CN113379497B - Order regulation method, order regulation device, computer equipment and computer readable storage medium - Google Patents

Order regulation method, order regulation device, computer equipment and computer readable storage medium Download PDF

Info

Publication number
CN113379497B
CN113379497B CN202110656684.0A CN202110656684A CN113379497B CN 113379497 B CN113379497 B CN 113379497B CN 202110656684 A CN202110656684 A CN 202110656684A CN 113379497 B CN113379497 B CN 113379497B
Authority
CN
China
Prior art keywords
abnormal
order
regulation
anomaly
target
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN202110656684.0A
Other languages
Chinese (zh)
Other versions
CN113379497A (en
Inventor
曾帆
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Koubei Shanghai Information Technology Co Ltd
Original Assignee
Koubei Shanghai Information Technology Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Koubei Shanghai Information Technology Co Ltd filed Critical Koubei Shanghai Information Technology Co Ltd
Priority to CN202110656684.0A priority Critical patent/CN113379497B/en
Publication of CN113379497A publication Critical patent/CN113379497A/en
Application granted granted Critical
Publication of CN113379497B publication Critical patent/CN113379497B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q30/00—Commerce
    • G06Q30/06—Buying, selling or leasing transactions
    • G06Q30/0601—Electronic shopping [e-shopping]
    • G06Q30/0633—Lists, e.g. purchase orders, compilation or processing
    • G06Q30/0635—Processing of requisition or of purchase orders
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G06Q10/063—Operations research, analysis or management
    • G06Q10/0631—Resource planning, allocation, distributing or scheduling for enterprises or organisations
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G06Q10/067—Enterprise or organisation modelling

Landscapes

  • Business, Economics & Management (AREA)
  • Engineering & Computer Science (AREA)
  • Human Resources & Organizations (AREA)
  • Strategic Management (AREA)
  • Economics (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Marketing (AREA)
  • Physics & Mathematics (AREA)
  • General Business, Economics & Management (AREA)
  • General Physics & Mathematics (AREA)
  • Development Economics (AREA)
  • Theoretical Computer Science (AREA)
  • Operations Research (AREA)
  • Game Theory and Decision Science (AREA)
  • Quality & Reliability (AREA)
  • Tourism & Hospitality (AREA)
  • Educational Administration (AREA)
  • Accounting & Taxation (AREA)
  • Finance (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

The application discloses an order regulation and control method, an order regulation and control device, computer equipment and a computer readable storage medium, which relate to the technical field of Internet, and are used for realizing the prediction of abnormal conditions in a region to be regulated and controlled through a plurality of abnormal prediction models capable of predicting different types of abnormalities, further determining a related target regulation and control strategy to regulate and control the order, reducing the occurrence possibility of the abnormal conditions, realizing the advanced regulation and control of order allocation in the region, enabling the regulated and controlled content to be more attached to the current actual condition of the region, reducing the occurrence probability of false judgment events of the order regulation and control, and having better intelligence. The method comprises the following steps: acquiring a plurality of abnormal prediction models; inputting regional parameters of the region to be regulated and controlled into a plurality of abnormal prediction models to obtain an abnormal probability set of the region to be regulated and controlled; and determining a target regulation strategy according to the abnormal probability set, and regulating and controlling the current order to be distributed of the region to be regulated according to the target regulation strategy.

