CN117391644A

CN117391644A - Parameter adjustment method, device, equipment and storage medium

Info

Publication number: CN117391644A
Application number: CN202311705784.3A
Authority: CN
Inventors: 张瑞; 田园; 丁亚斐; 朱丽丽; 李彬; 宋婷; 李晨鹏; 雷有寿
Original assignee: State Grid Materials Co Ltd
Current assignee: State Grid Materials Co Ltd
Priority date: 2023-12-12
Filing date: 2023-12-12
Publication date: 2024-01-12
Anticipated expiration: 2043-12-12
Also published as: CN117391644B

Abstract

The application provides a parameter adjustment method, device, equipment and storage medium, which are applied to the technical field of data processing. Acquiring service data and each configuration characteristic value corresponding to the service data in a current environment, pre-analyzing each configuration characteristic value corresponding to the service data by utilizing a preset data processing rule to obtain a pre-analysis result, when each configuration characteristic value in the pre-analysis result is in a preset range, performing configuration characteristic value validity judgment on each configuration characteristic value corresponding to the service data by utilizing a configuration characteristic value validity judgment rule on the basis of the service data to obtain a configuration characteristic value validity judgment result, performing trial calculation on each configuration characteristic value by utilizing a preset trial calculation rule on the basis of the configuration characteristic value validity judgment result to obtain a trial calculation result, and generating a configuration characteristic adjustment value corresponding to each configuration characteristic value and pushing the configuration characteristic value to the front end for content display on the basis of the trial calculation result and a preset prompt strategy. And intelligent parameter adjustment is performed through data analysis, so that the efficiency and scientificity of parameter adjustment are improved.

Description

Parameter adjustment method, device, equipment and storage medium

Technical Field

The present disclosure relates to the field of data processing technologies, and in particular, to a method, an apparatus, a device, and a storage medium for parameter adjustment.

Background

With the development of internet technology, electronic office work is a trend of selecting enterprise office modes. An important component in the business offices is contract management, and in the contract management process, the business needs to rely on more data enabling and intelligent support in order to improve the timeliness and accuracy of contract processing.

In order to promote the achievement of contract process timeliness, the existing management platform basically has functional modules of service flows, can complete the functions of traditional multi-stage approval, countersignature and the like, and meanwhile, each node is provided with corresponding parameters to early warn and prompt specific business processes such as contract signing and the like. For example, a business person can set standard time of each time node on the system, perform user portrayal on time timeliness of processing contract nodes by a system user, and set a trigger threshold of flow clinical early warning for specific business processes such as contract signing and the like. However, these parameters are often required to be manually analyzed and further empirically judged by a large number of business personnel through a large number of statistical analysis reports, the efficiency is low, and the effect is unexpected due to insufficient scientificity of parameter setting.

Therefore, how to provide an effective parameter adjustment method to improve the efficiency and scientificity of parameter adjustment is a problem to be solved.

Disclosure of Invention

In view of this, the present application provides a method, apparatus, device and storage medium for parameter adjustment, which aims to improve efficiency and scientificity of parameter adjustment.

In a first aspect, the present application provides a parameter adjustment method, which is characterized in that the method includes:

acquiring service data in a current environment and configuration characteristic values corresponding to the service data; the configuration characteristic value is a threshold value and a parameter value corresponding to the service data;

pre-analyzing each configuration characteristic value corresponding to the service data by utilizing a preset data processing rule based on the service data to obtain a pre-analysis result; the pre-analysis result comprises whether each configuration characteristic value is in a preset range or not;

when each configuration characteristic value in the pre-analysis result is in a preset range, carrying out configuration characteristic value validity judgment on each configuration characteristic value corresponding to the service data by utilizing a configuration characteristic value validity judgment rule based on the service data to obtain a configuration characteristic value validity judgment result;

Based on the configuration characteristic value validity judgment result, carrying out trial calculation on each configuration characteristic value through a preset trial calculation rule to obtain a trial calculation result;

based on the trial calculation result and a pre-configured prompting strategy, generating configuration feature adjustment values corresponding to the configuration feature values, and pushing the configuration feature adjustment values to the front end for content display.

Optionally, the preset data processing rule includes a data extraction rule, a data rejection rule and a data analysis rule, and the pre-analyzing each configuration feature value corresponding to the service data by using the preset data processing rule based on the service data to obtain a pre-analysis result includes:

performing data extraction in the service data by utilizing the data extraction rule to obtain first pre-analysis data;

removing the maximum value and the minimum value in the first pre-analysis data by using the data removing rule to obtain second pre-analysis data;

and determining the parameter value standard reaching rate of the parameter value corresponding to the service data based on the second pre-analysis data, and pre-analyzing the parameter value standard reaching rate corresponding to the parameter value by utilizing the data analysis rule to obtain a pre-analysis result.

Optionally, the configuration feature value validity judging rule includes a plurality of sub-judging rules, where the plurality of sub-judging rules at least includes a fact judging rule, a prediction judging rule, an association judging rule and a threshold critical judging rule, and when each configuration feature value in the pre-analysis result is in a preset range, performing configuration feature value validity judgment on each configuration feature value corresponding to the service data by using the configuration feature value validity judging rule based on the service data to obtain a configuration feature value validity judging result, where the configuration feature value validity judging result includes:

When the parameter value corresponding to the parameter value in the pre-analysis result is in a preset range, carrying out fact judgment on each configuration characteristic value corresponding to the service data by utilizing a fact judgment rule based on the service data to obtain a fact judgment result;

based on the service data, carrying out prediction judgment on each configuration characteristic value corresponding to the service data by using a prediction judgment rule to obtain a prediction judgment result;

performing association judgment on each configuration characteristic value corresponding to the service data by using an association judgment rule based on the service data to obtain an association judgment result;

threshold critical judgment is carried out on each configuration characteristic value corresponding to the service data by utilizing threshold critical judgment based on the service data to obtain a threshold critical judgment result;

and obtaining a configuration characteristic value validity judgment result by utilizing probability superposition based on the fact judgment result, the prediction judgment result, the association judgment result and the threshold critical judgment result.

Optionally, the performing association judgment on each configuration feature value corresponding to the service data by using an association judgment rule based on the service data to obtain an association judgment result includes:

presetting deviation scale values according to the configuration characteristic values; the deviation scale value is used for acquiring an association judgment object when association judgment is carried out;

And acquiring an association judgment object through the deviation scale value, and comparing each configuration characteristic value corresponding to the business data of the association judgment object with each configuration characteristic value corresponding to the business data to obtain an association judgment result.

Optionally, the deviation scale value at least includes a first deviation scale value and a second deviation scale value, and the obtaining the associated judgment object through the deviation scale value includes:

when the result of acquiring the association judgment object through the first deviation scale value does not exist, adjusting the first deviation scale value to be a second deviation scale value, and acquiring the association judgment object through the second deviation scale value; the second deviation scale value is larger than the first deviation scale value, and the second deviation scale value is an integer multiple of the first deviation scale.

Optionally, the method further comprises:

and pre-configuring preset data processing rules, data processing logics corresponding to configuration feature value validity judging rules and preset trial calculation rules and parameters corresponding to the data processing logics.

Optionally, after generating the configuration feature adjustment value corresponding to each configuration feature value based on the trial calculation result and the pre-configured prompting policy and pushing the configuration feature adjustment value to the front end for content display, the method further includes:

Responding to the selection of the configuration characteristic adjustment value corresponding to each configuration characteristic value by a user, and adjusting each configuration characteristic value into the configuration characteristic adjustment value;

monitoring the validity of the configuration characteristic adjustment value according to a preset period and service data corresponding to the preset period;

when the configuration feature adjustment value does not have effectiveness, adjusting the configuration feature adjustment value based on a preset adjustment rule to monitor the configuration feature adjustment value;

and adjusting each configuration characteristic value to be the monitoring configuration characteristic adjustment value.

In a second aspect, the present application provides a parameter adjustment device, the device comprising:

the acquisition module is used for acquiring service data in the current environment and each configuration characteristic value corresponding to the service data; the configuration characteristic value is a threshold value and a parameter value corresponding to the service data;

the pre-analysis module is used for pre-analyzing each configuration characteristic value corresponding to the service data by utilizing a preset data processing rule based on the service data to obtain a pre-analysis result; the pre-analysis result comprises whether each configuration characteristic value is in a preset range or not;

the configuration characteristic value validity judging module is used for carrying out configuration characteristic value validity judgment on each configuration characteristic value corresponding to the service data by utilizing a configuration characteristic value validity judging rule based on the service data when each configuration characteristic value in the pre-analysis result is in a preset range so as to obtain a configuration characteristic value validity judging result;

The trial calculation module is used for carrying out trial calculation on each configuration characteristic value through a preset trial calculation rule based on the configuration characteristic value validity judgment result to obtain a trial calculation result;

and the prompt module is used for generating configuration characteristic adjustment values corresponding to the configuration characteristic values based on the trial calculation result and a pre-configured prompt strategy, and pushing the configuration characteristic adjustment values to the front end for content display.

