned ticket supervision method embodiment of the present application. Its implementation principle and technical effect are similar and will not be repeated here.
[0207] In one embodiment, the analysis module 12 includes: an analysis unit, a screening unit and an acquisition unit, wherein:
[0208] An analysis unit, used to perform principal component analysis based on each standard traffic penalty information to determine an initial ticket influencing factor corresponding to each standard traffic penalty information;
[0209] A screening unit, used for screening a plurality of ticket influencing factors from each of the initial ticket influencing factors according to the factor loading matrix and the rotational interpretation information of each of the initial ticket influencing factors;
[0210] The acquisition unit is used to acquire the influence factor measurement index of each ticket influence factor according to each rotated interpretation information and each standard traffic penalty information.
[0211] The ticket supervision device provided in the embodiment of the present application can be used to execute the technical solution in the above-mentioned ticket supervision method embodiment of the present application. Its implementation principle and technical effect are similar and will not be repeated here.
[0212] In one embodiment, the screening unit includes: an acquisition subunit, a screening subunit, a first determination subunit and a second determination subunit, wherein:
[0213] An acquisition subunit is used to acquire multiple target ticket variables from multiple preset ticket variables according to each rotated factor loading matrix;
[0214] A screening subunit, for screening a plurality of candidate ticket influencing factors from each initial ticket influencing factor based on each rotated interpretation information;
[0215] The first determination subunit is used to determine the factor score of each target ticket variable under each candidate ticket influencing factor according to the factor loading matrix corresponding to each candidate ticket influencing factor and the corresponding rotated interpretation information;
[0216] The second determination subunit is used to determine each ticket impact factor from each candidate ticket impact factor based on the score of each factor.
[0217] The ticket supervision device provided in the embodiment of the present application can be used to execute the technical solution in the above-mentioned ticket supervision method embodiment of the present application. Its implementation principle and technical effect are similar and will not be repeated here.
[0218] In one embodiment, the screening subunit is specifically used for:
[0219] Perform the maximum variance rotation on each standard traffic penalty information to obtain the factor loading matrix of each initial ticket influencing factor under multiple preset ticket variables;
[0220] Perform factor rotation on the factor loading matrix corresponding to each initial ticket influencing factor to obtain the rotated factor loading matrix of each initial ticket influencing factor;
[0221] Correspondingly, according to each rotated factor loading matrix, multiple target ticket variables are obtained from multiple preset ticket variables, including:
[0222] According to the rotated factor loading matrix of each initial ticket influencing factor, the common degree of each ticket variable explained by the common factor is determined; the common factor represents the factors that exist in common in each standard traffic penalty information;
[0223] The ticket variables corresponding to multiple commonalities greater than a preset threshold among the commonalities are determined as target ticket variables.
[0224] The ticket supervision device provided in the embodiment of the present application can be used to execute the technical solution in the above-mentioned ticket supervision method embodiment of the present application. Its implementation principle and technical effect are similar and will not be repeated here.
[0225] In one embodiment, the first determining subunit is specifically configured to:
[0226] According to the factor loading matrix corresponding to each candidate ticket influencing factor and the corresponding rotational interpretation information, determine the component score coefficient matrix of each candidate ticket influencing factor under each target ticket variable;
[0227] According to the component score coefficient matrix, the factor score of each target ticket variable under each candidate ticket influencing factor is obtained.
[0228] The ticket supervision device provided in the embodiment of the present application can be used to execute the technical solution in the above-mentioned ticket supervision method embodiment of the present application. Its implementation principle and technical effect are similar and will not be repeated here.
[0229] In one embodiment, each impact factor measurement indicator includes the contribution of each standard traffic penalty information to the construction of each ticket impact factor and / or the correlation described by each ticket impact factor; the acquisition unit includes: a processing subunit, a third determination subunit and a fourth determination subunit, wherein:
[0230] A processing subunit is used to perform linear combination processing on each standard traffic penalty information to obtain the contribution degree of each ticket influencing factor; and / or,
[0231] A third determination subunit is used to determine the comprehensive score coefficient of any ticket influencing factor according to the contribution degree of the ticket influencing factor and the corresponding rotated interpretation information;
[0232] The fourth determination subunit is used to determine the correlation of the ticket influencing factors according to the comprehensive score coefficients of the ticket influencing factors.