Description

Order regulation method, order regulation device, computer equipment and computer readable storage medium
Technical Field
The present disclosure relates to the field of internet technologies, and in particular, to an order adjustment method, an order adjustment device, a computer device, and a computer readable storage medium.
Background
With the continuous development of internet technology, in most of service industries, the requirements of users on experience are higher and higher, so that more and more services can be provided for users by the terminals, for example, the takeaway industry has become an important component of people's daily life, and people can enjoy consumption and services without going out of home. In the current takeout industry, in order to provide more convenient services for users, many online platforms providing takeout services determine factors such as the time length of taking out an order, whether logistical responsibility is canceled, and the like, which affect whether users will continue to take out single points in the online platform, and regulate the processing process of the order according to the factors.
In the related art, an online platform counts abnormal orders which are overtime, complaint or not taken in a certain city in a historical process, analyzes the abnormal orders, determines factors influencing the urban orders, intervenes in the processes of assigning, taking and the like of the urban orders according to the determined factors, and realizes the regulation and control of the orders, so that the quantity of the abnormal orders in the city is reduced.
In carrying out the present application, the applicant has found that the related art has at least the following problems:
The factors which are determined for the city and influence the order are determined according to the abnormal orders in the history process, but the factors which actually influence the city are changed when the factors are determined along with the development of the city, so that the risk of misjudgment exists in the regulation and control of the order, and the intelligence is poor.
Disclosure of Invention
In view of this, the present application provides a method, an apparatus, a computer device and a computer readable storage medium for regulating orders, which mainly aims to solve the problems of misjudgment risk and poor intelligence in regulating orders at present.
According to a first aspect of the present application, there is provided an order regulating method, the method comprising:
acquiring a plurality of abnormality prediction models, wherein the plurality of abnormality prediction models are obtained by training order features of a plurality of abnormal orders of different abnormality types;
inputting regional parameters of a region to be regulated and controlled into the plurality of abnormal prediction models to obtain an abnormal probability set of the region to be regulated and controlled, wherein the abnormal probability set comprises a plurality of abnormal probabilities which are predicted by the plurality of abnormal prediction models based on the regional parameters;
and determining a target regulation strategy according to the abnormal probability set, and regulating and controlling the current order to be distributed of the region to be regulated according to the target regulation strategy.
Optionally, the acquiring a plurality of anomaly prediction models includes:
determining a historical time period, and counting a plurality of abnormal orders occurring in the historical time period;
reading the abnormal type of each abnormal order in the plurality of abnormal orders, and dividing the orders with the same abnormal type into the same order group to obtain a plurality of abnormal order groups;
for each abnormal order group in the abnormal order groups, extracting sample order features of all abnormal orders included in the abnormal order groups to obtain a plurality of sample order features, wherein the sample order features comprise one or more of order occurrence area feature parameters, order distribution feature parameters and order information;
training the plurality of sample order features according to the occurrence probability of each sample order feature in the abnormal order group, and constructing an abnormal prediction model of the abnormal order group;
and respectively carrying out sample order feature extraction and sample order feature training on each abnormal order group to obtain a plurality of abnormal prediction models of the plurality of abnormal order groups.
Optionally, the inputting the regional parameters of the region to be regulated to the plurality of anomaly prediction models to obtain the anomaly probability set of the region to be regulated includes:
Inputting the regional parameters to the anomaly prediction model for each anomaly prediction model in the plurality of anomaly prediction models, wherein the regional parameters at least comprise one or more of distribution resource supply characteristics, regional geographic characteristics and historical order characteristics of the region to be regulated;
determining at least one target sample order feature matched with the regional parameters based on the anomaly prediction model, counting occurrence probability corresponding to the at least one target sample order feature in the anomaly prediction model, and outputting anomaly probability, wherein the anomaly probability indicates the probability of occurrence of an anomaly type order corresponding to the anomaly prediction model in the region to be regulated;
respectively inputting the regional parameters into each of the plurality of anomaly prediction models to obtain the plurality of anomaly probabilities output by the plurality of anomaly prediction models;
the plurality of anomaly probabilities are taken as the anomaly probability set.
Optionally, the determining a target regulation strategy according to the abnormal probability set includes:
acquiring a preset probability threshold, comparing the plurality of abnormal probabilities with the preset probability threshold respectively, and outputting a plurality of comparison results;
Dividing the plurality of abnormal probabilities into a plurality of abnormal grades according to the plurality of comparison results;
extracting a highest abnormal level and a lowest abnormal level from the plurality of abnormal levels, and determining a level difference between the highest abnormal level and the lowest abnormal level;
acquiring a preset regulation and control standard, analyzing the grade gap based on the preset regulation and control standard, and outputting an analysis result;
and determining the target regulation strategy according to the analysis result.
Optionally, the analyzing the level difference based on the preset regulation and control standard, and outputting an analysis result includes:
if the preset regulation and control standard indicates that analysis is carried out according to a gap threshold, comparing the grade gap with the gap threshold, and outputting the analysis result for indicating the size relationship between the grade gap and the gap threshold;
if the preset regulation and control standard indicates that the regulation and control quantity is analyzed, a plurality of candidate grade gaps of a plurality of candidate regulation and control areas are obtained, the candidate grade gaps and the grade gaps are ranked from large to small to obtain a ranking result, the designated grade gaps of the regulation and control quantity are extracted at the head of the ranking result, and the analysis result for indicating whether the designated grade gaps of the regulation and control quantity comprise the grade gaps is output.
Optionally, the determining the target regulation strategy according to the analysis result includes:
when the analysis result indicates that the grade gap is smaller than or equal to a gap threshold or the designated grade gap of the regulation quantity does not comprise the grade gap, determining a first preset regulation strategy associated with the analysis result as the target regulation strategy;
when the analysis result indicates that the grade gap is larger than the gap threshold or the designated grade gap of the regulation quantity comprises the grade gap, determining a plurality of second preset regulation strategies related to the analysis result, inquiring a plurality of strategy information corresponding to the second preset regulation strategies, determining target strategy information matched with the regional parameters of the region to be regulated in the plurality of strategy information, and taking the second preset regulation strategies corresponding to the target strategy information as the target regulation strategies.
Optionally, the adjusting the order generated in the to-be-adjusted region according to the target adjusting strategy includes:
when the target regulation strategy indicates regulation and control of the allocation process of the order, pushing the order to be allocated to a plurality of secondary distribution resources and/or pushing the order to be allocated to a plurality of preset distribution resources in parallel;
And when the target regulation strategy indicates to regulate the distribution reaching resource quantity of the order, acquiring the subsidy resource quantity, and superposing the subsidy resource quantity to the rated distribution reaching resource quantity of the order to be distributed.
According to a second aspect of the present application, there is provided an order regulating device, comprising:
the acquisition module is used for acquiring a plurality of abnormal prediction models, wherein the plurality of abnormal prediction models are obtained by training order features of abnormal orders of different abnormal types;
the input module is used for inputting regional parameters of a region to be regulated to the plurality of abnormal prediction models to obtain an abnormal probability set of the region to be regulated, wherein the abnormal probability set comprises a plurality of abnormal probabilities which are predicted by the plurality of abnormal prediction models based on the regional parameters;
and the regulation and control module is used for determining a target regulation and control strategy according to the abnormal probability set and regulating and controlling the current order to be distributed of the region to be regulated and controlled according to the target regulation and control strategy.
Optionally, the acquiring module is configured to determine a historical time period, and count a plurality of abnormal orders occurring in the historical time period; reading the abnormal type of each abnormal order in the plurality of abnormal orders, and dividing the orders with the same abnormal type into the same order group to obtain a plurality of abnormal order groups; for each abnormal order group in the abnormal order groups, extracting sample order features of all abnormal orders included in the abnormal order groups to obtain a plurality of sample order features, wherein the sample order features comprise one or more of order occurrence area feature parameters, order distribution feature parameters and order information; training the plurality of sample order features according to the occurrence probability of each sample order feature in the abnormal order group, and constructing an abnormal prediction model of the abnormal order group; and respectively carrying out sample order feature extraction and sample order feature training on each abnormal order group to obtain a plurality of abnormal prediction models of the plurality of abnormal order groups.