In a third aspect, the present application provides a parameter adjustment device comprising a memory for storing instructions or code and a processor for executing the instructions or code to cause the device to perform the parameter adjustment method of any one of the preceding aspects.

In a fourth aspect, the present application provides a computer storage medium having code stored therein, which when executed, implements the parameter adjustment method of any one of the preceding first aspects.

The application provides a parameter adjustment method. When the method is executed, firstly, service data in a current environment and each configuration characteristic value corresponding to the service data are obtained, then each configuration characteristic value corresponding to the service data is pre-analyzed by utilizing a preset data processing rule based on the service data to obtain a pre-analysis result, then when each configuration characteristic value in the pre-analysis result is in a preset range, each configuration characteristic value corresponding to the service data is subjected to configuration characteristic value validity judgment by utilizing a configuration characteristic value validity judgment rule based on the service data to obtain a configuration characteristic value validity judgment result, then each configuration characteristic value is subjected to trial calculation by utilizing a preset trial calculation rule based on the configuration characteristic value validity judgment result to obtain a trial calculation result, and finally, a configuration characteristic adjustment value corresponding to each configuration characteristic value is generated and pushed to the front end for content display based on the trial calculation result and a preset prompt strategy. Through the step-by-step analysis and result trial calculation of a large amount of data, objective and timely adjustment of each configuration characteristic value in the service execution node is ensured, meanwhile, intelligent parameter adjustment is realized through data analysis, the influence of manpower waste and manual subjective participation is reduced, and the efficiency of parameter adjustment is improved.

Drawings

In order to more clearly illustrate the present embodiments or the technical solutions in the prior art, the drawings that are required for the embodiments or the description of the prior art will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present application, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.

Fig. 1 is a flowchart of a parameter adjustment method provided in an embodiment of the present application;

fig. 2 is a schematic technical principle diagram of a parameter adjustment method according to an embodiment of the present application;

fig. 3 is a schematic content presentation diagram of a configuration feature adjustment value of a contractual service according to an embodiment of the present disclosure;

FIG. 4 is a flowchart of a possible implementation of step S102 according to an embodiment of the present application;

FIG. 5 is a flowchart of a possible implementation of step S103 according to an embodiment of the present application;

FIG. 6 is a diagram of an automatic layout of a fact-judging system according to an embodiment of the present application;

fig. 7 is a schematic diagram of probability stacking according to an embodiment of the present application;

fig. 8 is a schematic structural diagram of a parameter adjusting device according to an embodiment of the present application.

Detailed Description

As described above, in order to promote the achievement of the timeliness of the contract process, the existing management platform basically has functional modules of service flows, and can complete the functions of traditional multi-stage approval, meeting and the like, and meanwhile, each node is provided with corresponding parameters to early warn and prompt specific business processes such as the contract signing and the like. For example, a business person can set standard time of each time node on the system, perform user portrayal on time timeliness of processing contract nodes by a system user, and set a trigger threshold of flow clinical early warning for specific business processes such as contract signing and the like. However, whether the time of the process node set by the user is reasonable, whether the time threshold setting of the user portrait business processing is reasonable, whether the time threshold setting of the process temporary early warning is reasonable and the like are set by a large amount of data statistics analysis report forms based on experience at present, so that the setting of the corresponding parameters does not have scientific data support, a large amount of manpower and material resources are wasted, and meanwhile, the efficiency of parameter value adjustment is low and not timely due to the limitation of manpower.

In view of this, the present application provides a parameter adjustment method. When the method is executed, firstly, service data in a current environment and each configuration characteristic value corresponding to the service data are obtained, then each configuration characteristic value corresponding to the service data is pre-analyzed by utilizing a preset data processing rule based on the service data to obtain a pre-analysis result, then when each configuration characteristic value in the pre-analysis result is in a preset range, each configuration characteristic value corresponding to the service data is subjected to configuration characteristic value validity judgment by utilizing a configuration characteristic value validity judgment rule based on the service data to obtain a configuration characteristic value validity judgment result, then each configuration characteristic value is subjected to trial calculation by utilizing a preset trial calculation rule based on the configuration characteristic value validity judgment result to obtain a trial calculation result, and finally, a configuration characteristic adjustment value corresponding to each configuration characteristic value is generated and pushed to the front end for content display based on the trial calculation result and a preset prompt strategy.

Therefore, through the multi-step analysis judgment and result trial calculation of a large amount of data, objective and timely adjustment of each configuration characteristic value in the service execution node is ensured, intelligent parameter adjustment is realized through data analysis, the influence of manpower waste and human subjective participation is reduced, and the parameter adjustment efficiency is improved.

It should be noted that, the embodiment of the present application does not limit the execution subject of the parameter adjustment method, and for example, the parameter adjustment method of the embodiment of the present application may be applied to a data processing device such as a terminal device or a server. The terminal device may be a smart phone, a computer, a personal digital assistant (Personal Digital Assistant, PDA), a tablet computer, or the like. The servers may be stand alone servers, clustered servers, or cloud servers.

For the purpose of making the objects, technical solutions and advantages of the embodiments of the present application more clear, the technical solutions of the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is apparent that the described embodiments are only some embodiments of the present application, but not all embodiments. All other embodiments, which can be made by one of ordinary skill in the art without undue burden from the present disclosure, are within the scope of the present disclosure.

Referring to fig. 1, fig. 1 is a flowchart of a parameter adjustment method according to an embodiment of the present application. As described in connection with fig. 1, the parameter adjustment method provided in the embodiment of the present application may include:

s101: and acquiring service data in the current environment and each configuration characteristic value corresponding to the service data.

In this embodiment, the configuration feature value is a threshold value and a parameter value corresponding to the service data. Specifically, before executing the service flow, different monitoring parameters and thresholds are set according to different service flows and nodes corresponding to the service flow, so as to ensure that the service can be efficiently operated in the execution process. Taking contract service as an example, a user can configure a node time standard value, a user service processing timeliness portrayal threshold value, a flow temporary early warning threshold value and the like of each contract executing node in the foreground. The node time standard value, the user service processing timeliness portrait threshold and the process temporary early warning threshold are parameters and thresholds for evaluating and monitoring service and process performances. The node time standard value refers to the expected time required by each node or step in one process or service. The node time standard value is determined according to business requirements, best practices or protocols and is used for measuring and monitoring the execution time of each node. The node time criteria may help identify delays, bottlenecks, or inefficient nodes and support improved and optimized flows. The user traffic processing timeliness portraits threshold is a threshold for measuring and evaluating the timeliness of user traffic processing. The user traffic processing time-rate portrayal threshold represents the proportion of traffic or tasks completed on time within a particular time window. The specific threshold setting depends on different business requirements and industry standards. The user traffic processing timeliness portrayal threshold may be used to monitor and optimize traffic flows, ensuring timely processing of user requests and demands. The flow deadline early warning threshold is a threshold set for tasks or activities in the flow that have an expiration date or time limit. The flow deadline early warning threshold represents a point in time of an early warning or reminder triggered just before the expiration date. The setting of the flow deadline early warning threshold can help the manager and the participant take action in time to avoid delay or timeout. The specific threshold may be customized based on the complexity, importance, and business requirements of the process.

In this embodiment, according to the execution condition of the user in the contract service, service data in the current environment and each configuration feature value corresponding to the service data are obtained.

In an alternative embodiment, in order to analyze the acquired service data and the parameter values corresponding to the service data, each rule that each configuration feature value is adjusted needs to be configured in advance, so as to be used for analyzing and adjusting each configuration feature value. Thus, the method further comprises: and pre-configuring preset data processing rules, data processing logics corresponding to configuration feature value validity judging rules and preset trial calculation rules and parameters corresponding to the data processing logics.

The preset data rules may include a data extraction rule, a data rejection rule and a data analysis rule, and the configuration feature value validity judgment rule may include a plurality of sub-judgment rules, where the plurality of sub-judgment rules at least includes a fact judgment rule, a prediction judgment rule, an association judgment rule and a threshold critical judgment rule. When the preset data processing rule and the configuration characteristic value validity judging rule are set, because the data types of extraction analysis are similar, the same items corresponding to the preset data processing rule and the configuration characteristic value validity judging rule in the setting process of the preset data processing rule and the configuration characteristic value validity judging rule can have range difference in the setting of the rule, but the substantial process is consistent. For example, a fact judgment rule and a preset data processing rule in the feature value validity judgment rule are configured, and the fact judgment rule is further expanded and subdivided on the basis of time granularity on the data extraction rule compared with the preset data processing rule, so that the purpose of verifying and refining a pre-analysis result on the fact judgment dimension is achieved.

