A method for synchronous updating of data governance and highway standards
Through a data governance method combining real-time data analysis and user feedback, the highway standard update process is optimized, and the problems of lagging update process and insufficient data utilization in the existing technology are solved, and flexible and efficient updates of highway standards are achieved.
Patent Information
- Application Number
- CN202410894462.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-04
- Publication Date
- 2024-12-06
- Estimated Expiration
- 2044-07-04
AI Technical Summary
The existing highway standard update process lacks flexibility and real-timeness, cannot adapt to technological development and user needs in a timely manner, and does not fully utilize on-site data, resulting in a lack of data support in the decision-making process.
By collecting and analyzing road usage and condition data in real time, combining user feedback and expert evaluation, generating highway standard update recommendations, and optimizing and adjusting through data governance methods, ensuring transparency and responsiveness of the update process.
Real-time and reliability of highway standard updates are achieved, ensuring the effectiveness and adaptability of update suggestions, and making full use of data to support the decision-making process.
Smart Images

Figure CN118733601B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of digitalization of highway transportation infrastructure, and in particular to a method for synchronously updating data management and highway standards. Background Art
[0002] With the development of digitalization and automation technology, highway infrastructure management faces more and more challenges and opportunities. In the traditional model, the update of highway standards often lags behind the development of technology and changes in market demand. The highway standard update process is usually manual, periodic, static, and slow to respond to market and technological changes. It lacks the necessary flexibility and real-time performance. In addition, the existing update process usually does not make full use of the large amount of data collected on site, resulting in a lack of data support in the decision-making process. In addition, due to the lack of a flexible and adaptive update mechanism, it is unable to fully reflect the changes in actual operation and use. In other words, the existing standards may not be able to adapt to new technological developments, environmental changes and user needs in a timely manner.
[0003] Therefore, the present invention proposes a method for synchronous updating of data governance and highway standards. Summary of the invention
[0004] The present invention provides a method for synchronous updating of data governance and highway standards, which is used to preliminarily determine update suggestions by collecting and analyzing highway usage and status data in real time, determining highway trends and problems, usage and maintenance needs, and enhancing the transparency and responsiveness of the highway standards update process through a combination of user feedback and expert evaluation, thereby ensuring the reliability of the suggestions.
[0005] The present invention provides a method for synchronously updating data management and highway standards, comprising:
[0006] Step 1: Collect static data of highway infrastructure and pre-process the static data;
[0007] Step 2: Analyze the preprocessed data to extract useful information and identify trends and problems of the highway. At the same time, use statistical analysis and regression analysis methods to analyze historical data to evaluate the usage and maintenance needs of the highway.
[0008] Step 3: generating an initial standard update recommendation for the highway based on the trends and issues, usage and maintenance needs;
[0009] Step 4: Collect feedback information on the highway from different users based on the online feedback platform, optimize and adjust the initial standard update suggestion, obtain expert evaluation of the optimized and adjusted update suggestion, and select the best standard update suggestion;
[0010] Step 5: Associating the best standard update suggestion with the feedback information and storing the associated best standard update suggestion into the standard knowledge base for base update.
[0011] Preferably, preprocessing the static data includes:
[0012] Extracting all data identifiers based on the static data, and inputting all data identifiers into an identifier blank table in sequence, to obtain identifier rows and identifier columns of the static data;
[0013] Calculate a first distribution probability based on each identification row according to the total number of all identification rows and the first overlap degree and the first row fullness of each identification row;
[0014] Calculate the second distribution probability based on each identification column according to the total number of columns of all identification columns and the second overlap degree and second row satisfaction of each identification column;
[0015] Based on the overlapping result of the identification row and the identification column, and in combination with the first distribution probability and the second distribution probability, an existence label is set for each identification, and preprocessed data is obtained.
[0016] Preferably, an existence label is set for each identifier, and preprocessed data is obtained, including:
[0017] Obtain the appearance time of each sign from the highway database and construct a time-position array of the corresponding sign;
[0018] Obtain all adjacent time differences based on the same position in the time-position array, and obtain the minimum time difference and the time variance from all adjacent time differences under the same position;
[0019] When the minimum time difference Variance with time The sum is not less than the preset difference threshold When, according to Setting a second coefficient to a corresponding identifier;
[0020] Based on all setting coefficients of the same identifier, the overlap coefficient assigned to the same identifier based on the overlap result, the first distribution probability and the second distribution probability, a final coefficient of the corresponding identifier is obtained;
[0021]
[0022] in, represents the final coefficient of the corresponding identification; Indicates the overlap coefficient of the corresponding logo, the value is 1 or 2; Indicates the first distribution probability of the identification row where the corresponding identification is located; represents the second distribution probability of the corresponding identifier column; ln represents the sign of the logarithmic function; Indicates that the corresponding identification is based on the average value of all first coefficients; Indicates that the corresponding identification is based on the average value of all second coefficients; Indicates the total number of the first coefficients under the corresponding mark; Indicates the total number of second coefficients under the corresponding mark; represents the variance of the corresponding identity based on all first coefficients; represents the weight set to the first coefficient;
[0023] A blank introduction tag consistent with the highway infrastructure management type matching the corresponding identification is set, and the corresponding final coefficient is attached to the blank introduction tag to obtain the existence tag of the corresponding identification.