[0233] The ticket supervision device provided in the embodiment of the present application can be used to execute the technical solution in the above-mentioned ticket supervision method embodiment of the present application. Its implementation principle and technical effect are similar and will not be repeated here.
[0234] In one embodiment, the impact factor measurement index includes contribution degree and / or correlation degree; the supervision module 13 includes: a clustering processing unit and a classification supervision unit, wherein:
[0235] A cluster processing unit, used for inputting each standard traffic penalty information, each contribution degree and / or each correlation degree into the dormant ticket cluster analysis model, and determining at least one standard traffic penalty information under each ticket influencing factor;
[0236] The classification supervision unit is used to classify and supervise dormant tickets corresponding to different standard traffic penalty information under each ticket influencing factor.
[0237] The ticket supervision device provided in the embodiment of the present application can be used to execute the technical solution in the above-mentioned ticket supervision method embodiment of the present application. Its implementation principle and technical effect are similar and will not be repeated here.
[0238] In one embodiment, the ticket supervision device includes: a verification module and a determination module, wherein:
[0239] The test module is used to perform a reliability test of factor analysis on each standard traffic penalty information and obtain the test results;
[0240] The determination module determines whether each standard traffic penalty information meets the reliability evaluation conditions when the inspection result is that the reliability passes.
[0241] The ticket supervision device provided in the embodiment of the present application can be used to execute the technical solution in the above-mentioned ticket supervision method embodiment of the present application. Its implementation principle and technical effect are similar and will not be repeated here.
[0242] For the specific definition of the ticket supervision device, please refer to the definition of the ticket supervision method above, which will not be repeated here. Each module in the above-mentioned ticket supervision device can be implemented in whole or in part by software, hardware and a combination thereof. The above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.
[0243] In one embodiment, a computer device is provided. The computer device may be a server, and its internal structure diagram may be as follows: Fig.13 As shown. The computer device includes a processor, a memory and a network interface connected through a system bus. Among them, the processor of the computer device is used to provide processing power. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store the original traffic penalty information of the sleeping tickets of each driving user. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a ticket supervision method is implemented.
[0244] Those skilled in the art will understand that Fig.13The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.
[0245] In one embodiment, a computer device is also provided, including a memory and a processor, wherein a computer program is stored in the memory, and when the processor executes the computer program, the technical solution in the above-mentioned ticket supervision method embodiment of the present application is implemented. The implementation principle and technical effect are similar and will not be repeated here.
[0246] In one embodiment, a computer-readable storage medium is also provided, on which a computer program is stored. When the computer program is executed by a processor, the technical solution of the above-mentioned ticket supervision method of the present application is implemented. The implementation principle and technical effect are similar and will not be repeated here.
[0247] In one embodiment, a computer program product is also provided, including a computer program. When the computer program is executed by a processor, the technical solution of the above-mentioned ticket supervision method of the present application is implemented. The implementation principle and technical effect are similar and will not be repeated here.
[0248] Those of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory or optical memory, etc. Volatile memory can include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM).
[0249] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.
[0250] The above-mentioned embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the patent of the present application shall be subject to the attached claims.
Claims
1. A ticket supervision method, characterized in that: The method comprises: Get the standard traffic penalty information of each driver's sleeping ticket; When each of the standard traffic penalty information meets the preset reliability evaluation conditions, an impact factor measurement index corresponding to each of the standard traffic penalty information is obtained; the impact factor measurement index represents the importance of each of the standard traffic penalty information when described by multiple different ticket impact factors; According to each of the standard traffic penalty information and each of the impact factor measurement indicators, at least one dormant ticket corresponding to each ticket impact factor is supervised.
2. The method according to claim 1, characterized in that: The obtaining of the impact factor measurement index corresponding to each of the standard traffic penalty information includes: Performing principal component analysis based on each of the standard traffic penalty information to determine the initial ticket influencing factors corresponding to each of the standard traffic penalty information; According to the factor loading matrix and the rotated interpretation information of each of the initial ticket influencing factors, a plurality of ticket influencing factors are screened from each of the initial ticket influencing factors; According to each of the rotated interpretation information and each of the standard traffic penalty information, an influence factor measurement index of each of the ticket influence factors is obtained.