Optionally, the input module is configured to input, for each anomaly prediction model in the plurality of anomaly prediction models, the regional parameter to the anomaly prediction model, where the regional parameter includes at least one or more of a distribution resource supply feature, a regional geographic feature, and a historical order feature of the region to be regulated; determining at least one target sample order feature matched with the regional parameters based on the anomaly prediction model, counting occurrence probability corresponding to the at least one target sample order feature in the anomaly prediction model, and outputting anomaly probability, wherein the anomaly probability indicates the probability of occurrence of an anomaly type order corresponding to the anomaly prediction model in the region to be regulated; respectively inputting the regional parameters into each of the plurality of anomaly prediction models to obtain the plurality of anomaly probabilities output by the plurality of anomaly prediction models; the plurality of anomaly probabilities are taken as the anomaly probability set.
Optionally, the regulation module is configured to obtain a preset probability threshold, compare the multiple abnormal probabilities with the preset probability threshold respectively, and output multiple comparison results; dividing the plurality of abnormal probabilities into a plurality of abnormal grades according to the plurality of comparison results; extracting a highest abnormal level and a lowest abnormal level from the plurality of abnormal levels, and determining a level difference between the highest abnormal level and the lowest abnormal level; acquiring a preset regulation and control standard, analyzing the grade gap based on the preset regulation and control standard, and outputting an analysis result; and determining the target regulation strategy according to the analysis result.
Optionally, the regulation and control module is configured to compare the level difference with a gap threshold if the preset regulation and control standard indicates that analysis is performed according to the gap threshold, and output the analysis result for indicating the magnitude relation between the level difference and the gap threshold; if the preset regulation and control standard indicates that the regulation and control quantity is analyzed, a plurality of candidate grade gaps of a plurality of candidate regulation and control areas are obtained, the candidate grade gaps and the grade gaps are ranked from large to small to obtain a ranking result, the designated grade gaps of the regulation and control quantity are extracted at the head of the ranking result, and the analysis result for indicating whether the designated grade gaps of the regulation and control quantity comprise the grade gaps is output.
Optionally, the regulation and control module is configured to determine, when the analysis result indicates that the level difference is less than or equal to a difference threshold or a specified level difference of a regulation and control number does not include the level difference, a first preset regulation and control policy associated with the analysis result as the target regulation and control policy; when the analysis result indicates that the grade gap is larger than the gap threshold or the designated grade gap of the regulation quantity comprises the grade gap, determining a plurality of second preset regulation strategies related to the analysis result, inquiring a plurality of strategy information corresponding to the second preset regulation strategies, determining target strategy information matched with the regional parameters of the region to be regulated in the plurality of strategy information, and taking the second preset regulation strategies corresponding to the target strategy information as the target regulation strategies.
Optionally, the adjusting and controlling module is configured to push the to-be-allocated order to a plurality of secondary distribution resources and/or push the to-be-allocated order to a plurality of preset distribution resources in parallel when the target adjusting and controlling policy indicates to adjust and control the allocation process of the order; and when the target regulation strategy indicates to regulate the distribution reaching resource quantity of the order, acquiring the subsidy resource quantity, and superposing the subsidy resource quantity to the rated distribution reaching resource quantity of the order to be distributed.
According to a third aspect of the present application there is provided a computer device comprising a memory storing a computer program and a processor implementing the steps of the method of any of the first aspects described above when the computer program is executed by the processor.
According to a fourth aspect of the present application there is provided a computer readable storage medium having stored thereon a computer program which when executed by a processor performs the steps of the method of any of the first aspects described above.
By means of the technical scheme, the order regulation method, the order regulation device, the computer equipment and the computer readable storage medium are used for obtaining the plurality of abnormal prediction models through training order features of a plurality of abnormal orders of different abnormal types, inputting regional parameters of a region to be regulated into the plurality of abnormal prediction models to obtain an abnormal probability set comprising a plurality of abnormal probabilities, determining a target regulation strategy according to the abnormal probability set, regulating and controlling the current generated order to be distributed of the region to be regulated according to the target regulation strategy, and realizing the prediction of abnormal conditions in the region to be regulated through the plurality of abnormal prediction models capable of predicting different types, further determining the related target regulation strategy to regulate and control the order, reducing the occurrence probability of abnormal conditions, realizing the early regulation and control of order distribution in the region, enabling regulated contents to be more fit with the current actual conditions of the region, reducing the occurrence probability of false judgment events of order regulation and control, and being good in intelligence.
The foregoing description is only an overview of the technical solutions of the present application, and may be implemented according to the content of the specification in order to make the technical means of the present application more clearly understood, and in order to make the above-mentioned and other objects, features and advantages of the present application more clearly understood, the following detailed description of the present application will be given.
Drawings
Various other advantages and benefits will become apparent to those of ordinary skill in the art upon reading the following detailed description of the preferred embodiments. The drawings are only for purposes of illustrating the preferred embodiments and are not to be construed as limiting the application. Also, like reference numerals are used to designate like parts throughout the figures. In the drawings:
fig. 1 shows a schematic flow chart of an order regulating method according to an embodiment of the present application;
fig. 2 shows a schematic flow chart of an order regulating method according to an embodiment of the present application;
fig. 3 is a schematic structural diagram of an order regulating device according to an embodiment of the present application;
fig. 4 shows a schematic device structure of a computer device according to an embodiment of the present application.
Detailed Description
Exemplary embodiments of the present application will be described in more detail below with reference to the accompanying drawings. While exemplary embodiments of the present application are shown in the drawings, it should be understood that the present application may be embodied in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.
The embodiment of the application provides an order regulation method, as shown in fig. 1, which comprises the following steps:
101. and acquiring a plurality of abnormality prediction models, wherein the plurality of abnormality prediction models are obtained by training order features of a plurality of abnormal orders of different abnormality types.
102. The regional parameters of the region to be regulated and controlled are input into a plurality of abnormal prediction models to obtain an abnormal probability set of the region to be regulated and controlled, wherein the abnormal probability set comprises a plurality of abnormal probabilities, and the abnormal probabilities are obtained by the plurality of abnormal prediction models based on regional parameter prediction.
103. And determining a target regulation strategy according to the abnormal probability set, and regulating and controlling the current order to be distributed of the region to be regulated according to the target regulation strategy.
According to the method provided by the embodiment of the application, the order characteristics of the abnormal orders of different abnormal types are trained to obtain a plurality of abnormal prediction models, the regional parameters of the region to be regulated are input into the plurality of abnormal prediction models to obtain the abnormal probability set comprising a plurality of abnormal probabilities, the target regulation strategy is determined according to the abnormal probability set, the order to be distributed, which is currently generated in the region to be regulated, is regulated according to the target regulation strategy, the abnormal conditions in the region to be regulated are predicted through the plurality of abnormal prediction models capable of predicting the different types, the related target regulation strategy is further determined to regulate the order, the occurrence possibility of the abnormal conditions is reduced, the early regulation of order distribution in the region is realized, the regulated content is more attached to the current actual condition of the region, the occurrence probability of an order regulation misjudgment event is reduced, and the intelligence is better.
The embodiment of the application provides an order regulation method, as shown in fig. 2, which comprises the following steps:
201. a plurality of anomaly prediction models are obtained.
In recent years, internet technology is continuously developed, more and more users start shopping based on an online platform, and whether the possibility of precipitation exists is an important factor affecting the stability of the online platform. However, the reasons influencing the user precipitation are often complex and difficult to quantify, such as factors like the order taking time length and whether the order is canceled, so that many online platforms can count overtime, complaint or unreceived abnormal orders of cities or areas in the history process by taking the cities or areas as units, analyze the abnormal orders, determine factors influencing the orders of the cities or areas, intervene in the subsequent order-generating assignment, taking and other processes in the cities or areas according to the determined factors, and realize the regulation and control of the orders, thereby reducing the quantity of the abnormal orders in the cities or areas. However, the applicant realizes that the intervention process is actually a judgment after the abnormality has occurred, that is, the abnormal event has occurred, the determined intervention means is determined based on the past abnormal event, and it is very likely that as the city or the area develops, when the order is regulated based on the intervention means, factors actually affecting the city or the area have changed, so that there is a risk of misjudgment on regulation of the order, the intelligence is poor, and the purpose of truly satisfying the user is difficult.