[0024] Preferably, identifying trends and problems on the highway comprises:
[0025] The useful information is split according to the set identification index to obtain index information for each set identification index;
[0026] Determine the mutual exclusion coefficient of each indicator information and each remaining information respectively, obtain the unity of the corresponding indicator information, and set the information weight to the corresponding indicator information;
[0027] According to the indicator weight and information weight of each indicator information, key entities are matched from the double weight-indicator comparison table to perform key extraction on the corresponding indicator information to obtain a single anomaly;
[0028] All single anomalies are combined according to the construction deployment structure of the highway to obtain the trends and problems of the highway.
[0029] Preferably, generating an initial standard update proposal for the highway based on the trends and issues, usage and maintenance needs includes:
[0030] Standardize the trends and issues, usage and maintenance needs to obtain the input data set ,in, represents the input data at time t, and n1 represents the total number of time; represents the sub-data of the i-th input type in the input data at the t-th time; n2 represents the total number of input types involved in the input data at the t-th time;
[0031] Inputting the input data into the decision model to generate an initial comprehensive update value at a corresponding moment and initial individual update values of different input types in the input data;
[0032] An initial standard update suggestion for the highway is generated according to the initial comprehensive update value and the initial individual update value.
[0033] Preferably, collecting feedback information on the highway from different users based on an online feedback platform and optimizing and adjusting the initial standard update suggestion include:
[0034] The feedback source of each user's feedback information collected by the online feedback platform is locked, and at the same time, the source noise path of each feedback source is determined based on the source-noise mapping relationship table;
[0035] According to the feedback proportion of each feedback source in a preset time period, a first cleaning rule is set for the corresponding source noise path;
[0036] Extracting the cleaning logic in the first cleaning rule and the feedback logic of each feedback information under the corresponding feedback source, and performing distance judgment on the cleaning logic and feedback logic under the same feedback source;
[0037] According to the distance judgment result, each feedback logic under the same feedback source is divided into a range, and a cleaning label set is set for the corresponding feedback source based on the number of logics under the same divided range and the average distance value, wherein the cleaning label set includes: the cleaning time and cleaning supervision sensitivity of each divided range under the corresponding feedback source;
[0038] Determine the logic type of the feedback logic under all feedback sources, and classify all cleaning label sets according to the logic type to obtain type cleaning sets of the same logic type;
[0039] Input the cleaning labels in the type cleaning set of the same logical type into the blank arrangement table in sequence, and obtain the redundancy level based on each cleaning label;
[0040] Traverse all redundancy levels under the same logic type and obtain the maximum level and the minimum level;
[0041] Based on the cleaning time and cleaning supervision sensitivity of the cleaning label, the cleaning value of each cleaning label under the same logic type is calculated;
[0042] Determine a corresponding cleaning variance according to all cleaning values under the same logic type, and optimize the cleaning logic based on the cleaning variance and the logic type to obtain a second cleaning rule;
[0043] The feedback information is cleaned based on the second cleaning rule to obtain optimized supplementation, and the initial standard update suggestion is optimized and adjusted.
[0044] Preferably, calculating the cleaning value of each cleaning tag under the same logic type includes:
[0045]
[0046] in, Indicates the cleaning value of the corresponding cleaning tag; Respectively represent the setting weights; Indicates the cleaning time of the corresponding cleaning label; Indicates the average cleaning time of all cleaning tags under the same logical type; Indicates the cleaning supervision sensitivity to which the corresponding cleaning label belongs; It represents the average value of the cleaning supervision sensitivity of all cleaning labels under the same logic type; Indicates the redundancy level of the corresponding cleaning tag; Indicates the maximum level of the same logical type; Indicates the minimum level under the same logical type; Indicates the adjustment amount; Represents the variance of all redundancy levels under the same logical type; Represents the average value of all redundancy levels under the same logic type; Indicates the total number of redundancy levels under the same logic type.
[0047] Preferably, associating the optimal standard update suggestion with the feedback information includes:
[0048] Extracting some update information from the best standard update proposal;
[0049] The similarity coefficients of each update information and each feedback information are determined respectively according to the similarity function, and an association is established between the feedback information with a similarity coefficient greater than a preset coefficient and the corresponding update information.
[0050] Compared with the prior art, the beneficial effects of this application are as follows:
[0051] By collecting and analyzing highway usage and condition data in real time, we can identify highway trends and problems, usage and maintenance needs to preliminarily determine update recommendations. By combining user feedback with expert evaluation, we can enhance the transparency and responsiveness of the highway standards update process and ensure the reliability of the recommendations.