3. The method according to claim 2, characterized in that According to the factor loading matrix and the rotated interpretation information of each of the initial ticket influencing factors, a plurality of ticket influencing factors are screened from each of the initial ticket influencing factors, including: According to each of the rotated factor loading matrices, obtaining a plurality of target ticket variables from a plurality of preset ticket variables; Based on each of the rotated interpretation information, screening a plurality of candidate ticket impact factors from each of the initial ticket impact factors; Determine the factor score of each target ticket variable under each candidate ticket influencing factor according to the factor loading matrix corresponding to each candidate ticket influencing factor and the corresponding rotated interpretation information; Based on the scores of the factors, each of the ticket impact factors is determined from among the candidate ticket impact factors.
4. The method according to claim 3, characterized in that Before obtaining a plurality of target ticket variables from a plurality of preset ticket variables according to each of the rotated factor loading matrices, the method further includes: Performing a maximum variance rotation on each of the standard traffic penalty information to obtain a factor loading matrix of each of the initial ticket influencing factors under multiple preset ticket variables; Performing factor rotation on the factor loading matrix corresponding to each of the initial ticket influencing factors to obtain a rotated factor loading matrix of each of the initial ticket influencing factors; Correspondingly, the step of obtaining a plurality of target ticket variables from a plurality of preset ticket variables according to each of the rotated factor loading matrices includes: Determine the commonality of each ticket variable explained by a common factor according to the rotated factor loading matrix of each of the initial ticket influencing factors; the common factor represents the factor that exists in common in each of the standard traffic penalty information; The ticket variables corresponding to multiple commonalities greater than a preset threshold among the commonalities are determined as the target ticket variables.
5. The method according to claim 3, characterized in that: Determining the factor score of each target ticket variable under each candidate ticket influencing factor according to the factor loading matrix corresponding to each candidate ticket influencing factor and the corresponding rotated interpretation information includes: Determine the component score coefficient matrix of each candidate ticket influencing factor under each target ticket variable according to the rotated factor loading matrix corresponding to each candidate ticket influencing factor; According to the component score coefficient matrix, the factor score of each target ticket variable under each candidate ticket influencing factor is obtained.
6. The method according to any one of claims 2 to 5, characterized in that: Each of the impact factor measurement indicators includes the contribution of each of the standard traffic penalty information to the construction of each of the ticket impact factors and / or the correlation described by each of the ticket impact factors; the impact factor measurement indicator of each of the ticket impact factors is obtained according to each of the rotated interpretation information and each of the standard traffic penalty information, including: Performing linear combination processing on each of the standard traffic penalty information to obtain the contribution degree of each of the ticket influencing factors; and / or, For any ticket influencing factor, determine the comprehensive score coefficient of the ticket influencing factor according to the contribution degree of the ticket influencing factor and the corresponding rotated interpretation information; The correlation degree of the ticket influencing factors is determined according to the comprehensive score coefficients of the ticket influencing factors.
7. The method according to any one of claims 1 to 5, characterized in that The impact factor measurement index includes a contribution degree and / or a correlation degree; the monitoring of at least one dormant ticket corresponding to each ticket impact factor according to each of the standard traffic penalty information and each of the impact factor measurement indexes includes: Inputting each of the standard traffic penalty information, each of the contribution degrees and / or each of the correlation degrees into a dormant ticket cluster analysis model to determine at least one standard traffic penalty information under each ticket influencing factor; The dormant tickets corresponding to the different standard traffic penalty information under each ticket influencing factor are classified and supervised.
8. The method according to any one of claims 1 to 5, characterized in that The method further comprises: Conducting a reliability test of factor analysis on each of the standard traffic penalty information to obtain a test result; If the inspection result is that the reliability is passed, it is determined that each of the standard traffic penalty information meets the reliability evaluation condition.
9. A ticket supervision device, characterized in that: The device comprises: An information acquisition module, used to obtain standard traffic penalty information of each driving user's sleeping ticket; An analysis module, for obtaining an impact factor measurement index corresponding to each of the standard traffic penalty information when each of the standard traffic penalty information meets a preset reliability evaluation condition; the impact factor measurement index represents the importance of each of the standard traffic penalty information when described by a plurality of different ticket impact factors; The supervision module is used to supervise at least one dormant ticket corresponding to each ticket impact factor according to each of the standard traffic penalty information and each of the impact factor measurement indicators.
10. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 8 are implemented.