Therefore, the method for regulating and controlling the order is provided, a plurality of abnormal prediction models are obtained by training order features of abnormal orders of different types, regional parameters of a region to be regulated and controlled are input into the plurality of abnormal prediction models, an abnormal probability set comprising a plurality of abnormal probabilities is obtained, a target regulation and control strategy is determined according to the abnormal probability set, the current order to be distributed of the region to be regulated and controlled is regulated and controlled according to the target regulation and control strategy, abnormal conditions in the region to be regulated and controlled are predicted through the plurality of abnormal prediction models capable of predicting different types, further, the related target regulation and control strategy is determined to regulate and control the order, the occurrence possibility of the abnormal conditions is reduced, the early regulation and control of order distribution in the region is realized, the regulated and controlled content is more fit with the current actual condition of the region, the occurrence probability of an order regulation and control misjudgment event is reduced, and the intelligence is better.
In order to implement the order regulation and control method in the embodiment of the present application, a plurality of abnormality prediction models are obtained by training order features of a plurality of abnormal orders of different abnormality types in advance, and then regional parameters of a region to be regulated and controlled are predicted according to the plurality of abnormality prediction models, so as to determine the probability of occurrence of abnormal orders of corresponding abnormality types in the region to be regulated and controlled. The specific process of obtaining a plurality of anomaly prediction models is as follows:
Firstly, determining a historical time period, and counting a plurality of abnormal orders occurring in the historical time period, wherein the historical time period can be 15 days in the past, 30 days in the past, and the like, and the abnormal orders are orders which cause bad user experience due to abnormal actions of insufficient distribution resources or distribution resources in the process of ordering or distribution, for example, orders which are overtime in distribution, not picked up for a long time, complained and the like can be used as abnormal orders. Further, the abnormal orders may be orders occurring in different areas or cities, and are not limited to orders occurring in areas to be regulated.
And then, reading the abnormal type of each abnormal order in the abnormal orders, and dividing the orders with the consistent abnormal types into the same order group to obtain a plurality of abnormal order groups. In practice, the process of dividing orders according to the abnormal type, that is, dividing the bad experience type caused by the abnormal order to the user, for example, dividing the abnormal order with overtime delivery into the same order group, dividing the abnormal order with complaint into the same order group, and dividing the abnormal order which is not taken for a long time into the same order group.
Then, an abnormality prediction model is built for each abnormal order group, and a plurality of abnormality prediction models are obtained. Specifically, when the abnormality prediction model is constructed, for each of a plurality of abnormality order groups, sample order features of all abnormality orders included in the abnormality order group are extracted, and a plurality of sample order features are obtained. The sample order feature may include one or more of an order generation area feature parameter, an order distribution feature parameter, and order information. The characteristic parameters of the order generation area can be the distribution position of the order or the configuration quantity of a rider of a city or a district where a merchant position is located, weather information, the topography characteristics of the merchant position, the geographic information of the city or the district, the supply characteristic information of the city or the district and the like; the order delivery characteristic parameter may be the time of pick-up of the order, the information of the delivery personnel delivering the order, etc.; the order information may be basic information of the order such as the delivery location of the order, merchant location, amount of the order, etc. After the plurality of sample order features are extracted, training the plurality of sample order features according to the occurrence probability of each sample order feature in the abnormal order group in the plurality of sample order features, and constructing an abnormal prediction model of the abnormal order group. The model training process for a certain group of abnormal order groups is completed through the process, the abnormal prediction model of the abnormal order groups is obtained, the training process is respectively executed for each abnormal order group, sample order feature extraction and sample order feature training are carried out, a plurality of abnormal prediction models of a plurality of abnormal order groups can be obtained, the probability of abnormal orders of corresponding abnormal types in a certain area is predicted based on the abnormal prediction models, and allocation and regulation of the orders are carried out according to the probability.
It should be noted that, the process of obtaining the plurality of abnormal prediction models described above is a process of training the plurality of abnormal prediction models, and in the process of practical application, the server carried by the on-line platform may construct the plurality of abnormal prediction models in advance, store the plurality of abnormal prediction models, and directly obtain the plurality of stored abnormal prediction models for use when the region to be regulated needs to be predicted and the order is regulated, without instant training, so as to improve the regulation efficiency of the order. Furthermore, a training period can be set in the server, the process of training the abnormal prediction model is executed every training period, a new abnormal prediction model is trained, and the new abnormal prediction model is used for covering the previous abnormal prediction model, so that the follow-up prediction result of the region to be regulated is more attached to the current actual condition of the region to be regulated.
202. And inputting the regional parameters of the region to be regulated and controlled into a plurality of abnormal prediction models to obtain an abnormal probability set of the region to be regulated and controlled.
In the embodiment of the application, after the plurality of abnormality prediction models are obtained, the probability of occurrence of an abnormal order of a corresponding abnormality type in the region to be regulated can be predicted based on the plurality of abnormality prediction models, so that the influence degree of each abnormality type on the region to be regulated can be determined through the probability later, and the order is regulated through a proper regulation strategy, so that the aim of reducing the risk of occurrence of an abnormal event in the region to be regulated is fulfilled. The area to be regulated and controlled can be a city, or a business district, a district in a city, or a fence area with a certain preset range, and the range of the area to be regulated and controlled is not particularly limited in the application. Furthermore, since the plurality of anomaly prediction models are obtained, actually predicting the region to be regulated may obtain a plurality of anomaly probabilities, that is, an anomaly probability set, a process of generating the anomaly probability set is described below:
For each of the plurality of anomaly prediction models, first, a regional parameter is input to the anomaly prediction model, the regional parameter including at least one or more of a distribution resource supply characteristic, a regional geographic characteristic, and a historical order characteristic of the region to be regulated. Wherein the distribution resource supply characteristic can be the total amount of distribution resources with terminal distribution capability, historical distribution evaluation and the like provided by the area to be regulated, and the distribution resources can be distribution staff, intelligent distribution robots, intelligent distribution vehicles and the like; the regional geographic features may be weather features, geographic location features, etc. of the region to be regulated; the historical order characteristics may be the number of historical orders for the area to be conditioned, the demand for the historical orders, the pick-up of the historical orders, the basic information of the historical orders, and so on.
And then, determining at least one target sample order feature matched with the regional parameters based on the abnormality prediction model, counting the occurrence probability of the at least one target sample order feature corresponding to the abnormality prediction model, and outputting the abnormality probability. The abnormal probability indicates the probability of occurrence of an abnormal type order corresponding to the abnormal prediction model in the region to be regulated. When the occurrence probability of at least one target sample order feature corresponding to the abnormality prediction model is counted, a weight algorithm may be adopted, the occurrence probability of at least one target sample order feature is calculated according to the feature weight corresponding to each target sample order feature, so as to obtain the abnormality probability, or the occurrence probability of at least one target sample order feature may be obtained by calculating the average value of the occurrence probabilities of at least one target sample order feature, or the occurrence probability of at least one target sample order feature may be aggregated based on an aggregation algorithm, so as to output the abnormality probability, which is not particularly limited in the application.
By repeatedly executing the generation process of the abnormal probabilities, the region parameters are respectively input into each of the plurality of abnormal prediction models, so that a plurality of abnormal probabilities output by the plurality of abnormal prediction models can be obtained, and the plurality of abnormal probabilities are used as an abnormal probability set.
203. And determining a target regulation strategy according to the abnormal probability set.