[0052] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or understood by practicing the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures particularly pointed out in the written description and the accompanying drawings.
[0053] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS
[0054] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:
[0055] Figure 1 This is a flow chart of a method for synchronously updating data management and highway standards in an embodiment of the present invention;
[0056] Figure 2 It is a structural diagram of the data management and highway standard synchronous updating device in an embodiment of the present invention. DETAILED DESCRIPTION
[0057] The preferred embodiments of the present invention are described below in conjunction with the accompanying drawings. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention, and are not used to limit the present invention.
[0058] The present invention provides a method for synchronously updating data management and highway standards, such as Figure 1 As shown, including:
[0059] Step 1: Collect static data of highway infrastructure and pre-process the static data;
[0060] Step 2: Analyze the preprocessed data to extract useful information and identify trends and problems of the highway. At the same time, use statistical analysis and regression analysis methods to analyze historical data to evaluate the usage and maintenance needs of the highway.
[0061] Step 3: generating an initial standard update recommendation for the highway based on the trends and issues, usage and maintenance needs;
[0062] Step 4: Collect feedback information on the highway from different users based on the online feedback platform, optimize and adjust the initial standard update suggestion, obtain expert evaluation of the optimized and adjusted update suggestion, and select the best standard update suggestion;
[0063] Step 5: Associating the best standard update suggestion with the feedback information and storing the associated best standard update suggestion into the standard knowledge base for base update.
[0064] In this embodiment, if Figure 2 As shown, step 1 is implemented based on the data collection and preprocessing module, step 2 is implemented based on the data analysis and processing module, step 3 is implemented based on the standard update suggestion generation module, step 4 is implemented based on the user feedback mechanism, and step 5 is implemented based on the standard knowledge base update module.
[0065] In this embodiment, the static data includes: traffic flow statistics, meteorological data, road condition survey reports, accident records, etc.
[0066] In this embodiment, preprocessing refers to the identification analysis and processing of static data, wherein various types of data involved in the static data have their corresponding set identifications, which are set in advance. Which identification represents which data meaning, etc. For example, Sg%1 indicates that a vehicle collision occurred in lane 1 of section A.
[0067] In this embodiment, the extracted useful information is obtained by removing irrelevant data from the pre-processed data, and the result obtained is the useful information.
[0068] In this embodiment, trends and problems refer to abnormal conditions existing on the corresponding highway, such as road maintenance, severe road wear, etc.
[0069] In this embodiment, the process of using the statistical analysis and regression analysis method includes:
[0070]
[0071] in, Represent two variables of time series data, They represent the mean respectively; ,in, is the predicted value, is the regression coefficient, is the independent variable.
[0072] In this embodiment, historical data refers to the collected data of the highway at historical time, which is consistent with the data type involved in static data.
[0073] In this embodiment, the initial standard update suggestions include but are not limited to:
[0074] recommended highway maintenance standards based on historical weather data;
[0075] Recommended traffic signal cycles and lane usage standards based on traffic flow statistics;
[0076] Survey report on pavement wear and tear and recommendations on maintenance frequency adjustments;
[0077] Use rule engines, such as decision tree models, to automatically generate update recommendations based on predefined rules and conditions.
[0078] In this embodiment, the feedback information includes feedback on highway conditions and standards submitted by highway users and stakeholders.
[0079] In this embodiment, a multi-criteria decision analysis method (such as AHP and TOPSIS) is used to comprehensively consider different factors and select the best standard update suggestion.
[0080] In this embodiment, the knowledge base is maintained and updated regularly to ensure the accuracy and completeness of its data.
[0081] In this embodiment, the evaluation process is actually a further optimization of the recommendations.
[0082] In this embodiment, the purpose of association is to determine the association relationship between the update suggestion and the feedback.
[0083] In this embodiment, the online feedback platform is pre-deployed, and the platform is connected to multiple feedback channels.
[0084] The beneficial effects of the above technical solution are: by collecting and analyzing highway usage and condition data in real time, determining highway trends and problems, usage and maintenance needs to preliminarily determine update recommendations, and by combining user feedback with expert evaluation to enhance the transparency and responsiveness of the highway standards update process and ensure the reliability of the recommendations.
[0085] The present invention provides a method for synchronously updating data management and highway standards, which pre-processes the static data, including:
[0086] Extracting all data identifiers based on the static data, and inputting all data identifiers into an identifier blank table in sequence, to obtain identifier rows and identifier columns of the static data;
[0087] Calculate a first distribution probability based on each identification row according to the total number of all identification rows and the first overlap degree and the first row fullness of each identification row;
[0088] Calculate the second distribution probability based on each identification column according to the total number of columns of all identification columns and the second overlap degree and second row satisfaction of each identification column;
[0089] Based on the overlapping result of the identification row and the identification column, and in combination with the first distribution probability and the second distribution probability, an existence label is set for each identification, and preprocessed data is obtained.