Taking contract service as an example, a foreground is used for managing rule configuration methods for adjusting configuration characteristic values (including parameters, threshold values and the like). For example, the monitoring data content setting and the corresponding rejection rule configuration may specifically include a setting method such as a monitoring data duration setting, a data amount sample amount setting, an abnormal data rejection range setting, and a data comparison analysis control group setting. For another example, the monitoring rules and analysis rules of the configuration data may specifically include a weight ratio of different time intervals for data prediction such as standard or threshold adjustment, overall user processing timeliness, and process temporary early warning occurrence rate, a threshold for determining user traffic processing timeliness portraits threshold, process temporary early warning threshold setting validity (forming validity judgment), a monitoring parameter for determining a feature to be adjusted by a node time standard, an analysis dimension combination setting for accurately determining node timeliness, user traffic processing timeliness portraits threshold, and process temporary early warning threshold to be adjusted, such as a target value ±bias threshold, a continuous month comparison positive or negative month number, and the like, an adjustment scale setting for trial adjustment of the target value, and an available judgment feature value rule setting for trial adjustment of the target value.

The method achieves the aim of parameter value adjustment by respectively configuring analysis rules such as monitoring analysis threshold and standard effectiveness, monitoring data elimination, data analysis comparison judgment rules, judgment and trial calculation scales, trend prediction judgment and trial calculation rules, comparison parameters, deviation judgment critical threshold, trial calculation adjustable value judgment rules and the like.

S102: pre-analyzing each configuration characteristic value corresponding to the service data by utilizing a preset data processing rule based on the service data to obtain a pre-analysis result; the pre-analysis result includes whether the configuration feature values are in a preset range.

After the service data and each configuration characteristic value corresponding to the service data in the current environment are obtained, each configuration characteristic value corresponding to the service data is pre-analyzed according to the service data of the user, so that the service execution condition of the user under the constraint of each configuration characteristic value is obtained, and a pre-analysis result whether each configuration characteristic value accords with a preset standard is obtained.

In addition, when the pre-analysis is carried out, necessary processing is automatically carried out on the data, firstly, the maximum value and the minimum value are removed, secondly, the data access range of the analysis data with the sample size not reaching the standard is automatically enlarged, and the removal processing can be carried out if the analysis data does not reach the standard.

It should be noted that the preset standard is whether each configuration characteristic value is within a preset range. Specifically, comparing the parameter value standard reaching rate corresponding to each configuration characteristic value with the target standard reaching rate corresponding to each configuration characteristic value, and if the parameter value standard reaching rate corresponding to each configuration characteristic value is lower than the lower limit, indicating that the parameter value is set too strictly; if the parameter value standard reaching rate corresponding to each configuration characteristic value is higher than the upper limit, the parameter value is set too loosely.

Taking contract service as an example, the pre-analysis result can comprise a node time standard, a service processing timeliness portrayal threshold value and a flow temporary early warning threshold value, the value of the service processing timeliness under the corresponding parameter value, the timeliness portrayal threshold value of the service processing and the time completion proportion of the service after monitoring the flow temporary early warning are monitored to reach a monitoring target value or not, and the deviation degree is large, if the deviation degree is larger than a certain value, the preliminary judgment is to be adjusted; and analyzing and judging the timeliness of the node processing, monitoring the timeliness of the service processing under the standard, setting a threshold value for monitoring, and preliminarily judging that the timeliness is to be adjusted when the timeliness does not reach a certain value.

Specifically, taking a node time standard value as an example, if a node is a node time standard value of 5 working days, obtaining that the service processing time rate under the node time standard is 100% by analyzing service data, namely, all people complete the service corresponding to the node within a specified time, and comparing the data to obtain that the service processing time rate target value under the node time standard is 80% ± deviation threshold value of 5%, 100% exceeds the upper limit of 85%, and the suspected standard is loose; if the service processing time rate under the node time standard is 60% by analyzing the service data, 60% is lower than the lower limit of 75%, and the suspected standard is strict. When each configuration characteristic value is not in a preset range (including lower limit and higher limit), the parameter value is indicated to have a problem, and adjustment is needed. It can be understood that if the parameter value standard reaching rate corresponding to each configuration characteristic value is at the target standard reaching rate corresponding to each configuration characteristic value, the parameter value is set reasonably, and no adjustment is needed.

In this embodiment, when pre-analysis is performed, a data mining technique is used to extract data from the acquired service data. Data mining is a process of extracting useful information and patterns from a large amount of data by using statistical, machine learning, artificial intelligence, and the like methods. Common data mining techniques include cluster analysis, classification analysis, association rule mining, anomaly detection, predictive modeling, and the like. The data mining technology can help discover hidden modes, association relations and trends in service data, the data mining analysis technology can provide deep insight into the data, help organizations to know key features, trends and modes in the data, and meanwhile, the data mining technology can automatically process a large amount of data and quickly identify the modes and relations. Therefore, the data mining technology is adopted in the embodiment, so that the efficiency of data analysis and decision making can be greatly improved, and the requirement of manual operation is reduced. It should be noted that, the user may select a corresponding data mining technology according to the pre-analysis requirement, and the embodiment of the present application does not limit a specific data mining technology in the application process.

S103: and when each configuration characteristic value in the pre-analysis result is in a preset range, carrying out configuration characteristic value validity judgment on each configuration characteristic value corresponding to the service data by utilizing a configuration characteristic value validity judgment rule based on the service data to obtain a configuration characteristic value validity judgment result.

In this embodiment, when each configuration feature value obtained through the pre-analysis result is in a preset range exceeding the upper limit or lower than the lower limit, it is indicated that each corresponding configuration feature value needs to be adjusted. However, since only a single data mining is adopted during the pre-analysis, the details of the service data may be lost, and thus, a further more accurate judgment needs to be performed on the pre-analysis result to ensure the validity of the threshold value corresponding to each parameter. Specifically, the validity judgment of the configuration feature value can be carried out according to the fact judgment of data deviation, the comparison judgment of the data deviation and the average value of the comparison group, the future trend prediction judgment, the adjustment threshold critical judgment and other four dimensions further judgment, and the configuration feature value validity judgment result of each configuration feature value is formed to obtain the accurate judgment result. Taking contract business as an example, accurate judgment of node time standard, flow threshold value and user business processing timeliness portrayal threshold value adjustment can be formed. Therefore, in this embodiment, the configuration feature value validity judgment rule is constructed, and more detailed and accurate configuration feature value validity judgment is performed on each configuration feature value corresponding to the service data based on the service data, and meanwhile, the adjustment direction, such as adjustment increase or adjustment decrease, can be judged due to the existence of the adjustment threshold critical judgment.

It should be noted that, the configuration feature value validity judgment rule may include four parts, namely, a fact judgment, a prediction judgment, an association judgment and a threshold critical judgment. Further, in order to ensure accuracy of the judgment result, the order of the four steps may be defined, for example, the judgment may be performed according to the first step of facts, the second step of prediction judgment, the third step of association judgment, the fourth step of threshold critical judgment, and if one step of judgment is not performed, the process is automatically skipped to the next step or the process is automatically ended. Furthermore, in the process of carrying out fact judgment, prediction judgment, association judgment and threshold critical judgment, the two-eight principle can be combined for different judgment, and a majority of passing methods are adopted, so that a probability superposition method is adopted, and the accuracy of judgment is greatly improved.

Wherein, the fact judgment comprises the combination of total quantity judgment and month judgment, and the judgment is carried out according to the principle that the sample quantity is enough, the total accords with the total quantity and the majority accords with the month (the majority accords with more than 80 percent month); the prediction judgment comprises the steps of meeting a trigger adjustment condition according to the enough sample size and the predicted value; the association judgment comprises comparison of comparison parameters with consistent other conditions and different standards or thresholds, and is usually compared with the average of the whole network and other provincial company data forms, and the comparison group judges that the trigger threshold or the standard is adjustable; the threshold critical judgment is combined with the comparison data judgment, combined with the trial calculation adjustment scale, the current value is calculated by trial calculation with +1 scales, and the judgment is adjustable or most of months are adjustable. And through superposition judgment of multiple dimensions, the probability of the superposition judgment is greater than 80%, and the comprehensive adjustability is considered to be improved. Meanwhile, the threshold critical judgment outputs each configuration characteristic value to be adjusted to be large or small.

Specifically, probability superposition is that the first step, the second step, the third step and the fourth step respectively borrow the results of the previous step and reference, and the principle of probability superposition is applied to accurately judge the standard to be adjusted.

The first accurate judgment adopts 3 characteristic values on the same side of the whole body, the month, the statistical data month and the whole body exceeding 80% to carry out probability superposition judgment, the predicted value of the second step and 2 characteristic value judgments on the same side of the predicted value data and the statistical data exceeding 80% are superposed, 3 characteristic values are respectively superposed in the third step, and the value (single or multiple value) obtained in the association judgment is in the upper limit or the lower limit of the super-target value combination critical threshold value compared with the target value combination critical threshold value; (2) If the values are a plurality of values, more than 80% of the correlation judgment results in that the monitoring values are all at the lower limit or the upper limit of the target value combination critical threshold; (3) The upper limit or the lower limit of the threshold value combination critical threshold value of the predicted value and the upper limit or the lower limit obtained by the fact judgment are positioned on the same side, and the threshold value critical judgment characteristic value of the fourth step (the trial value is calculated in a mode of adding or subtracting 1 time of the scale value; the condition that the judgment of the trial value can not be continued is that the conclusion obtained by the last calculation is not adjustable according to the accurate judgment method, the conclusion obtained by the previous calculation can be adjusted according to the accurate judgment method, and the adjustment value suggested by the trial calculation takes the previous value obtained by the last trial calculation) and the like, thereby increasing the judgment accuracy.

In an alternative embodiment, in the accurate determination, in order to further exclude abnormal data generated by human in the actual application process, an abnormal data removing rule may be preset, so as to automatically remove, according to the abnormal data removing rule, the type of the actually suspected management type problem that does not need to calculate the adjustment threshold or parameter yet. Among them, typical management class problems are 2 classes, in which, in the case of the same standard, service data of only a few targets (e.g., 50 target institutions, only 1 target institution) does not reach the standard, or in the case of the same standard, service data of a plurality of targets (e.g., 50 target institutions, 49 target institutions) does not reach the expectations. For example, the threshold value of the processing time rate of the contract business of a certain branch office is the same as that of other comparison groups, but the actual index of the branch structure is automatically compared with the data of the comparison groups under the threshold value, the processing time rate of the contract business of the branch office is recognized to be obviously lower than that of the comparison groups for a plurality of months or a plurality of weeks, and management intervention advice can be automatically pushed and terminated when the situation occurs. Thereby improving the efficiency of parameter adjustment.

S104: and carrying out trial calculation on each configuration characteristic value through a preset trial calculation rule based on the configuration characteristic value validity judgment result to obtain a trial calculation result.

In this embodiment, the configuration feature value validity determination result including the adjustment direction has been obtained in step S103, so that the preset adjustment value trial calculation can be performed one by one according to the current value and the minimum adjustment scale based on the corresponding adjustment direction and the trial adjustment scale. According to the adjustment scale, the direction to be adjusted and the adjustment direction are determined by combining the preamble, and the simulation trial calculation is performed, wherein the simulation trial calculation mode is also a three-step simulation mode based on the fact, the prediction mode, the comparison data mode, the characteristic value superposition judgment mode and the like.

Specifically, the trial calculation can further judge according to the four dimensions of data deviation fact judgment, comparison judgment with a comparison group average value, future trend prediction judgment, adjustment threshold critical judgment and the like, so as to form accurate adjustment judgment of node time standard, flow clinical threshold and user business processing timeliness portrait threshold, and judge whether parameter values are adjustable according to a trial calculation validity rule. That is, in this embodiment, when the parameter value needs to be adjusted, the preset calculation rule is adopted to calculate each configuration feature value in a trial manner so as to obtain a parameter value that meets the expectation, that is, the parameter value obtained after the trial calculation meets the parameter value target achievement rate.

Specifically, the preset calculation rule may obtain a parameter value a for the current parameter value+the minimum adjustment scale, perform configuration feature value validity judgment on the parameter value a as a parameter value corresponding to service data, if the configuration feature value validity judgment result of the parameter value a does not conform to the expected feature, it indicates that the parameter value a may be further processed as a configuration feature adjustment value, and if the configuration feature value validity judgment result of the parameter value a conforms to the expected feature, it indicates that the parameter value a still does not meet the parameter value target standard rate for the service data, and may not be further processed as the configuration feature adjustment value. The process of determining the validity of the configuration feature value is already described, and will not be described in detail herein.

It should be noted that, after the calculation according to the above steps is completed and the judgment is feasible, since the parameter value a still does not meet the target standard rate of the parameter value for the service data, it cannot be further used as the configuration feature adjustment value, and therefore, it is necessary to add a scale to calculate again until the increased configuration feature adjustment value meets the target standard rate of the parameter value.

In this embodiment, after the preliminary analysis judgment and the accurate analysis judgment, one or more adjustable configuration feature values can be obtained, and the adjustment direction of the parameter value, whether the scale is adjusted positively or negatively, can be obtained. The trial calculation value is calculated according to the mode that the scale value is 1 time, 2 times and 3 times are added or subtracted in sequence, and the condition that the trial calculation value is judged to be unable to continue is that the conclusion obtained in the last calculation is unable to be adjusted according to the accurate judgment method, the conclusion obtained in the previous calculation can be adjusted according to the accurate judgment method, and the adjustment value of the trial calculation proposal takes the previous value of the last trial calculation.

Taking contract service as an example, the method can comprise node time standard value, user service processing timeliness portrait threshold and flow early warning threshold adjustment, and the following embodiment exemplarily provides a process for adjusting the parameters.

For the node time standard, under the condition of definitely adjusting the direction by scales, the node time standard is continuously calculated according to scales of 2 times, 3 times and X times, and the condition that the trial calculation value is judged to be unable to be continued is that the conclusion obtained by the last calculation is unable to be adjusted according to the accurate judgment method, the conclusion obtained by the previous calculation can be adjusted according to the accurate judgment method, and the adjustment value of the trial calculation proposal takes the previous value of the last trial calculation.

And (3) the flow early warning threshold value is adjustable, or the timeliness portrait threshold value of user business processing is adjustable, under the condition of clear scale adjustment direction, the correlation judgment of trial calculation properties is adopted according to scales of 2 times, 3 times and X times, and the condition that the trial calculation value is judged to be unable to continue is that the conclusion obtained by the last calculation is unable to be adjusted according to an accurate judgment method, the conclusion obtained by the previous calculation is adjustable according to the accurate judgment method, and the adjustment value of the trial calculation proposal is taken as the previous value of the last trial calculation. Meanwhile, for the threshold adjustment of the standard tightening type, a satisfaction principle can be utilized, a marginal effect decreasing method is started, the improvement amplitude of an expected result under the standard after the tightening is compared with the threshold standard before the tightening, and if the improvement amplitude shows a descending or narrowing trend and the effectiveness threshold is triggered proportionally, the method is considered to be not feasible according to the latest adjustment scheme.

The satisfaction principle (satisficing) is also called utility maximization principle, and is a common criterion in decision theory. It teaches that decisions should be pursued to maximize satisfaction or utility of the results when making decisions. Marginal effect, sometimes also referred to as marginal contribution, refers to the effect of an increase or decrease in a variable on the result. The decrease in the marginal effect is a phenomenon in which the effect on the result gradually decreases as a certain variable increases. In short, a decreasing marginal effect means that as a variable increases, its contribution to the result gradually decreases. In the embodiment of the application, the satisfaction principle and the marginal effect decreasing method are combined, so that the parameter value obtained by trial calculation is more scientific.

Fig. 2 is a schematic diagram of a technical implementation process of a parameter adjustment method provided in an embodiment of the present application, and in fig. 2, a is service data of service processing timeliness and a threshold corresponding to the service processing timeliness in the current environment in provinces a and B. B is a possible pre-analysis result presentation form after pre-analysis, namely, the requirement of adjusting the threshold value exists in the province B through a line diagram, and the suspected standard is strict. And then accurately judging through the fact judgment, the prediction judgment, the comparison judgment and the threshold critical judgment in the step c, namely, judging the validity of the configuration characteristic value to obtain a configuration characteristic value validity judgment result. And finally, carrying out trial calculation on the threshold value in the step B through the conclusion to be adjusted, the adjustment scale checking and the adjustment pre-calculation rule in the step d to obtain a trial calculation result.

S105: based on the trial calculation result and a pre-configured prompting strategy, generating configuration feature adjustment values corresponding to the configuration feature values, and pushing the configuration feature adjustment values to the front end for content display.

In this embodiment, the corresponding trial calculation result has been obtained in step S104, and fig. 3 is a schematic diagram showing the content of the configuration feature adjustment value of the contract service provided in this embodiment. In combination with the illustration in fig. 3 (the threshold adjustment suggestion of the counting rate in the process of user system contract proxy), for the trial calculation result, a corresponding manager can be pushed by using a pre-configured prompting strategy, the original standard value and the new standard suggestion value are marked, and a preliminary analysis judgment conclusion and a precise judgment conclusion under the original standard value are automatically generated as analysis basis, so that a user can check the calculation detail information by himself. The user may click on the pop-up interface to confirm that the adjusted criteria or threshold is in effect. In an alternative embodiment, the system may also be set in an automatic mode, and the system monitors data in the background of the system, and the system automatically adjusts and validates each configuration feature adjustment value, such as a contract service node time standard value, a contract service flow deadline early warning threshold value, and a contract service processing timeliness user image threshold value. It will be appreciated that the user can review the detailed analysis process based on the conclusion by clicking on the detail review.

In an optional embodiment, since the adjusted parameter value does not conform to the preset range due to other factors in the subsequent service execution process, the adjusted reference values need to be continuously monitored and adjusted again, so after the configuration feature adjustment values corresponding to the configuration feature values are generated based on the trial calculation result and the pre-configured prompting policy and pushed to the front end for content display, the method further includes: responding to the selection of the configuration characteristic adjustment value corresponding to each configuration characteristic value by a user, and adjusting each configuration characteristic value into the configuration characteristic adjustment value; monitoring the validity of the configuration characteristic adjustment value according to a preset period and service data corresponding to the preset period; when the configuration feature adjustment value does not have effectiveness, adjusting the configuration feature adjustment value based on a preset adjustment rule to monitor the configuration feature adjustment value; and adjusting each configuration characteristic value to be the monitoring configuration characteristic adjustment value.

In this embodiment, after a user adjusts a parameter value, service data corresponding to a preset period is obtained through the preset period, validity of an adjustment value of a configuration feature is monitored through analysis of the service data, specifically, a background of the system automatically monitors effect comparison before and after adjustment of the parameter value corresponding to the preset monitoring period, and automatically gives a valid or invalid conclusion based on the preset conclusion period, the preset monitoring period of the automatic monitoring can be 6 months, the preset conclusion period of the conclusion can be 1 month, N can be 6 in combination with a conclusion of scrolling N times, and if M times is greater than or equal to M times and is invalid, M can be 2, a monitoring period is expanded once, the preset period is expanded to 2 times and is continuously monitored, if the continuous monitoring comparison is still invalid, the parameter value automatically rolls back 1 scale, and the monitoring is continuously performed. If Q times less than or equal to the preset period are monitored to be invalid, Q can be 1, and the default rule or standard is adjusted to be valid.

In the embodiment, the integration and embedded system built by the service personnel based on experience ensures objective and timely adjustment of each configuration characteristic value through presetting multi-step data processing and analysis rules, and a large amount of data are learned, mined, trained and optimized on the basis of rule setting by the service personnel and through data mining and the like in the process of data acquisition, so that the effect of efficiently and scientifically adjusting the parameter values is achieved. For each configuration characteristic value of contract business, the following problems are solved in a targeted way: (1) The contract process time management standard effectiveness analysis, the contract business process temporary early warning threshold effectiveness analysis and the contract business processing user timeliness portrait threshold setting effectiveness analysis are performed, whether analysis opinion is effective or not is intelligently given, and manual analysis and judgment are omitted; (2) The multi-dimensional judgment problems of rule adjustment such as contract process time standard, business process temporary early warning threshold, business processing user timeliness image threshold and the like are intelligently judged, judgment comments are intelligently given, whether adjustment is needed or not, and a conclusion is enhanced based on data; (3) The contract process time standard, the business process temporary early warning threshold value and the business processing user timeliness portrait threshold value adjustment simulation trial calculation model support to automatically give a suggested value, confirm to be effective by one key and give the optimal adjustment opinion according to data operation.

In an alternative embodiment, the parameter values corresponding to the application may be extended based on different services, for example, taking a contract service as an example, and the payment monitoring and early warning threshold after the contract money reaches the pay-off condition in the contract service may be analyzed and the early warning threshold may be intelligently adjusted. The method has reference value for monitoring and adjusting the effectiveness of the early warning threshold of other processes, monitoring and adjusting the node time standard, monitoring and adjusting the effectiveness of the timeliness portrait threshold of the system business processing, and the like.

In an optional embodiment, when the method and the device implement automatic adjustment and analysis on each configuration characteristic value of service data, because big data operation modes such as data mining are used, in order to improve operation efficiency, a micro-service micro-application technology mode and data center station calculation power expansion data analysis can be adopted aiming at the big data operation processing mode, and the data operation analysis is carried out in a mode of avoiding daytime peak, so that the system performance of conventional service execution is prevented from being influenced.

In the embodiment of the present application, there are a plurality of possible implementations of step S102 described in fig. 1, and the following description will be given. It should be noted that the implementations presented in the following description are only exemplary and not representative of all implementations of the embodiments of the present application.

Fig. 4 is a flowchart of a possible implementation of step S102 in the embodiment of the present application. The preset data processing rule includes a data extraction rule, a data rejection rule, and a data analysis rule, and referring to fig. 4, the pre-analyzing each configuration feature value corresponding to the service data by using the preset data processing rule based on the service data to obtain a pre-analysis result may include:

s201: and carrying out data extraction in the service data by utilizing the data extraction rule to obtain first pre-analysis data.

In the present embodiment, the length of time and the number of samples of data extraction are determined in advance by defining a data extraction rule. The time length prescribes the time range of data extraction, the number of samples prescribes the number of samples of the data extraction, and when the samples with the prescribed number of samples cannot be extracted in the initial preset time range, the time length is increased to carry out data extraction again to obtain first pre-analysis data.

In an alternative embodiment, after the first pre-analysis data is extracted, the first pre-analysis data is primarily analyzed according to a data analysis rule based on the first pre-analysis data, and specific description is given below by taking a node time standard in contract service, a user service processing timeliness portrayal threshold value and a flow deadline early warning threshold value as examples.

For monitoring the time standard of the contract service node, the service processing time satisfaction rate is extracted according to the time length (the time length is colloquially the data of the latest 6 months), the sample data volume reaches the standard, the extraction data range (when the time standard is set by the measurement province company, the data range is always the data of a certain province company, the calculation formula = the number of the monitoring nodes reaching the corresponding province time standard/the total service processing number of the nodes in the corresponding province company in the extraction time range, when the time standard is set by a certain department in the measurement, the service process responsible for the department corresponds to the service processing satisfaction condition of the node, the calculation formula = the number of the nodes reaching the time standard of the department in the extraction time range/the total service processing number of the node in the corresponding department in the extraction time range) is judged in the condition that the service time standard of the corresponding certain node of the specific service department is measured, and when the sample volume fails to reach the standard, the length of the extraction time range is automatically expanded to the latest 1 year.

For the user business processing timeliness portrait threshold monitoring, all contract-related business processing timeliness data within half a year of the corresponding provincial company is extracted. The calculation formula s1=achieves the standard number of the corresponding provincial time in the extraction time range/the total business processing number in the provincial company corresponding to the extraction time range. The amount of monitoring data also needs to meet or exceed the sample size specification.

For the process deadline early warning threshold, the proportion of the business completed on time after the process early warning is generated by all the contract-related business processes in half a year of the corresponding provincial company is extracted. The calculation formula s2=the number of business process temporary early warning which is completed on time after the corresponding provincial company generates the process temporary early warning in the extraction time range/the number of business process temporary early warning which is generated in the corresponding provincial company in the extraction time range. The amount of monitoring data also needs to meet or exceed the sample size specification.

In an alternative embodiment, in an actual application process, because there may be an uncertainty interference situation of service data over a long span time, for example, service data over two years may not have a referential property for historical reasons, noise may be introduced after the service data is extracted, and reliability of the extracted data is reduced. Further, an alarm of insufficient sample size can be returned to the front page.

S202: and eliminating the maximum value and the minimum value in the first pre-analysis data by utilizing the data eliminating rule to obtain second pre-analysis data.

In this embodiment, since the individual abnormal samples have a larger deviation effect on the results, it is necessary to reject the data that may seriously affect the results from the service data. In this embodiment, a method of clipping an average value is specifically adopted to perform data rejection. Clipping averages, also called truncated averages, are a method used statistically to measure the concentration trend, similar to average and median. The clipping average is the average calculated after discarding some of the highest and lowest data in the probability distribution or sample, by discarding the same amount of data at both the highest and lowest ends. The discarded data may be expressed as a percentage of the overall data, but may also be expressed as a fixed number discarded. According to the average value cutting statistical method, uniformly defining a data eliminating rule, carrying out statistical calculation on the extracted data in a time range according to weeks, eliminating the highest value and the lowest value, and eliminating interference items of administrative or holiday or other factors in the contract business handling process to obtain second pre-analysis data.

S203: and determining the parameter value standard reaching rate of the parameter value corresponding to the service data based on the second pre-analysis data, and pre-analyzing the parameter value standard reaching rate corresponding to the parameter value by utilizing the data analysis rule to obtain a pre-analysis result.

After the data is removed in step S202, the formula defined by the data extraction is changed to: (original numerator value-extraction highest value corresponding numerator value-extraction lowest value corresponding numerator value)/(original denominator value-extraction highest value corresponding denominator value-extraction lowest value corresponding denominator value). And then comparing and primarily analyzing the second pre-analysis data by utilizing the data analysis rule. The method specifically comprises the step of comparing a recalculated value after data elimination with a corresponding critical threshold value. For a specific analysis process, the process of data analysis is described in step S201, and will not be described here. The pre-analysis result can comprise that the data compared with the node service time standard is the service processing time rate target value +/-deviation threshold value under the node time standard; less than the lower limit, the suspected standard is strict; exceeding the upper line indicates that the suspected standard is relaxed. The data of the user business processing timeliness portrait threshold monitoring comparison is a system user contract business processing timeliness target value +/-deviation threshold under the user portrait threshold; less than the lower limit, the suspected standard is strict; exceeding the upper line indicates that the suspected standard is relaxed.

The data compared with the flow period early warning threshold value is that the business after the flow period early warning generated by the system under the flow period threshold value completes the proportional target value + -deviation threshold value on time; the suspected standard is relaxed when the lower limit is not reached; exceeding the upper line indicates that the suspected standard is strict.

In summary, the conclusion output by the data comparison and preliminary analysis system: node business time standard: the monitoring value is a1, and the compared data is a business processing time rate target value b1 plus or minus a deviation threshold value c1% under the node time standard; less than the lower limit (b 1-c 1), the suspected standard is strict; or exceeds the upper line (b1+c1), the suspected standard is relaxed.

User traffic processing timeliness portrayal threshold: the monitoring value is a2, and the compared data is a system user contract business processing time rate target value b2+/-deviation threshold value c2% under the user portrait threshold value; less than the lower limit (b 2-c 2), the suspected standard is strict; or exceeds the upper line (b2+c2), the suspected standard is relaxed.

The monitoring value is a3, and compared data is a proportion target value b3+/-deviation threshold value c3 which is completed on time by the business after the system flow temporary early warning under the flow temporary threshold value; less than the lower limit (b 3-c 3), the suspected standard is relaxed; exceeding the upper line (b3+c3), the suspected standard is strict.

In this embodiment, through preliminary analysis of each configuration feature value of service data, whether each configuration feature value is to be adjusted is preliminarily determined. Providing basis for subsequent data analysis. And the efficiency of subsequent data analysis is improved.

In the embodiment of the present application, there are a plurality of possible implementations of step S103 described in fig. 1, and the following description will be given. It should be noted that the implementations presented in the following description are only exemplary and not representative of all implementations of the embodiments of the present application.

Fig. 5 is a flowchart of a possible implementation of step S103 in the embodiment of the present application. The configuration feature value validity judging rule includes a plurality of sub-judging rules, where the plurality of sub-judging rules at least includes a fact judging rule, a prediction judging rule, an association judging rule and a threshold critical judging rule, and referring to fig. 5, when each configuration feature value in the pre-analysis result is in a preset range, performing configuration feature value validity judgment on each configuration feature value corresponding to the service data based on the service data by using the configuration feature value validity judging rule may include:

S301: and when the parameter value corresponding to the parameter value in the pre-analysis result is in a preset range, carrying out fact judgment on each configuration characteristic value corresponding to the service data by utilizing a fact judgment rule based on the service data to obtain a fact judgment result.

In this embodiment, the fact judgment calculation rule is mainly based on preliminary analysis judgment, and extends to combining data analysis and disassembly in the time length range into monthly statistical analysis, and month judgment is performed by taking the time length as an example, that is, the judgment value is changed from a value of 1 half year to a value of 1 half year+a value of 6 months, the calculation rule of the month value is the same as the preliminary analysis calculation rule of data comparison, the difference item 1 is changed into month in the time length range, and the difference item 2 is the maximum and minimum values of the day are removed according to the month by the removal rule; in addition, it is necessary to comprehensively consider whether the most month data of the last X months is at the target value combination critical threshold lower limit or upper limit compared with the target value ± critical threshold. And combining the two-eight rule to obtain a fact judgment result. The basic idea of the two-eight rule, also known as pareto rule or 80/20 rule, is that there are in many cases imbalanced and asymmetric relationships. By identifying and focusing on those factors with the greatest impact, resources can be managed more efficiently and efficiency can be improved. Therefore, in the embodiment of the application, two-eight rules are fused in each sub-judgment rule of the configuration characteristic value validity judgment rule to improve the efficiency. Fig. 6 is an automatic picture illustration of the fact judgment system in the embodiment of the application, and in combination with the illustration in fig. 6, basic information of various data, a line graph and a monitoring conclusion of the fact judgment, which are used for accurately judging the service processing timeliness image threshold value of the contractual service user, are provided, so that the user can intuitively see the specific information of the step, and a direct theoretical basis is provided for the subsequent selection and adjustment of parameter values by the user.

The features that the fact judgment concludes affirmative include: 1. 80% or more of the monthly data meet the required adjustment characteristics; 2. more than 80% of the monitoring months are on one side of the lower limit or the upper limit of the target value combination critical threshold; 3. and the whole monitoring value and more than 80% of the month data monitoring value are both in the lower limit or the upper limit of the target value combination critical threshold value within the time length. And the system automatically monitors, analyzes and judges that the 3 characteristics are met, and indicates that the fact judgment is passed.

In this embodiment, by analyzing service data with finer granularity in the time dimension, it can be more precisely determined whether the pre-analysis result is accurate, and the reliability of the conclusion to be adjusted of the parameter value to be adjusted is ensured.

S302: and carrying out prediction judgment on each configuration characteristic value corresponding to the service data by utilizing a prediction judgment rule based on the service data to obtain a prediction judgment result.

In this embodiment, in addition to the above determination of the parameter value to be adjusted obtained by the pre-analysis result at a finer time granularity, the determination of the parameter value to be adjusted may also be performed by predicting the trend of future data. Therefore, the prediction judgment calculation rule can be utilized to carry out prediction judgment on each configuration characteristic value corresponding to the service data to obtain a prediction judgment result. Specifically, the prediction judgment of the present embodiment refers to 136 a prediction algorithm, also called 136 mean line prediction, which is an application of ARIMA (autoregressive moving average) model. An ARIMA (autoregressive moving average) model can be used to analyze and predict one-dimensional time series data. The method is characterized in that the predicted time length cannot be excessively long, and the probability of the latest data guiding the future is higher. In the embodiment trend prediction application, trend value predictions of future 1 month time period or other such as time-lapse rate are developed according to values of 1 month, 3 months, 6 months. The data of the last 1 month, the data of the last 3 months and the data of the last 6 months are respectively matched with weights, the data prediction value of the next 1 month is calculated backwards, and a calculation formula s3=the weight of the data of the last 1 month, the weight of the data of the last one month+the weight of the data summation of the data of the last 3 months/the weight of the data summation of the 3 months of the last 3 months+the weight of the data summation of the month of the last 6 months/the weight of the data average of the data of the 6 months of the last 6 months is calculated. Constraint 1: the weight of the most recent 1 month data > the weight of the most recent 3 months data > the weight of the most recent 6 months data mean; constraint 2: weight of the most recent 1 month data + weight of the most recent 3 months data + weight of the most recent 6 months data mean = 100%. The predictable value comprises the service processing time rate under the node time standard, the system user contract service processing time rate, the service flow on-time completion rate after the service flow temporary early warning, and the like.

The prediction judgment results in positive conclusion characterized in that: 1. the predicted value is in the upper limit or the lower limit of the super-target value combination critical threshold value compared with the target value combination critical threshold value; 2. the upper limit or the lower limit of the combined critical threshold of the predicted value and the target value is on the same side as the upper limit or the lower limit obtained by the fact judgment. And if the characteristic of the 2 items is met, the prediction judgment is passed. That is, through the prediction and judgment, the parameter value to be adjusted still has adverse effects in the execution process of the subsequent service, and needs to be adjusted.

S303: and carrying out association judgment on each configuration characteristic value corresponding to the service data by using an association judgment rule based on the service data to obtain an association judgment result.

In this embodiment, besides the above-mentioned determination of the parameter value to be adjusted obtained by the pre-analysis result in a finer time granularity, and the determination of the parameter value to be adjusted by predicting the trend of future data, further the execution condition of the parameter value at other monitoring points may be compared laterally. Therefore, the service data of other monitoring points corresponding to the parameter value can be searched by using the association judgment rule for association comparison. Specifically, the correlation judgment is also called comparison judgment, and the main purpose is to find the values of other provinces or departments for comparison analysis aiming at the same monitoring point, judge, if a plurality of available values are available, the conclusion according to 80% is in order, and the main purpose is to avoid the significant deviation of the standard or threshold adjustment from the statistical rule.

In an optional embodiment, the performing, based on the service data, association judgment on each configuration feature value corresponding to the service data by using an association judgment rule to obtain an association judgment result includes: presetting deviation scale values according to the configuration characteristic values; the deviation scale value is used for acquiring an association judgment object when association judgment is carried out; and acquiring an association judgment object through the deviation scale value, and comparing each configuration characteristic value corresponding to the business data of the association judgment object with each configuration characteristic value corresponding to the business data to obtain an association judgment result. In this embodiment, in order to more accurately determine whether the parameter value to be adjusted in the pre-analysis result needs to be adjusted, verification may be performed by associating service data corresponding to the same parameter value of other objects, but the service data of other objects cannot be directly compared, and association should be selected to be performed by the nearest difference between the parameter value to be adjusted, so that an offset scale value may be preset to obtain an association determination object when performing association determination, then the association determination object may be obtained through the preset offset scale value, and meanwhile service data of the association determination object may be obtained, and each configuration feature value of the service data of the association determination object may be compared with each configuration feature value corresponding to the service data, so as to obtain an association determination result.

In an alternative embodiment, the setting of the deviation scale value cannot be too large or too small, so an initial deviation scale value may be set, and the size of the deviation scale value is continuously adjusted to obtain an ideal association judgment object, so the deviation scale value at least includes a first deviation scale value and a second deviation scale value, and the obtaining the association judgment object through the deviation scale value includes: when the result of acquiring the association judgment object through the first deviation scale value does not exist, the first deviation scale value is adjusted to be a second deviation scale value, and the association judgment object is acquired through the second deviation scale value. It should be noted that the first deviation scale value and the second deviation scale value are only exemplary, and the deviation scale values may be adjusted multiple times according to the requirement in the actual application process, that is, there may be a third deviation scale value to an nth deviation scale value in the actual application process.

In general, the first deviation scale value is the minimum adjustment unit, and when the adjustment is performed a plurality of times, each deviation scale value is larger than the previous deviation scale value, and each deviation scale value is an integer multiple of the minimum adjustment unit (first deviation scale value). That is, the second deviation scale value is greater than the first deviation scale value, the second deviation scale value being an integer multiple of the first deviation scale. Similarly, the nth deviation scale value is larger than the nth-1 deviation scale value, and the nth deviation scale value and the nth-1 deviation scale value are integer multiples of the first deviation scale value. Specifically, it may be 1-fold, 2-fold, 3-fold, or the like. It can be understood that, in this embodiment, when the multiple of the setting adjustment is smaller, the acquired association judgment object is more accurate, and the speed of acquiring the association judgment object is faster when the multiple of the setting adjustment is larger.

The following is still an exemplary illustration of the relevant parameter values in the contractual business.

Specific to node service time standard value: taking province company data with the same or positive and negative deviation 1 adjustment scale as the association judgment basis (if department data is the department data, the values of other departments conforming to the data conditions can be taken, and the data values of the province company or other province companies can be taken as the comparison); if the value obtained by the preliminary analysis judgment of the node service time standard is lower deviation, the scale is positively valued, if no data exists in the adjustment scale of the deviation 1, the node service time standard is continuously added with 1 scale to search the data, and if no data exists, the key judgment calculation is not processed; if the value obtained by preliminary analysis is the upper deviation, the scale takes the value negatively, if no data exists in the deviation 1 adjustment scale, the data is continuously searched by 1 scale downwards, and if no data exists, the key judgment calculation is not processed; specifically to a process threshold: and taking the province company data with the same or positive and negative deviation 1 adjustment scale as the association judgment basis.

To the user traffic processing timeliness portrayal threshold: and taking the province company data with the same or positive and negative deviation 1 adjustment scale as the association judgment basis. If the value obtained through preliminary analysis is lower deviation, the scale is positively valued, if no data exists in the deviation 1 adjustment scale, 1 scale is continuously added to search data, and if no data exists, key judgment calculation is not processed; if the value obtained by preliminary analysis is the upper deviation, the scale takes the value negatively, if no data exists in the deviation 1 adjustment scale, the data is searched by continuing to decrease 1 scale downwards, and if no data exists, the key judgment calculation is not processed.

Specifically to a process threshold: taking province company data with the same or positive and negative deviation 1 adjustment scale as a correlation judgment basis; if the value obtained through preliminary analysis judgment is lower deviation, the scale is negatively valued, if no data exists in 1 adjustment scale of the deviation, the data is continuously searched by 1 scale downwards, and if no data exists, the key judgment calculation is not processed; if the value obtained by preliminary analysis is judged to be the upper deviation, the scale is positively valued, if no data exists in the deviation 1 adjustment scale, 1 scale is continuously added to search data, and if no data exists, the key judgment calculation is not processed; specifically to a process threshold: and taking the province company data with the same or positive and negative deviation 1 adjustment scale as the association judgment basis.

For the adjustment scale, the initial value of the three types of values after adjustment minus the initial value before adjustment is an integer multiple of the minimum adjustment unit (also called scale), wherein the minimum adjustment or adjustment unit is used for trial calculation and adjustment of the node time standard or flow critical early warning threshold or the business processing timeliness user image threshold. The calculation rule is the same as the rule in step S201, and will not be described here.

The correlation judgment results in positive conclusion characterized in that: 1. the value (single or multiple value) obtained by the association judgment is in the upper limit or the lower limit of the super-target value combination critical threshold value compared with the target value combination critical threshold value; 2. if the values are a plurality of values, more than 80% of the correlation judgment results in that the monitoring values are all at the lower limit or the upper limit of the target value combination critical threshold; 3. the upper limit or the lower limit of the combined critical threshold of the predicted value and the target value is on the same side as the upper limit or the lower limit obtained by the fact judgment. And if the 3 characteristics are met, the association judgment is passed.

S304: and carrying out threshold critical judgment on each configuration characteristic value corresponding to the service data by utilizing threshold critical judgment based on the service data to obtain a threshold critical judgment result.

In the case where the above-described fact judgment, prediction judgment, and association judgment pass, threshold critical judgment may be further performed so as to obtain the direction of parameter value adjustment.

Specifically, the threshold critical judgment is to determine the direction to be adjusted and the adjustment direction according to the adjustment scale and combining the preamble, and perform simulation trial calculation, wherein the simulation trial calculation mode is also based on the fact judgment, the prediction judgment, the comparison data judgment and the threshold critical judgment.

For the node time standard, the actual standard of the current execution of the data is adopted, according to the adjustment direction obtained by data comparison and preliminary analysis, the suspected standard is loosely calculated in a simulation mode according to-1 scales of the node time standard of the current execution. The node time standard which is executed currently is adjusted and reduced according to the preset deviation scale value through simulation trial calculation to obtain the node time standard corresponding to the new standard, the fact judgment calculation and analysis, the prediction judgment calculation and analysis and the association judgment calculation and analysis are carried out item by using data which meet the new standard under the new standard which is supposed after the adjustment scale, and the calculation logic is the same as the fact judgment, the prediction judgment and the association judgment respectively, and can be executed according to S301-S303, and the description is omitted.

And (3) for the node time standard, the suspected standard is strictly subjected to simulation trial calculation according to the +1 scale of the currently executed node time standard. The node time standard corresponding to the new standard is obtained by adjusting and increasing the currently executed node time standard according to the preset deviation scale value through simulation trial calculation, the fact judgment calculation and analysis, the prediction judgment calculation and analysis and the association judgment calculation and analysis are carried out item by using data conforming to the new standard in the data range and the data time length under the new standard which is supposed after the adjustment and increase scale, and the calculation logic is respectively the same as the fact judgment, the prediction judgment and the association judgment, and can be executed by referring to S301-S303, and the description is omitted.

And (3) for the flow early warning threshold and the user business processing timeliness portrait threshold, threshold critical judgment is not carried out, and association judgment is carried out.

The threshold critical judgment positive conclusion feature value is as follows: the trial calculation value is calculated in a mode of adding or subtracting 1 time of the scale value; the condition that the trial calculation value is judged to be unable to continue is that the conclusion obtained by the last calculation is unable to be adjusted according to the accurate judgment method, the conclusion obtained by the previous calculation can be adjusted according to the accurate judgment method, and the trial calculation suggestion adjustment value takes the previous value of the last trial calculation.

S305: and obtaining a configuration characteristic value validity judgment result by utilizing probability superposition based on the fact judgment result, the prediction judgment result, the association judgment result and the threshold critical judgment result.