In the embodiment of the application, the occurrence probability of a plurality of abnormal orders of different abnormal types is predicted for the region to be regulated, and after the abnormal probabilities are obtained, the influence of each abnormal type on the region to be regulated is different, for example, the severe degree of a timeout event is lower, and the influence on the region to be regulated is not great; however, frequent complaints are severe, and the influence on the area to be regulated is great. Therefore, actually, according to a plurality of abnormal probabilities included in the abnormal probability set, an abnormal grade is determined, and according to a grade gap between the abnormal grades, a range capable of being regulated by the abnormal grade of the region to be regulated is determined, so that a corresponding strategy is adopted to regulate the order of the region to be regulated, and the process of specifically determining the abnormal grades is as follows:
firstly, a preset probability threshold is obtained, a plurality of abnormal probabilities are respectively compared with the preset probability threshold, and a plurality of comparison results are output. The preset probability value is not limited to one value, and may be more than one value, or may be in interval form, for example, 0-20%, 20-50%, etc., which is not specifically limited in this application.
Then, since the plurality of abnormal probabilities are large and small compared with the preset probability threshold, in order to facilitate distinguishing the magnitudes of the plurality of abnormal probabilities, the plurality of abnormal probabilities may be divided into a plurality of abnormal grades according to a plurality of comparison results of the plurality of abnormal probabilities compared with the preset probability threshold. For example, assuming that the class corresponding to [ 0 to 20% ] is "mild severe", if an abnormal probability having a value of 10% exists among the plurality of abnormal probabilities, the abnormal probability is classified into the class of "mild severe".
204. Extracting the highest abnormal level and the lowest abnormal level from the plurality of abnormal levels, determining the level difference between the highest abnormal level and the lowest abnormal level, acquiring a preset regulation and control standard, analyzing the level difference based on the preset regulation and control standard, and outputting an analysis result.
In the embodiment of the application, since more than one area is actually waiting for order regulation, sometimes resources are limited, order regulation can be performed only on areas with large abnormal grade lifting space, and regulation strategies to be adopted by each area waiting for order regulation are different, therefore, grade gaps need to be calculated, preset regulation standards are acquired, the grade gaps are analyzed based on the preset regulation standards, analysis results are output, and whether the order regulation needs to be performed on the area to be regulated and how the order regulation needs to be performed are determined later according to the analysis results.
The preset regulation standard can be in two forms in practice, one form is to instruct to analyze according to a gap threshold, namely to determine the lifting space of an abnormal grade, and different regulation strategies are adopted for the areas with larger and smaller lifting space. Another form is to instruct to analyze according to the regulation quantity, namely, rank the differences of all the areas waiting to be regulated currently, select a certain quantity of areas arranged in front to regulate orders, and the following two forms of analysis processes of preset regulation standards are respectively described:
if the preset regulation and control standard indicates that the analysis is carried out according to the gap threshold, the grade gap is compared with the gap threshold, and an analysis result for indicating the relation between the grade gap and the gap threshold is output. For example, if the gap threshold is 5, for the gap threshold 4, an analysis result indicating that the gap threshold is smaller is output; for the gap threshold 6, an analysis result indicating that the gap threshold is larger is output.
If the preset regulation standard indicates that the regulation quantity is analyzed, a plurality of candidate grade gaps of a plurality of candidate regulation areas are obtained, the candidate grade gaps and the grade gaps are ranked from large to small to obtain a ranking result, the designated grade gaps of the regulation quantity are extracted at the head of the ranking result, and an analysis result for indicating whether the designated grade gaps of the regulation quantity comprise the grade gaps is output. For example, if the sorting result is (5, 4,3, 2), and the regulation number is 2, the specified level difference of the regulation number is (5, 4), so that when the level difference of the region to be regulated is 3, it can be determined that the specified level difference of the regulation number does not include 3, and an analysis result indicating that the level difference is not included is output; and when the level difference of the area to be regulated is 5, determining that the designated level difference of the regulation quantity comprises 5, and outputting an analysis result indicating that the level difference is included. The specific form of the preset regulation standard is not limited in the present application.
205. And determining a target regulation strategy according to the analysis result.
In the embodiment of the application, after the analysis result is determined, a target regulation strategy is determined according to the analysis result, and then order regulation is performed on the region to be regulated according to the target regulation strategy. The preset regulation standard has two forms, and the obtained analysis result has two different forms, so that the target regulation strategy needs to be determined according to different modes according to different analysis results, and the specific process is as follows:
when the analysis result indicates that the grade gap is smaller than or equal to the gap threshold or the designated grade gap of the regulation quantity does not include the grade gap, the space for indicating that the abnormal grade of the region to be regulated can be improved is smaller, or the region to be regulated is temporarily not in the consideration range of order regulation, therefore, a first preset regulation strategy associated with the analysis result is determined as a target regulation strategy, and regulation is properly performed according to the target regulation strategy. The number of the areas to be subjected to order regulation and control determined based on the analysis result is still small in practice, and most of the areas may not reach the preset regulation and control standard, so the first preset regulation and control strategy may be a regulation and control strategy commonly used for various types of areas, for example, the first preset regulation and control strategy may be to push the order to the secondary distribution resource in parallel, so that the first preset regulation and control strategy can be easily applied to the areas, and the areas which do not reach the preset regulation and control standard can be adapted to regulation and control.
When the analysis result indicates that the level difference is greater than the level difference threshold or the designated level difference of the regulation quantity includes the level difference, the space for improving the abnormal level of the to-be-regulated area is larger, or the to-be-regulated area is in the consideration range of order regulation, therefore, a plurality of second preset regulation strategies related to the analysis result need to be determined. Considering that the regional parameters of different regions to be regulated are different, if the capacity of some regions to be regulated is sufficient, and if the capacity of some regions to be regulated is insufficient, different regulation strategies need to be adopted for regulation, so that a plurality of strategy information corresponding to a plurality of second preset regulation strategies need to be queried, target strategy information matched with the regional parameters of the regions to be regulated is determined in the plurality of strategy information, and the second preset regulation strategy corresponding to the target strategy information is used as the target regulation strategy. For example, it is assumed that the regional parameters of the region a to be regulated indicate that the capacity of the region a is insufficient, the strategy information of the regulation strategy "a" in the plurality of second preset regulation strategies indicates that the region is suitable for the region with sufficient capacity, and the strategy information of the regulation strategy "b" indicates that the region with sufficient capacity is suitable for the region with sufficient capacity, so that the target regulation strategy determined for the region a is "b".
206. And regulating and controlling orders generated in the region to be regulated according to the target regulation and control strategy.
In the embodiment of the application, after the target regulation and control strategy is determined, the order generated in the region to be regulated and controlled can be regulated and controlled according to the target regulation and control strategy. The target regulation strategy indicates how to regulate each link or each parameter in the order allocation process, for example, the target regulation strategy can indicate to push an order to be allocated to preset distribution resources preset in some online platforms and push secondary distribution resources provided by some third party platforms, so that the receiving rate of the order to be allocated is improved; or may also instruct to increase the delivery fee of the order to be distributed, etc., and the specific content indicated by the target regulation policy is not limited in the application.
Thus, when the target regulation strategy indicates to regulate the allocation process of the order, the order to be allocated is pushed to a plurality of secondary distribution resources and/or is pushed to a plurality of preset distribution resources in parallel. The order to be distributed is subjected to degradation processing and pushed to secondary distribution resources provided by a third-party platform, or the orders to be distributed can be pushed to preset distribution resources provided by an online platform in parallel, so that parallel distribution of different transport capacities is realized.
When the target regulation strategy indicates to regulate the delivery touch-up resource quantity of the order, the patch resource quantity is obtained, and the patch resource quantity is superimposed to the rated delivery touch-up resource quantity of the order to be distributed, namely, the target regulation strategy indicates how much to increase the delivery cost, and the delivery cost of the order to be distributed is properly increased, so that the pick-up rate of the order to be distributed is improved, and overtime is avoided.
According to the method provided by the embodiment of the application, the order characteristics of the abnormal orders of different abnormal types are trained to obtain a plurality of abnormal prediction models, the regional parameters of the region to be regulated are input into the plurality of abnormal prediction models to obtain the abnormal probability set comprising a plurality of abnormal probabilities, the target regulation strategy is determined according to the abnormal probability set, the order to be distributed, which is currently generated in the region to be regulated, is regulated according to the target regulation strategy, the abnormal conditions in the region to be regulated are predicted through the plurality of abnormal prediction models capable of predicting the different types, the related target regulation strategy is further determined to regulate the order, the occurrence possibility of the abnormal conditions is reduced, the early regulation of order distribution in the region is realized, the regulated content is more attached to the current actual condition of the region, the occurrence probability of an order regulation misjudgment event is reduced, and the intelligence is better.