[0090] In this embodiment, the data identifier refers to a unique identifier representative for different data.
[0091] In this embodiment, the blank table of identifiers contains blank cells corresponding to different identifiers, which facilitates input and filling of identifiers involved in static data to obtain a filled table.
[0092] In this embodiment, each row of the filled table is regarded as an identification row, and each column is regarded as an identification column. Because the same data identification may exist several times in the static data, there will be multiple identical identifications in the same blank space, and the overlap degree can be obtained. There are multiple different identifications in the same identification row, and the fullness can be obtained.
[0093] In this embodiment, the first overlap degree=the overlap quantity of each blank space in the same mark row / the total quantity of blank space in the corresponding mark row.
[0094] The fullness of the first row = the number of non-empty blank spaces / the total number of blank spaces involved in the corresponding identification row.
[0095] .
[0096] Among them, the calculation method of the second overlap degree, the second row fullness, and the second distribution probability is similar to the calculation process of the first overlap degree, the first row fullness, and the first distribution probability, and will not be repeated here.
[0097] In this embodiment, the existence label is obtained by assigning a value to the corresponding label after the corresponding value is calculated.
[0098] In this embodiment, useful data is retained by setting an existence tag, and at this time, pre-processed data can be obtained.
[0099] The beneficial effect of the above technical solution is: by extracting data identifiers and performing overlapping, filling and distribution analysis on rows and columns of blank tables, existence labels are set to facilitate the effective acquisition of preprocessed data, provide a data basis for subsequent analysis, and ensure the rationality of highway standard updates.
[0100] The present invention provides a method for synchronously updating data management and highway standards, setting an existence label for each identifier and obtaining pre-processed data, including:
[0101] Obtain the appearance time of each sign from the highway database and construct a time-position array of the corresponding sign;
[0102] Obtain all adjacent time differences based on the same position in the time-position array, and obtain the minimum time difference and the time variance from all adjacent time differences under the same position;
[0103] When the minimum time difference Variance with time The sum is less than the preset difference threshold When, according to Setting a first coefficient to a corresponding identifier;
[0104] When the minimum time difference Variance with time The sum is not less than the preset difference threshold When, according to Setting a second coefficient to a corresponding identifier;
[0105] Based on all setting coefficients of the same identifier, the overlap coefficient assigned to the same identifier based on the overlap result, the first distribution probability and the second distribution probability, a final coefficient of the corresponding identifier is obtained;
[0106]
[0107] in, represents the final coefficient of the corresponding identification; Indicates the overlap coefficient of the corresponding logo, the value is 1 or 2; Indicates the first distribution probability of the identification row where the corresponding identification is located; represents the second distribution probability of the corresponding identifier column; ln represents the sign of the logarithmic function; Indicates that the corresponding identification is based on the average value of all first coefficients; Indicates that the corresponding identification is based on the average value of all second coefficients; Indicates the total number of the first coefficients under the corresponding mark; Indicates the total number of second coefficients under the corresponding mark; represents the variance of the corresponding identity based on all first coefficients; represents the weight set to the first coefficient;
[0108] A blank introduction tag consistent with the highway infrastructure management type matching the corresponding identification is set, and the corresponding final coefficient is attached to the blank introduction tag to obtain the existence tag of the corresponding identification.
[0109] In this embodiment, the highway database includes the setting time and identification representation of different identifications, wherein the setting time is the appearance time, because the same identification may appear at different locations and at different times, and each identification may be updated in time at a certain location, so, for example, the time-position array for identification 1 is: w1--t1 / t3 / t4, w2--t2, wherein t1, t2, t3, t4 represent the appearance time, and w1 and w2 represent the appearance location.
[0110] In this embodiment, the adjacent time differences for w1 are: t3-t1, t4-t3.
[0111] Minimum time difference: min(t3-t1,t4-t3).
[0112] In this embodiment, the difference threshold is preset It is pre-set.
[0113] In this embodiment, the first coefficient is: ;
[0114] The second coefficient is: .
[0115] In this embodiment, for example, the overlapping identifier in the first row of the identifier row and the first column of the identifier column corresponds to the coordinate (1, 1). At this time, the overlapping coefficient of the identifier at this coordinate is 2, otherwise, the overlapping coefficient is 1.
[0116] In this embodiment, the blank introduction label corresponding to each highway infrastructure management type is different, mainly for type distinction, and the final coefficient is added to assign a value to the label to ensure the substantiality of the label.
[0117] The beneficial effect of the above technical solution is: by setting a time-position array to set different coefficients for the identification, and combining the distribution probability and the overlap coefficient to calculate the final coefficient, it facilitates the setting of the existence label and effectively determines the trends and problems of the highway.