The accuracy judgment is realized by fact judgment, prediction judgment, association judgment and threshold critical judgment, and the two-eight principle is combined with different judgment, and a majority passing method is adopted, so that the probability superposition method is adopted, and the accuracy of the judgment is greatly improved. Fig. 7 is a schematic diagram of probability stacking provided in the embodiment of the present application, and in combination with the schematic diagram shown in fig. 7, the judgment accuracy is greatly improved through different random combinations of judgment. The bottom circle probability is 100%, the probability X is calculated according to the dimension 1 for the first time, the probability X1 is calculated according to the dimension 1 for the second time, the probability X2 is calculated according to the dimension in the third time, the probability X3 is calculated according to the dimension in the fourth time, and the probabilities X, X1, X2 and X3 are all smaller than 100%. In the case that the second calculation takes into account the first factor, the third calculation takes into account the second factor, and the fourth calculation takes into account the third factor, a superposition of probabilities is generated, that is, the estimated probability value gradually approaches 100% on the premise that the dimensions are switched several times and the previous analysis dimension is considered and continued. As shown in fig. 7, three times of estimation are performed to cover areas a, B, C, and D, and the areas gradually increase, while the areas of the areas not covered (the areas become smaller as the number of times of measurement increases in the gap portion existing at a distance of 100%).

The first step, the second step, the third step and the fourth step respectively borrow and refer to the results of the previous step, and the principle of probability superposition is applied to accurately judge the standard to be regulated, wherein the first step of accurate judgment adopts 3 characteristic values of the whole, the month and the statistics data month which are more than 80% of the same side to carry out probability superposition judgment, the predicted value of the second step and 2 characteristic values of the predicted value data and the statistics data which are more than 80% of the same side are superposed, the third step of association judgment 3 characteristic values, the fourth step of threshold critical judgment and the like are further superposed, and the judgment accuracy is increased.

The above provides some specific implementations of the parameter adjustment method for the embodiments of the present application, and based on this, the present application further provides a corresponding apparatus. The apparatus provided in the embodiments of the present application will be described from the viewpoint of functional modularization.

Fig. 8 is a schematic structural diagram of a parameter adjusting device according to an embodiment of the present application. Referring to the schematic structural diagram of the parameter adjustment apparatus 400 shown in fig. 8, the apparatus 400 includes an acquisition module 410, a pre-analysis module 420, a configuration feature value validity determination module 430, a trial calculation module 440, and a prompt module 450.

An obtaining module 410, configured to obtain service data in a current environment and each configuration feature value corresponding to the service data;

the pre-analysis module 420 is configured to pre-analyze each configuration feature value corresponding to the service data by using a preset data processing rule based on the service data to obtain a pre-analysis result; the pre-analysis result comprises whether each configuration characteristic value is in a preset range or not;

a configuration feature value validity judging module 430, configured to perform configuration feature value validity judgment on each configuration feature value corresponding to the service data by using a configuration feature value validity judging rule based on the service data when each configuration feature value in the pre-analysis result is in a preset range, so as to obtain a configuration feature value validity judging result;

the trial calculation module 440 is configured to perform trial calculation on each configuration feature value through a preset trial calculation rule based on the configuration feature value validity determination result to obtain a trial calculation result;

and the prompt module 450 is used for generating configuration feature adjustment values corresponding to the configuration feature values based on the trial calculation result and a pre-configured prompt strategy, and pushing the configuration feature adjustment values to the front end for content display.

The preset data processing rules include a data extraction rule, a data rejection rule, and a data analysis rule, and the pre-analysis module 420 specifically includes:

The data extraction unit is used for carrying out data extraction in the service data by utilizing the data extraction rule to obtain first pre-analysis data;

the data eliminating unit is used for eliminating the maximum value and the minimum value in the first pre-analysis data by utilizing the data eliminating rule to obtain second pre-analysis data;

and the data analysis unit is used for determining the parameter value standard reaching rate of the parameter value corresponding to the service data based on the second pre-analysis data, and pre-analyzing the parameter value standard reaching rate corresponding to the parameter value by utilizing the data analysis rule to obtain a pre-analysis result.

The configuration feature value validity judging rule includes a plurality of sub-judging rules, where the plurality of sub-judging rules include at least a fact judging rule, a prediction judging rule, an association judging rule and a threshold critical judging rule, and the configuration feature value validity judging module 430 specifically includes:

the fact judgment unit is used for carrying out fact judgment on each configuration characteristic value corresponding to the service data by utilizing a fact judgment rule based on the service data when the parameter value corresponding to the parameter value in the pre-analysis result is in a preset range to obtain a fact judgment result;

The prediction fact judging unit is used for predicting and judging each configuration characteristic value corresponding to the service data by utilizing a prediction judging rule based on the service data to obtain a prediction judging result;

the association judgment unit is used for carrying out association judgment on each configuration characteristic value corresponding to the service data by using an association judgment rule based on the service data to obtain an association judgment result;

the threshold critical judgment unit is used for carrying out threshold critical judgment on each configuration characteristic value corresponding to the service data by utilizing threshold critical judgment on the basis of the service data to obtain a threshold critical judgment result;

and the configuration characteristic value validity judging unit is used for obtaining a configuration characteristic value validity judging result by utilizing probability superposition based on the fact judging result, the prediction judging result, the association judging result and the threshold critical judging result.

The association judging unit is specifically configured to:

presetting deviation scale values according to the configuration characteristic values; the deviation scale value is used for acquiring an association judgment object when association judgment is carried out; and acquiring an association judgment object through the deviation scale value, and comparing each configuration characteristic value corresponding to the business data of the association judgment object with each configuration characteristic value corresponding to the business data to obtain an association judgment result.

The association judging unit is specifically configured to:

The apparatus 400 further comprises:

the pre-configuration module is used for pre-configuring each data processing logic corresponding to the preset data processing rule, the configuration characteristic value validity judgment rule and the preset calculation rule and the parameters corresponding to each data processing logic.

The embodiment of the application also provides corresponding parameter adjusting equipment and a computer storage medium, which are used for realizing the scheme provided by the embodiment of the application.

The device comprises a memory and a processor, wherein the memory is used for storing instructions or codes, and the processor is used for executing the instructions or codes to enable the device to execute the parameter adjustment method according to any embodiment of the application.

The computer storage medium stores code, and when the code is executed, a device executing the code implements the parameter adjustment method described in any embodiment of the present application.

The "first" and "second" in the names of "first", "second" (where present) and the like in the embodiments of the present application are used for name identification only, and do not represent the first and second in sequence.

From the above description of embodiments, it will be apparent to those skilled in the art that all or part of the steps of the above described example methods may be implemented in software plus general hardware platforms. Based on such understanding, the technical solutions of the present application may be embodied in the form of a software product, which may be stored in a storage medium, such as a read-only memory (ROM)/RAM, a magnetic disk, an optical disk, or the like, including several instructions for causing a computer device (which may be a personal computer, a server, or a network communication device such as a router) to perform the methods described in the embodiments or some parts of the embodiments of the present application.

In this specification, each embodiment is described in a progressive manner, and identical and similar parts of each embodiment are all referred to each other, and each embodiment mainly describes differences from other embodiments. In particular, for the device embodiments, since they are substantially similar to the method embodiments, the description is relatively simple, and reference is made to the description of the method embodiments for relevant points. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment. Those of ordinary skill in the art will understand and implement the present invention without undue burden.

The foregoing is merely exemplary embodiments of the present application and is not intended to limit the scope of the present application.

Claims

1. A method of parameter adjustment, the method comprising:

2. The method for adjusting parameters according to claim 1, wherein the preset data processing rules include a data extraction rule, a data rejection rule and a data analysis rule, and the pre-analyzing each configuration feature value corresponding to the service data by using the preset data processing rule based on the service data to obtain a pre-analysis result includes:

3. The method according to claim 1, wherein the configuration feature value validity judgment rule includes a plurality of sub-judgment rules, the plurality of sub-judgment rules including at least a fact judgment rule, a prediction judgment rule, an association judgment rule, and a threshold critical judgment rule, and when the configuration feature values in the pre-analysis result are in a preset range, performing configuration feature value validity judgment on the configuration feature values corresponding to the service data based on the service data by using the configuration feature value validity judgment rule to obtain a configuration feature value validity judgment result, including:

4. The method for adjusting parameters according to claim 3, wherein the performing association judgment on each configuration feature value corresponding to the service data by using association judgment rules based on the service data to obtain an association judgment result comprises:

5. The method according to claim 4, wherein the deviation scale value includes at least a first deviation scale value and a second deviation scale value, and the obtaining the association judgment object through the deviation scale value includes:

6. The parameter adjustment method according to claim 1, characterized in that the method further comprises:

7. The method for adjusting parameters according to claim 1, wherein after generating the configuration feature adjustment values corresponding to the configuration feature values based on the trial calculation result and the pre-configured prompting policy and pushing the configuration feature adjustment values to the front end for content display, the method further comprises:

8. A parameter adjustment device, the device comprising:

9. A parameter adjustment device, characterized in that the device comprises a memory for storing instructions or code and a processor for executing the instructions or code to cause the device to perform the parameter adjustment method of any one of claims 1-7.

10. A computer storage medium having code stored therein, which when executed, causes an apparatus executing the code to implement the parameter adjustment method of any one of claims 1-7.