Further, as a specific implementation of the method shown in fig. 1, an embodiment of the present application provides an order regulation device, as shown in fig. 3, where the device includes: the system comprises an acquisition module 301, an input module 302 and a regulation and control module 303.
The acquiring module 301 is configured to acquire a plurality of anomaly prediction models, where the plurality of anomaly prediction models are obtained by training order features of a plurality of anomaly orders with different anomaly types;
the input module 302 is configured to input a regional parameter of a region to be regulated to the plurality of anomaly prediction models, and obtain an anomaly probability set of the region to be regulated, where the anomaly probability set includes a plurality of anomaly probabilities, and the plurality of anomaly probabilities are predicted by the plurality of anomaly prediction models based on the regional parameter;
the adjusting and controlling module 303 is configured to determine a target adjusting and controlling policy according to the abnormal probability set, and adjust and control an order to be allocated currently generated in the area to be adjusted and controlled according to the target adjusting and controlling policy.
In a specific application scenario, the obtaining module 301 is configured to determine a historical time period, and count a plurality of abnormal orders that occur in the historical time period; reading the abnormal type of each abnormal order in the plurality of abnormal orders, and dividing the orders with the same abnormal type into the same order group to obtain a plurality of abnormal order groups; for each abnormal order group in the abnormal order groups, extracting sample order features of all abnormal orders included in the abnormal order groups to obtain a plurality of sample order features, wherein the sample order features comprise one or more of order occurrence area feature parameters, order distribution feature parameters and order information; training the plurality of sample order features according to the occurrence probability of each sample order feature in the abnormal order group, and constructing an abnormal prediction model of the abnormal order group; and respectively carrying out sample order feature extraction and sample order feature training on each abnormal order group to obtain a plurality of abnormal prediction models of the plurality of abnormal order groups.
In a specific application scenario, the input module 302 is configured to input, for each of the plurality of anomaly prediction models, the regional parameter to the anomaly prediction model, where the regional parameter includes at least one or more of a distribution resource supply feature, a regional geographic feature, and a historical order feature of the region to be regulated; determining at least one target sample order feature matched with the regional parameters based on the anomaly prediction model, counting occurrence probability corresponding to the at least one target sample order feature in the anomaly prediction model, and outputting anomaly probability, wherein the anomaly probability indicates the probability of occurrence of an anomaly type order corresponding to the anomaly prediction model in the region to be regulated; respectively inputting the regional parameters into each of the plurality of anomaly prediction models to obtain the plurality of anomaly probabilities output by the plurality of anomaly prediction models; the plurality of anomaly probabilities are taken as the anomaly probability set.
In a specific application scenario, the regulation module 304 is configured to obtain a preset probability threshold, compare the plurality of abnormal probabilities with the preset probability threshold, and output a plurality of comparison results; dividing the plurality of abnormal probabilities into a plurality of abnormal grades according to the plurality of comparison results; extracting a highest abnormal level and a lowest abnormal level from the plurality of abnormal levels, and determining a level difference between the highest abnormal level and the lowest abnormal level; acquiring a preset regulation and control standard, analyzing the grade gap based on the preset regulation and control standard, and outputting an analysis result; and determining the target regulation strategy according to the analysis result.
In a specific application scenario, the regulation and control module 304 is configured to compare the level difference with a difference threshold if the preset regulation and control standard indicates that analysis is performed according to the difference threshold, and output the analysis result for indicating the magnitude relationship between the level difference and the difference threshold; if the preset regulation and control standard indicates that the regulation and control quantity is analyzed, a plurality of candidate grade gaps of a plurality of candidate regulation and control areas are obtained, the candidate grade gaps and the grade gaps are ranked from large to small to obtain a ranking result, the designated grade gaps of the regulation and control quantity are extracted at the head of the ranking result, and the analysis result for indicating whether the designated grade gaps of the regulation and control quantity comprise the grade gaps is output.
In a specific application scenario, the regulation module 304 is configured to determine, when the analysis result indicates that the level difference is less than or equal to a difference threshold or a specified level difference of a regulation number does not include the level difference, a first preset regulation policy associated with the analysis result as the target regulation policy; when the analysis result indicates that the grade gap is larger than the gap threshold or the designated grade gap of the regulation quantity comprises the grade gap, determining a plurality of second preset regulation strategies related to the analysis result, inquiring a plurality of strategy information corresponding to the second preset regulation strategies, determining target strategy information matched with the regional parameters of the region to be regulated in the plurality of strategy information, and taking the second preset regulation strategies corresponding to the target strategy information as the target regulation strategies.
In a specific application scenario, the adjusting and controlling module 304 is configured to push the order to be allocated to a plurality of secondary distribution resources and/or push the order to be allocated to a plurality of preset distribution resources in parallel when the target adjusting and controlling policy indicates to adjust and control the allocation process of the order; and when the target regulation strategy indicates to regulate the distribution reaching resource quantity of the order, acquiring the subsidy resource quantity, and superposing the subsidy resource quantity to the rated distribution reaching resource quantity of the order to be distributed.
According to the device provided by the embodiment of the application, the plurality of abnormal prediction models are obtained by training the order characteristics of the abnormal orders of the different types, the regional parameters of the region to be regulated are input into the plurality of abnormal prediction models, the abnormal probability set comprising the plurality of abnormal probabilities is obtained, the target regulation strategy is determined according to the abnormal probability set, the current order to be distributed of the region to be regulated is regulated according to the target regulation strategy, the abnormal conditions in the region to be regulated are predicted through the plurality of abnormal prediction models capable of predicting the different types, the related target regulation strategy is further determined to regulate the order, the occurrence possibility of the abnormal conditions is reduced, the early regulation of order distribution in the region is realized, the regulated content is more attached to the current actual condition of the region, the occurrence probability of an order regulation misjudgment event is reduced, and the intelligence is better.
It should be noted that, other corresponding descriptions of each functional unit related to the order regulating device provided in the embodiment of the present application may refer to corresponding descriptions in fig. 1 and fig. 2, and are not repeated herein.
In an exemplary embodiment, referring to fig. 4, there is further provided a device, which includes a communication bus, a processor, a memory, and a communication interface, and may further include an input-output interface and a display device, wherein the respective functional units may perform communication with each other through the bus. The memory stores a computer program and a processor for executing the program stored in the memory to execute the order regulating method in the above embodiment.
A computer readable storage medium having stored thereon a computer program which when executed by a processor performs the steps of the order regulating method.
From the above description of the embodiments, it will be apparent to those skilled in the art that the present application may be implemented in hardware, or may be implemented by means of software plus necessary general hardware platforms. Based on such understanding, the technical solution of the present application may be embodied in the form of a software product, which may be stored in a non-volatile storage medium (may be a CD-ROM, a U-disk, a mobile hard disk, etc.), and includes several instructions for causing a computer device (may be a personal computer, a server, or a network device, etc.) to perform the methods described in various implementation scenarios of the present application.
Those skilled in the art will appreciate that the drawings are merely schematic illustrations of one preferred implementation scenario, and that the modules or flows in the drawings are not necessarily required to practice the present application.
Those skilled in the art will appreciate that modules in an apparatus in an implementation scenario may be distributed in an apparatus in an implementation scenario according to an implementation scenario description, or that corresponding changes may be located in one or more apparatuses different from the implementation scenario. The modules of the implementation scenario may be combined into one module, or may be further split into a plurality of sub-modules.
The foregoing application serial numbers are merely for description, and do not represent advantages or disadvantages of the implementation scenario.
The foregoing disclosure is merely a few specific implementations of the present application, but the present application is not limited thereto and any variations that can be considered by a person skilled in the art shall fall within the protection scope of the present application.