[0118] The present invention provides a method for synchronously updating data governance and highway standards, identifying trends and problems of the highway, including:
[0119] The useful information is split according to the set identification index to obtain index information for each set identification index;
[0120] Determine the mutual exclusion coefficient of each indicator information and each remaining information respectively, obtain the unity of the corresponding indicator information, and set the information weight to the corresponding indicator information;
[0121] According to the indicator weight and information weight of each indicator information, key entities are matched from the double weight-indicator comparison table to perform key extraction on the corresponding indicator information to obtain a single anomaly;
[0122] All single anomalies are combined according to the construction deployment structure of the highway to obtain the trends and problems of the highway.
[0123] In this embodiment, the identification index is set in advance, and then information is extracted according to the index. Each index corresponds to an index information, which is equivalent to splitting the useful information. For example, there are index 1 and index 2. After splitting, index 1 corresponds to index information a1, and index 2 corresponds to index information a2, and index information a1 and index information a2 constitute useful information.
[0124] In this embodiment, the mutual exclusion coefficient=1-sim (corresponding indicator information, corresponding remaining information).
[0125] In this embodiment, singularity=mutually exclusive coefficient / preset coefficient, and the value of the preset coefficient is generally 1+the variance of all mutually exclusive coefficients.
[0126] In this embodiment, information weight=square root of the product of singularity and index weight.
[0127] In this embodiment, the dual weight-indicator comparison table contains a combination of different indicator weights, information weights, and key entities matching the combination, and the key entity refers to a key identifier to extract information, obtain key information, and then determine the anomaly of the key information, that is, a single anomaly.
[0128] In this embodiment, for example, the key entity is ^1. At this time, information related to ^1 is extracted from the indicator information to determine the anomaly in the extracted information based on the indicator standard. For example, the current road wear information is r1, and the standard is that it needs to be less than r0. At this time, r1 is greater than r0, and the single anomaly is: road wear anomaly.
[0129] In this embodiment, the construction deployment structure refers to the road construction structure of the highway.
[0130] In this embodiment, the purpose of the combination is to sequentially combine all single anomalies, and then obtain trends and problems based on the combination-problem database matching, and the database contains different anomaly combinations and the trends and problems that match the combinations, which are pre-planned.
[0131] The beneficial effect of the above technical solution is: by splitting the useful information according to the set identification indicators and calculating the mutually exclusive coefficients, and combining the double weights, a single anomaly is effectively obtained, which provides a basis for the subsequent acquisition of trends and problems and ensures the initial rationality of the recommended acquisition.
[0132] The present invention provides a method for synchronously updating data governance and highway standards, which generates an initial standard update suggestion for the highway based on the trends and problems, usage and maintenance requirements, including:
[0133] Standardize the trends and issues, usage and maintenance needs to obtain the input data set in, represents the input data at time t, and n1 represents the total number of time; represents the sub-data of the i-th input type in the input data at the t-th time; n2 represents the total number of input types involved in the input data at the t-th time;
[0134] Inputting the input data into the decision model to generate an initial comprehensive update value at a corresponding moment and initial individual update values of different input types in the input data;
[0135] An initial standard update suggestion for the highway is generated according to the initial comprehensive update value and the initial individual update value.
[0136] In this embodiment, the decision model is pre-trained based on various single inputs and outputs, and various comprehensive inputs and outputs are obtained by training the neural network model for samples, so that the initial comprehensive update value and the initial individual update value can be directly obtained.
[0137] In this embodiment, the initial standard update suggestion is obtained based on a value analysis model, which is obtained by training a neural network model based on samples of different values, different types, and expert analysis suggestions on value and type combinations.
[0138] The beneficial effects of the above technical solution are: ensuring the uniformity of data through standardization, and obtaining suggestions based on decision-making models and value analysis models, providing a basis for subsequent adjustments and evaluations.
[0139] The present invention provides a method for synchronously updating data management and highway standards, which collects feedback information of different users on the highway based on an online feedback platform, and optimizes and adjusts the initial standard update suggestion, including:
[0140] The feedback source of each user's feedback information collected by the online feedback platform is locked, and at the same time, the source noise path of each feedback source is determined based on the source-noise mapping relationship table;
[0141] According to the feedback proportion of each feedback source in a preset time period, a first cleaning rule is set for the corresponding source noise path;
[0142] Extracting the cleaning logic in the first cleaning rule and the feedback logic of each feedback information under the corresponding feedback source, and performing distance judgment on the cleaning logic and feedback logic under the same feedback source;
[0143] According to the distance judgment result, each feedback logic under the same feedback source is divided into a range, and a cleaning label set is set for the corresponding feedback source based on the number of logics under the same divided range and the average distance value, wherein the cleaning label set includes: the cleaning time and cleaning supervision sensitivity of each divided range under the corresponding feedback source;
[0144] Determine the logic type of the feedback logic under all feedback sources, and classify all cleaning label sets according to the logic type to obtain type cleaning sets of the same logic type;
[0145] Input the cleaning labels in the type cleaning set of the same logical type into the blank arrangement table in sequence, and obtain the redundancy level based on each cleaning label;
[0146] Traverse all redundancy levels under the same logic type and obtain the maximum level and the minimum level;
[0147] Based on the cleaning time and cleaning supervision sensitivity of the cleaning label, the cleaning value of each cleaning label under the same logic type is calculated;
[0148] Determine a corresponding cleaning variance according to all cleaning values under the same logic type, and optimize the cleaning logic based on the cleaning variance and the logic type to obtain a second cleaning rule;
[0149] The feedback information is cleaned based on the second cleaning rule to obtain optimized supplementation, and the initial standard update suggestion is optimized and adjusted.
[0150] In this embodiment, the feedback source refers to the input address where the user inputs the feedback information. For example, user 1 implements feedback from website 1, and user 2 implements feedback from website 2, and website 1 and website 2 respectively establish communication connections with the online feedback platform, that is, information can be directly communicated to the online feedback platform through feedback based on these two websites.
[0151] In this embodiment, the source-noise mapping relationship table includes different feedback sources and source noise paths matching the sources, and the source noise path is mainly the communication path between the website and the online feedback platform, and noise may exist in the path.
[0152] In this embodiment, the feedback ratio=the feedback amount of the corresponding feedback source within the preset time period / the total feedback amount of all feedback sources within the preset time period.
[0153] In this embodiment, the first cleaning rule is obtained based on the matching of the proportion-source-rule comparison table, and the table includes different feedback sources, the feedback proportions matching the sources, and the cleaning rules pre-set by experts. The purpose of the cleaning rules is to clean missing values, supplement missing values, etc., to ensure the rationality of the data.
[0154] In this embodiment, each cleaning rule has its corresponding cleaning logic, which is pre-planned.
[0155] In this embodiment, the feedback logic is obtained by inputting feedback information into an information logic analysis model, and the information logic analysis model is obtained by training a neural network model based on different feedback information and experts' logical decomposition of the feedback information into samples.
[0156] In this embodiment, the distance judgment is calculated based on the Euclidean distance formula.
[0157] In this embodiment, for example, under feedback source 1, there is feedback logic 1 with a distance a1, feedback logic 2 with a distance a2, and feedback logic 3 with a distance a3. At this time, distance a1 and distance a2 can be divided together, and distance a3 can be divided together. At this time, the result of dividing distance a1 and distance a2 together is: the number of logics is 2, and so on.
[0158] In this embodiment, each division range corresponds to a cleaning time that is pre-set, and the cleaning supervision sensitivity is determined based on the logical number and the average distance value, and is obtained by matching based on the range-number-distance-sensitivity comparison table, and the comparison table includes the division range, the logical number based on the division range, the average distance value and the cleaning supervision sensitivity that matches it.
[0159] In this embodiment, the logic type is, for example, the indication logic for highway signs, the wear logic for road wear pavement, etc. For example, feedback source 1: feedback logic 01, feedback logic 02;
[0160] Feedback source 2: feedback logic 11, feedback logic 12;
[0161] At this time, feedback logic 01 and feedback logic 11 belong to the same logic type a1;
[0162] Feedback logic 02 and feedback logic 12 belong to logic type b1;
[0163] Therefore, after classification according to the logic type, all feedback logics involved in the same logic type can be obtained.
[0164] In this embodiment, the blank arrangement table includes a preset type header, mainly for effectively arranging the corresponding cleaning labels in the table, and the rows in the table are label results under the same logical type.
[0165] In this embodiment, the redundancy level=the occurrence frequency of the same cleaning label under the same logic type / the number of non-empty cells in the corresponding column.
[0166] In this embodiment, an optimization factor matching the cleaning variance and the logic type is obtained from the optimization database, and the cleaning logic is optimized based on the optimization factor to obtain the second cleaning rule.
[0167] For example, the first cleaning rule is to delete data similar to r01 in the data, and the optimization factor refers to expanding r01 to obtain r02. At this time, the second cleaning rule includes deleting data similar to r02 in the data.
[0168] In this embodiment, the purpose of optimization and supplementation is to optimize and adjust the initial suggestions, and the second cleaning rule obtains more effective data after data cleaning of the feedback information. By analyzing the problems existing in the more effective data, an optimization solution can be obtained to adjust the initial suggestions. For example, the road surface of section A was originally to be restored to primary wear, but the optimization supplement is that the vehicle passing rate of section A is much higher than that of normal sections, and secondary wear restoration is required.
[0169] The beneficial effect of the above technical solution is: by determining the source noise path and the feedback ratio to set the first cleaning rule, and further through the distance judgment analysis between the feedback logic and the cleaning logic, and the classification under the type logic, the redundancy level is effectively calculated, thereby achieving the optimization and supplement of the cleaning rules, ensuring the rationality and transparent integrity of the suggestions.
[0170] The present invention provides a method for synchronously updating data management and highway standards, which calculates the cleaning value of each cleaning tag under the same logical type, including:
[0171]
[0172] in, Indicates the cleaning value of the corresponding cleaning tag; Respectively represent the setting weights; Indicates the cleaning time of the corresponding cleaning label; Indicates the average cleaning time of all cleaning tags under the same logical type; Indicates the cleaning supervision sensitivity to which the corresponding cleaning label belongs; It represents the average value of the cleaning supervision sensitivity of all cleaning labels under the same logic type; Indicates the redundancy level of the corresponding cleaning tag; Indicates the maximum level of the same logical type; Indicates the minimum level under the same logical type; Indicates the adjustment amount; Represents the variance of all redundancy levels under the same logical type; Represents the average value of all redundancy levels under the same logic type; Indicates the total number of redundancy levels under the same logic type.
[0173] The beneficial effect of the above technical solution is: the method of size comparison is used to discuss the method of obtaining the adjustment amount in two cases, and further the cleaning value is calculated from three aspects through cleaning time, cleaning supervision sensitivity and redundancy level to ensure its accuracy and provide a basis for the follow-up.
[0174] The present invention provides a method for synchronously updating data management and highway standards, which associates the optimal standard update suggestion with feedback information, including:
[0175] Extracting some update information from the best standard update proposal;
[0176] The similarity coefficients of each update information and each feedback information are determined respectively according to the similarity function, and an association is established between the feedback information with a similarity coefficient greater than a preset coefficient and the corresponding update information.
[0177] In this embodiment, the best standard update suggestion is obtained by further evaluating the adjusted suggestion, which belongs to the prior art. For example, the initial suggestion is The adjusted recommendation is The recommendations after evaluation are .
[0178] In this embodiment, sim() is a similarity function, which belongs to the prior art.
[0179] In this embodiment, the preset coefficient is pre-set and has a value of 0.5.
[0180] The beneficial effect of the above technical solution is: by extracting the update information and combining the similarity function with the feedback information to perform similarity comparison analysis, the effective establishment of association is achieved.
[0181] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.
Claims
1. A method for synchronous updating of data management and highway standards, characterized in that: include: Step 1: Collect static data of highway infrastructure and pre-process the static data; Step 2: Analyze the preprocessed data to extract useful information and identify trends and problems of the highway. At the same time, use statistical analysis and regression analysis methods to analyze historical data to evaluate the usage and maintenance needs of the highway. Step 3: generating an initial standard update recommendation for the highway based on the trends and issues, usage and maintenance needs; Step 4: Collect feedback information on the highway from different users based on the online feedback platform, optimize and adjust the initial standard update suggestion, obtain expert evaluation of the optimized and adjusted update suggestion, and select the best standard update suggestion; Step 5: Associating the best standard update suggestion with the feedback information and storing the associated best standard update suggestion into the standard knowledge base for base update; The static data is preprocessed, including: Extracting all data identifiers based on the static data, and inputting all data identifiers into an identifier blank table in sequence, to obtain identifier rows and identifier columns of the static data; Calculate a first distribution probability based on each identification row according to the total number of all identification rows and the first overlap degree and the first row fullness of each identification row; Calculate the second distribution probability based on each identification column according to the total number of all identification columns and the second overlap degree and the second row fullness of each identification column; Based on the overlapping results of the identification row and the identification column, and in combination with the first distribution probability and the second distribution probability, an existence label is set for each identification, and preprocessed data is obtained; Among them, an existence label is set for each identifier, and preprocessed data is obtained, including: Obtain the appearance time of each sign from the highway database and construct a time-position array of the corresponding sign; Obtain all adjacent time differences based on the same position in the time-position array, and obtain the minimum time difference and the time variance from all adjacent time differences under the same position; When the minimum time difference t min With the time variance σ 2 When the sum of is less than the preset difference threshold Yt, Setting a first coefficient to a corresponding identifier; When the minimum time difference t min With the time variance σ 2 When the sum of is not less than the preset difference threshold Yt, Setting a second coefficient to a corresponding identifier; Based on all setting coefficients of the same identifier, the overlap coefficient assigned to the same identifier based on the overlap result, the first distribution probability and the second distribution probability, a final coefficient of the corresponding identifier is obtained; Among them, Zx represents the final coefficient of the corresponding mark; Cx represents the overlap coefficient of the corresponding mark, which takes a value of 1 or 2; f1 represents the first distribution probability of the mark row where the corresponding mark is located; f2 represents the second distribution probability of the mark column where the corresponding mark is located; ln represents the logarithmic function symbol; d1 ave Indicates that the corresponding identification is based on the average value of all first coefficients; d2 ave Indicates the average value of all second coefficients based on the corresponding identifier; m1 indicates the total number of first coefficients under the corresponding identifier; m2 indicates the total number of second coefficients under the corresponding identifier; represents the variance of the corresponding identifier based on all first coefficients; ε1 represents the weight set to the first coefficient; Setting a blank introduction tag consistent with the highway infrastructure management type matching the corresponding identification, and appending the corresponding final coefficient to the blank introduction tag to obtain an existence tag of the corresponding identification; Among them, irrelevant data is eliminated from the preprocessed data, and the result obtained is useful information; Wherein, the first overlap degree = the overlap number of each blank space in the same mark row / the total number of blank spaces in the corresponding mark row; The fullness of the first row = the number of non-empty blank spaces / the total number of blank spaces involved in the corresponding identification row.
2. The method for synchronously updating data management and highway standards according to claim 1 is characterized in that: Identify trends and issues on the highway, including: The useful information is split according to the set identification index to obtain index information for each set identification index; Determine the mutual exclusion coefficient of each indicator information and each remaining information respectively, obtain the unity of the corresponding indicator information, and set the information weight to the corresponding indicator information; According to the indicator weight and information weight of each indicator information, key entities are matched from the double weight-indicator comparison table to perform key extraction on the corresponding indicator information to obtain a single anomaly; All single anomalies are combined according to the construction deployment structure of the highway to obtain the trends and problems of the highway.
3. The method for synchronously updating data management and highway standards according to claim 1 is characterized in that: Based on the trends and issues, usage and maintenance needs, generate initial standards update recommendations for the highway, including: The trends and issues, usage and maintenance requirements are standardized to obtain the input data set S t ={x(t),t=1,2,3,...n1},x(t)={r ti ,i=1,2,3,...,n2}, where x(t) represents the input data at time t, n1 represents the total number of time; r ti represents the sub-data of the i-th input type in the input data at the t-th time; n2 represents the total number of input types involved in the input data at the t-th time; Inputting the input data into the decision model to generate an initial comprehensive update value at a corresponding moment and initial individual update values of different input types in the input data; An initial standard update suggestion for the highway is generated according to the initial comprehensive update value and the initial individual update value.
4. The method for synchronous updating of data management and highway standards according to claim 1 is characterized in that: Collect feedback information on the highway from different users based on the online feedback platform, and optimize and adjust the initial standard update suggestions, including: The feedback source of each user's feedback information collected by the online feedback platform is locked, and at the same time, the source noise path of each feedback source is determined based on the source-noise mapping relationship table; According to the feedback proportion of each feedback source in a preset time period, a first cleaning rule is set for the corresponding source noise path; Extracting the cleaning logic in the first cleaning rule and the feedback logic of each feedback information under the corresponding feedback source, and performing distance judgment on the cleaning logic and feedback logic under the same feedback source; According to the distance judgment result, each feedback logic under the same feedback source is divided into a range, and a cleaning label set is set for the corresponding feedback source based on the number of logics under the same divided range and the average distance value, wherein the cleaning label set includes: the cleaning time and cleaning supervision sensitivity of each divided range under the corresponding feedback source; Determine the logic type of the feedback logic under all feedback sources, and classify all cleaning label sets according to the logic type to obtain type cleaning sets of the same logic type; Input the cleaning labels in the type cleaning set of the same logical type into the blank arrangement table in sequence, and obtain the redundancy level based on each cleaning label; Traverse all redundancy levels under the same logic type and obtain the maximum level and the minimum level; Based on the cleaning time and cleaning supervision sensitivity of the cleaning label, the cleaning value of each cleaning label under the same logic type is calculated; Determine a corresponding cleaning variance according to all cleaning values under the same logic type, and optimize the cleaning logic based on the cleaning variance and the logic type to obtain a second cleaning rule; The feedback information is cleaned based on the second cleaning rule to obtain optimized supplementation, and the initial standard update suggestion is optimized and adjusted.
5. The method for synchronous updating of data management and highway standards according to claim 1 is characterized in that: Calculate the cleaning value of each cleaning tag under the same logical type, including: Wherein, Qz represents the cleaning value of the corresponding cleaning label; δ1, δ2, and δ3 represent the set weights respectively; Tq represents the cleaning time of the corresponding cleaning label; Tave represents the average cleaning time of all cleaning labels under the same logic type; Lq represents the cleaning supervision sensitivity of the corresponding cleaning label; Lave represents the average cleaning supervision sensitivity of all cleaning labels under the same logic type; Rq represents the redundancy level of the corresponding cleaning label; Rmax represents the maximum level under the same logic type; Rmin represents the minimum level under the same logic type; βq represents the adjustment amount; represents the variance of all redundancy levels under the same logic type; Rave represents the average value of all redundancy levels under the same logic type; M1 represents the total number of redundancy levels under the same logic type.
6. The method for synchronous updating of data management and highway standards according to claim 1, characterized in that: Correlate the best standards update recommendations with feedback information, including: Extracting some update information from the best standard update proposal; The similarity coefficients of each update information and each feedback information are determined respectively according to the similarity function, and an association is established between the feedback information with a similarity coefficient greater than a preset coefficient and the corresponding update information.