Claims (16)

1. An order regulating method, comprising:
acquiring a plurality of abnormality prediction models, wherein the plurality of abnormality prediction models are obtained by training sample order features of a plurality of abnormal orders with different abnormality types; the sample order feature comprises one or more of order generation area feature parameters, order distribution feature parameters and order information; the sample order feature is an order feature extracted from all abnormal orders included in each abnormal order group in a plurality of abnormal order groups; the plurality of abnormal order groups are obtained by dividing orders with consistent abnormal types into the same order group according to the read abnormal type of each abnormal order in the plurality of abnormal orders;
Inputting regional parameters of a region to be regulated and controlled into the plurality of abnormal prediction models to obtain an abnormal probability set of the region to be regulated and controlled, wherein the abnormal probability set comprises a plurality of abnormal probabilities which are predicted by the plurality of abnormal prediction models based on the regional parameters; the abnormal probability indicates the probability of occurrence of an abnormal type order corresponding to the abnormal prediction model in the region to be regulated;
determining a target regulation strategy according to the abnormal probability set, and regulating and controlling the current to-be-allocated order generated by the to-be-regulated area according to the target regulation strategy;
the determining a target regulation strategy according to the abnormal probability set specifically comprises the following steps:
determining an anomaly level according to the plurality of anomaly probabilities included in the anomaly probability set; determining the target regulation strategy according to the grade difference between the abnormal grades;
the step of regulating the order generated in the region to be regulated according to the target regulation strategy comprises the following steps:
when the target regulation strategy indicates regulation and control of the allocation process of the order, the order to be allocated is pushed to a plurality of secondary distribution resources and/or is pushed to a plurality of preset distribution resources in parallel.
2. The method of claim 1, wherein the obtaining a plurality of anomaly prediction models comprises:
determining a historical time period, and counting a plurality of abnormal orders occurring in the historical time period;
reading the abnormal type of each abnormal order in the plurality of abnormal orders, and dividing the orders with the same abnormal type into the same order group to obtain a plurality of abnormal order groups;
for each abnormal order group in the abnormal order groups, extracting sample order features of all abnormal orders included in the abnormal order groups to obtain a plurality of sample order features;
training the plurality of sample order features according to the occurrence probability of each sample order feature in the abnormal order group, and constructing an abnormal prediction model of the abnormal order group;
and respectively carrying out sample order feature extraction and sample order feature training on each abnormal order group to obtain a plurality of abnormal prediction models of the plurality of abnormal order groups.
3. The method according to claim 1, wherein the inputting the regional parameters of the region to be regulated into the plurality of anomaly prediction models to obtain the anomaly probability set of the region to be regulated includes:
Inputting the regional parameters to the anomaly prediction model for each anomaly prediction model in the plurality of anomaly prediction models, wherein the regional parameters at least comprise one or more of distribution resource supply characteristics, regional geographic characteristics and historical order characteristics of the region to be regulated;
determining at least one target sample order feature matched with the regional parameters based on the anomaly prediction model, counting the occurrence probability corresponding to the at least one target sample order feature in the anomaly prediction model, and outputting anomaly probability;
respectively inputting the regional parameters into each of the plurality of anomaly prediction models to obtain the plurality of anomaly probabilities output by the plurality of anomaly prediction models;
the plurality of anomaly probabilities are taken as the anomaly probability set.
4. The method of claim 1, wherein determining a target regulatory strategy from the set of anomaly probabilities comprises:
acquiring a preset probability threshold, comparing the plurality of abnormal probabilities with the preset probability threshold respectively, and outputting a plurality of comparison results;
dividing the plurality of abnormal probabilities into a plurality of abnormal grades according to the plurality of comparison results;
Extracting a highest abnormal level and a lowest abnormal level from the plurality of abnormal levels, and determining a level difference between the highest abnormal level and the lowest abnormal level;
acquiring a preset regulation and control standard, analyzing the grade gap based on the preset regulation and control standard, and outputting an analysis result;
and determining the target regulation strategy according to the analysis result.
5. The method of claim 4, wherein analyzing the level gap based on the preset regulation criteria, and outputting an analysis result, comprises:
if the preset regulation and control standard indicates that analysis is carried out according to a gap threshold, comparing the grade gap with the gap threshold, and outputting the analysis result for indicating the size relationship between the grade gap and the gap threshold;
if the preset regulation and control standard indicates that the regulation and control quantity is analyzed, a plurality of candidate grade gaps of a plurality of candidate regulation and control areas are obtained, the candidate grade gaps and the grade gaps are ranked from large to small to obtain a ranking result, the designated grade gaps of the regulation and control quantity are extracted at the head of the ranking result, and the analysis result for indicating whether the designated grade gaps of the regulation and control quantity comprise the grade gaps is output.
6. The method of claim 4, wherein determining the target regulatory strategy based on the analysis results comprises:
when the analysis result indicates that the grade gap is smaller than or equal to a gap threshold or the designated grade gap of the regulation quantity does not comprise the grade gap, determining a first preset regulation strategy associated with the analysis result as the target regulation strategy;
when the analysis result indicates that the grade gap is larger than the gap threshold or the designated grade gap of the regulation quantity comprises the grade gap, determining a plurality of second preset regulation strategies related to the analysis result, inquiring a plurality of strategy information corresponding to the second preset regulation strategies, determining target strategy information matched with the regional parameters of the region to be regulated in the plurality of strategy information, and taking the second preset regulation strategies corresponding to the target strategy information as the target regulation strategies.
7. The method of claim 1, wherein the regulating the order generated in the region to be regulated according to the target regulation strategy further comprises:
and when the target regulation strategy indicates to regulate the distribution reaching resource quantity of the order, acquiring the subsidy resource quantity, and superposing the subsidy resource quantity to the rated distribution reaching resource quantity of the order to be distributed.
8. An order regulating device, comprising:
the acquisition module is used for acquiring a plurality of abnormal prediction models, wherein the plurality of abnormal prediction models are obtained by training sample order features of abnormal orders of different abnormal types; the sample order feature comprises one or more of order generation area feature parameters, order distribution feature parameters and order information; the sample order feature is extracted from the order features of all abnormal orders included in each abnormal order group in the plurality of abnormal order groups; the plurality of abnormal order groups are obtained by dividing orders with consistent abnormal types into the same order group according to the read abnormal type of each abnormal order in the plurality of abnormal orders;
the input module is used for inputting regional parameters of a region to be regulated to the plurality of abnormal prediction models to obtain an abnormal probability set of the region to be regulated, wherein the abnormal probability set comprises a plurality of abnormal probabilities which are predicted by the plurality of abnormal prediction models based on the regional parameters; the abnormal probability indicates the probability of occurrence of an abnormal type order corresponding to the abnormal prediction model in the region to be regulated;
The regulation and control module is used for determining a target regulation and control strategy according to the abnormal probability set, and regulating and controlling the current to-be-distributed order of the to-be-regulated area according to the target regulation and control strategy;
the regulation and control module is specifically used for:
determining an anomaly level according to the plurality of anomaly probabilities included in the anomaly probability set; determining the target regulation strategy according to the grade difference between the abnormal grades;
the regulation and control module is used for pushing the order to be distributed to a plurality of secondary distribution resources and/or pushing the order to be distributed to a plurality of preset distribution resources in parallel when the target regulation and control strategy indicates to regulate and control the distribution process of the order.
9. The apparatus of claim 8, wherein the acquisition module is configured to determine a historical time period, and to count a plurality of abnormal orders occurring during the historical time period; reading the abnormal type of each abnormal order in the plurality of abnormal orders, and dividing the orders with the same abnormal type into the same order group to obtain a plurality of abnormal order groups; for each abnormal order group in the abnormal order groups, extracting sample order features of all abnormal orders included in the abnormal order groups to obtain a plurality of sample order features; training the plurality of sample order features according to the occurrence probability of each sample order feature in the abnormal order group, and constructing an abnormal prediction model of the abnormal order group; and respectively carrying out sample order feature extraction and sample order feature training on each abnormal order group to obtain a plurality of abnormal prediction models of the plurality of abnormal order groups.
10. The apparatus of claim 8, wherein the input module is configured to input, for each of the plurality of anomaly prediction models, the regional parameters to the anomaly prediction model, the regional parameters including at least one or more of a distribution resource supply characteristic, a regional geographic characteristic, and a historical order characteristic of the region to be regulated; determining at least one target sample order feature matched with the regional parameters based on the anomaly prediction model, counting the occurrence probability corresponding to the at least one target sample order feature in the anomaly prediction model, and outputting anomaly probability; respectively inputting the regional parameters into each of the plurality of anomaly prediction models to obtain the plurality of anomaly probabilities output by the plurality of anomaly prediction models; the plurality of anomaly probabilities are taken as the anomaly probability set.
11. The apparatus of claim 8, wherein the regulation module is configured to obtain a preset probability threshold, compare the plurality of abnormal probabilities with the preset probability threshold, and output a plurality of comparison results; dividing the plurality of abnormal probabilities into a plurality of abnormal grades according to the plurality of comparison results; extracting a highest abnormal level and a lowest abnormal level from the plurality of abnormal levels, and determining a level difference between the highest abnormal level and the lowest abnormal level; acquiring a preset regulation and control standard, analyzing the grade gap based on the preset regulation and control standard, and outputting an analysis result; and determining the target regulation strategy according to the analysis result.
12. The apparatus of claim 11, wherein the regulation module is configured to compare the level difference with a gap threshold if the preset regulation standard indicates that the analysis is performed according to the gap threshold, and output the analysis result indicating a relationship between the level difference and the gap threshold; if the preset regulation and control standard indicates that the regulation and control quantity is analyzed, a plurality of candidate grade gaps of a plurality of candidate regulation and control areas are obtained, the candidate grade gaps and the grade gaps are ranked from large to small to obtain a ranking result, the designated grade gaps of the regulation and control quantity are extracted at the head of the ranking result, and the analysis result for indicating whether the designated grade gaps of the regulation and control quantity comprise the grade gaps is output.
13. The apparatus of claim 12, wherein the regulation module is configured to determine a first preset regulation strategy associated with the analysis result as the target regulation strategy when the analysis result indicates that the level difference is less than or equal to a level difference threshold or a specified level difference of a regulation number does not include the level difference; when the analysis result indicates that the grade gap is larger than the gap threshold or the designated grade gap of the regulation quantity comprises the grade gap, determining a plurality of second preset regulation strategies related to the analysis result, inquiring a plurality of strategy information corresponding to the second preset regulation strategies, determining target strategy information matched with the regional parameters of the region to be regulated in the plurality of strategy information, and taking the second preset regulation strategies corresponding to the target strategy information as the target regulation strategies.
14. The apparatus of claim 8, wherein the means for adjusting is configured to obtain a subsidized resource amount and to superimpose the subsidized resource amount to a nominal delivery reach resource amount of the order to be allocated when the target adjustment policy indicates adjustment of the delivery reach resource amount of the order.
15. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that the processor implements the steps of the method of any of claims 1 to 7 when the computer program is executed.
16. A computer readable storage medium, on which a computer program is stored, characterized in that the computer program, when being executed by a processor, implements the steps of the method of any of claims 1 to 7.
CN202110656684.0A 2021-06-11 2021-06-11 Order regulation method, order regulation device, computer equipment and computer readable storage medium Active CN113379497B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202110656684.0A CN113379497B (en) 2021-06-11 2021-06-11 Order regulation method, order regulation device, computer equipment and computer readable storage medium

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202110656684.0A CN113379497B (en) 2021-06-11 2021-06-11 Order regulation method, order regulation device, computer equipment and computer readable storage medium

Publications (2)

Publication Number Publication Date
CN113379497A CN113379497A (en) 2021-09-10
CN113379497B true CN113379497B (en) 2023-12-26

Family

ID=77574261

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202110656684.0A Active CN113379497B (en) 2021-06-11 2021-06-11 Order regulation method, order regulation device, computer equipment and computer readable storage medium

Country Status (1)

Country Link
CN (1) CN113379497B (en)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114708021A (en) * 2022-03-30 2022-07-05 北京有竹居网络技术有限公司 Seat allocation method and device, storage medium and electronic equipment

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US7305351B1 (en) * 2000-10-06 2007-12-04 Qimonda Ag System and method for managing risk and opportunity
CN105719112A (en) * 2015-05-22 2016-06-29 北京小度信息科技有限公司 Determination method and device for distribution abnormal state, and server
CN107092974A (en) * 2016-11-29 2017-08-25 北京小度信息科技有限公司 Dispense pressure prediction method and device
CN110378522A (en) * 2019-07-02 2019-10-25 北京三快在线科技有限公司 Method, apparatus, storage medium and the electronic equipment of prediction dispatching status information
CN111275507A (en) * 2018-12-04 2020-06-12 北京嘀嘀无限科技发展有限公司 Order abnormity identification and order risk management and control method and system
CN111459675A (en) * 2020-03-31 2020-07-28 拉扎斯网络科技（上海）有限公司 Data processing method and device, readable storage medium and electronic equipment

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US7305351B1 (en) * 2000-10-06 2007-12-04 Qimonda Ag System and method for managing risk and opportunity
CN105719112A (en) * 2015-05-22 2016-06-29 北京小度信息科技有限公司 Determination method and device for distribution abnormal state, and server
CN107092974A (en) * 2016-11-29 2017-08-25 北京小度信息科技有限公司 Dispense pressure prediction method and device
CN111275507A (en) * 2018-12-04 2020-06-12 北京嘀嘀无限科技发展有限公司 Order abnormity identification and order risk management and control method and system
CN110378522A (en) * 2019-07-02 2019-10-25 北京三快在线科技有限公司 Method, apparatus, storage medium and the electronic equipment of prediction dispatching status information
CN111459675A (en) * 2020-03-31 2020-07-28 拉扎斯网络科技（上海）有限公司 Data processing method and device, readable storage medium and electronic equipment

Also Published As

Publication number Publication date
CN113379497A (en) 2021-09-10

Similar Documents

Publication Publication Date Title
CN110988422B (en) Electricity stealing identification method and device and electronic equipment
US10664931B2 (en) Data analyzing system and method
CN109615122B (en) Distribution range generation method and device, electronic equipment and storage medium
CN109636013B (en) Distribution range generation method and device, electronic equipment and storage medium
CN110582064B (en) Short message distribution method, device, equipment and medium
CN111523748A (en) Case distribution method, device and system
CN113379497B (en) Order regulation method, order regulation device, computer equipment and computer readable storage medium
CN112730938A (en) Electricity stealing user judgment method based on electricity utilization collection big data
CN112297934B (en) Charging station processing method and device, storage medium and processor
CN114819213A (en) Intelligent operation and maintenance management method and system for electric automobile public charging facility
CN112183943A (en) Order distribution method and device and electronic equipment
CN106059816B (en) Transfer storage facility site selecting method and system
CN111340581B (en) Data processing method and device, readable storage medium and electronic equipment
CN108099641A (en) The energy control method and device of charging station
CN107977855B (en) Method and device for managing user information
CN107846614B (en) Video traffic scheduling method and device and electronic equipment
CN110070392B (en) User loss early warning method and device
CN111429190A (en) Automatic generation method and system of material purchase order, server and medium
CN116107849B (en) Data center station energy consumption management system based on artificial intelligence
CN113705710A (en) Method and device for identifying default electricity consumption of electric vehicle charging facility
CN111833595B (en) Shared automobile auxiliary vehicle configuration method, electronic device and storage medium
CN109981325B (en) Alarm threshold setting method and device, computing equipment and medium
CN111723169A (en) Map display method and device, electronic equipment and storage medium
CN117236532B (en) Load data-based electricity consumption peak load prediction method and system
CN117937548B (en) Capacity configuration method and device for novel energy storage power station, storage medium and processor

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant