CN114612029A - Risk waybill data identification method and device, computer equipment and storage medium - Google Patents
Risk waybill data identification method and device, computer equipment and storage medium Download PDFInfo
- Publication number
- CN114612029A CN114612029A CN202011423686.7A CN202011423686A CN114612029A CN 114612029 A CN114612029 A CN 114612029A CN 202011423686 A CN202011423686 A CN 202011423686A CN 114612029 A CN114612029 A CN 114612029A
- Authority
- CN
- China
- Prior art keywords
- abnormal
- waybill data
- historical
- data
- waybill
- 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.)
- Pending
Links
Images
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
- G06Q10/00—Administration; Management
- G06Q10/08—Logistics, e.g. warehousing, loading or distribution; Inventory or stock management
- G06Q10/083—Shipping
- G06Q10/0838—Historical data
-
- 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
- G06Q10/08—Logistics, e.g. warehousing, loading or distribution; Inventory or stock management
- G06Q10/083—Shipping
Landscapes
- Business, Economics & Management (AREA)
- Engineering & Computer Science (AREA)
- Economics (AREA)
- Quality & Reliability (AREA)
- Entrepreneurship & Innovation (AREA)
- Human Resources & Organizations (AREA)
- Marketing (AREA)
- Operations Research (AREA)
- Development Economics (AREA)
- Strategic Management (AREA)
- Tourism & Hospitality (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The application relates to a risk waybill data identification method, a risk waybill data identification device, computer equipment and a storage medium, wherein the method comprises the following steps: acquiring historical waybill data; determining target characteristics associated with preset abnormal operation according to the historical waybill data; determining an abnormal score corresponding to each historical waybill data and the target feature based on a local abnormal factor detection algorithm and the target feature; and determining the historical waybill data with the abnormal score larger than a preset threshold value as the risk waybill data with abnormal operation. When abnormal operation needs to be identified, the method utilizes characteristic information related to the abnormal operation, simultaneously outputs abnormal scores corresponding to historical waybill data based on a local abnormal factor detection algorithm, screens out risk waybill data possibly having abnormal operation from all historical waybill data by combining with a preset threshold value, and greatly improves the identification efficiency of the risk waybill data due to the utilization of the local abnormal factor detection algorithm.
Description
Technical Field
The present application relates to the field of logistics technologies, and in particular, to a method and an apparatus for identifying risk waybill data, a computer device, and a storage medium.
Background
Currently, in the field of logistics, a receiving and dispatching person has a chance to perform illegal false operation, and the illegal false operation of a receiving and dispatching end may bring loss to a company or bring bad experience to a customer, so that waybill data with illegal and false operation needs to be identified as far as possible by a certain method, so that the loss of the company is reduced and the user experience is improved.
In order to identify the waybill data which possibly has illegal false operations, statistics can be carried out on collecting and dispatching data of each person to be collected and dispatched, and then a threshold value is taken according to experience to judge whether the waybill has the illegal false operations. However, this approach requires significant time and is inefficient.
Disclosure of Invention
In view of the above, it is necessary to provide a risk waybill data identification method, an apparatus, a computer device and a storage medium, which can improve efficiency.
A method for identifying risk manifest data, the method comprising:
acquiring historical waybill data;
determining target characteristics associated with preset abnormal operation according to the historical waybill data;
determining an abnormal score corresponding to each historical waybill data and the target feature based on a local abnormal factor detection algorithm and the target feature;
and determining the historical waybill data with the abnormal score larger than a preset threshold value as the risk waybill data with abnormal operation.
In one embodiment, before determining that the historical waybill data with the abnormal score larger than the preset threshold is the risk waybill data with abnormal operation, the method further includes:
sequencing all the abnormal scores in a descending order to obtain an abnormal score sequencing result;
and determining a third quartile in the abnormal score sorting result as the preset threshold.
In one embodiment, after determining the historical waybill data with the abnormal score larger than the preset threshold as the risk waybill data with abnormal operation, the method further includes:
pushing the risk waybill data to a corresponding target object, and acquiring feedback information of the target object;
determining a real abnormal freight note with abnormal operation according to the feedback information;
taking more than two k values, and respectively determining a kth quartile corresponding to each k value;
respectively taking each kth quartile as a candidate threshold value, and determining a predicted abnormal freight note in each historical freight note data;
calculating a recall rate and a precision rate corresponding to each kth quartile based on the real abnormal waybill and the predicted abnormal waybill;
and determining the preset threshold according to the recall rate and the precision rate corresponding to each kth quartile.
In one embodiment, the historical waybill data includes whether each order contained in the historical waybill is cancelled; the preset abnormal operation comprises false acquisition; the target features include: the number of times of sending operation is cancelled by the same person who receives and sends the mail.
In one embodiment, the determining an abnormality score corresponding to each historical waybill data and the target feature based on a local abnormality factor detection algorithm and the target feature includes:
taking the operation times of the same receiving and dispatching person for cancelling the consignor in all the historical waybill data as a characteristic value, and constructing a local abnormal factor detection model based on a local abnormal factor detection algorithm; and acquiring an abnormal score corresponding to the target characteristic and each historical waybill data output by the local abnormal factor detection model.
In one embodiment, the historical waybill data comprises a blanket address corresponding to a historical waybill, and a place of appropriate operation of the historical waybill; the preset abnormal operation comprises non-recipient address dispatching; the target features include: the distance between the appropriate operation location and the recipient address.
In one embodiment, the determining an abnormality score corresponding to each historical waybill data and the target feature based on a local abnormality factor detection algorithm and the target feature includes:
taking the distance between the appropriate operation place and the addressee in all the historical waybill data as a characteristic value, and constructing a local abnormal factor detection model based on a local abnormal factor detection algorithm; and acquiring an abnormal score corresponding to the target characteristic and each historical waybill data output by the local abnormal factor detection model.
A risk manifest data identification apparatus, the apparatus comprising:
the acquisition module is used for acquiring historical waybill data;
the characteristic determining module is used for determining target characteristics associated with preset abnormal operation according to the historical waybill data;
an abnormal score determining module, configured to determine an abnormal score corresponding to the target feature for each piece of historical waybill data based on a local abnormal factor detection algorithm and the target feature;
and the identification module is used for determining the historical waybill data with the abnormal score larger than the preset threshold value as the risk waybill data with abnormal operation.
A computer device comprising a memory and a processor, the memory storing a computer program, the processor implementing the steps of the above-mentioned method for identifying risk waybill data when executing the computer program.
A computer-readable storage medium, on which a computer program is stored which, when being executed by a processor, carries out the steps of the above-mentioned method for identifying risk manifest data.
According to the risk waybill data identification method, the risk waybill data identification device, the computer equipment and the storage medium, after historical waybill data are obtained, target characteristics associated with preset abnormal operations are determined according to the historical waybill data, abnormal scores corresponding to the historical waybill are determined based on the target characteristics and a local abnormal factor detection algorithm, the abnormal scores are compared with a preset threshold value, and the historical waybill data larger than the preset threshold value are determined as risk waybill data. When abnormal operation needs to be identified, the method utilizes characteristic information related to the abnormal operation, simultaneously outputs abnormal scores corresponding to historical waybill data based on a local abnormal factor detection algorithm, screens out risk waybill data possibly having abnormal operation from all historical waybill data by combining with a preset threshold value, and greatly improves the identification efficiency of the risk waybill data due to the utilization of the local abnormal factor detection algorithm.
Drawings
FIG. 1 is a schematic flow chart diagram of a method for identifying risk waybill data in one embodiment;
FIG. 2 is a schematic flow chart diagram illustrating a method for identifying risk manifest data in another embodiment;
FIG. 3 is a schematic flow chart diagram illustrating a method for identifying risk manifest data in another embodiment;
FIG. 4 is a schematic flow chart diagram illustrating a method for identifying risk waybill data in an exemplary embodiment;
FIG. 5 is a block diagram of the risk waybill data identification device in one embodiment;
FIG. 6 is a block diagram showing the structure of a risk waybill data identification device in another embodiment;
FIG. 7 is a block diagram showing the structure of a risk waybill data identification device in another embodiment;
FIG. 8 is a diagram illustrating an internal structure of a computer device according to an embodiment.
Detailed Description
In order to make the objects, technical solutions and advantages of the present application more apparent, the present application is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the present application and are not intended to limit the present application.
In an embodiment, as shown in fig. 1, a method for identifying risk waybill data is provided, and this embodiment is illustrated by applying the method to a terminal, it is to be understood that the method may also be applied to a server, and may also be applied to a system including the terminal and the server, and is implemented by interaction between the terminal and the server. In this embodiment, the method includes steps S110 to S140.
And step S110, obtaining historical waybill data.
The historical waybill data refers to waybill data generated in a historical time period, and the historical time period can be set according to actual conditions, for example, the historical time period can be set to 1 month in history, half month in history, or the previous day and the like. In one embodiment, historical waybill data can be divided into data for collecting waybills and data for dispatching waybills; the receiving shipping note is a shipping note which is usually picked up by a person to be dispatched from a client of a sender and needs to be dispatched, and the dispatching shipping note is a shipping note which is usually dispatched to a client of a receiver by the person to be dispatched.
In one embodiment, the historical waybill is a blanket waybill, and the historical waybill data includes: the order number, the collection time, the collection personnel, the network sites where the collection personnel are located and whether the order is cancelled or not are included in the collected freight note. The order can be determined to be detained according to the waybill data of whether the order is cancelled or not, and the detained reason is to cancel the sending; in one embodiment, the waybill data of whether the order was canceled is obtained through the FVP routing table in the bargun operating system. The bargun is a handheld terminal device used in the logistics express industry, is provided with an operating system and a scanning engine, and is a carrier for data storage in the logistics express industry. The bayonets are used as intelligent equipment for rapid data acquisition, work services of all express links such as collecting scanning, sorting tracking, dispatching scanning and loading scanning are integrated, and intelligent work and management can be realized. All transport links of the express are monitored through bargun scanning, real-time tracking recording can be synchronously carried out on the position of the express at the background, managers and clients can conveniently inquire goods information through a system in time, and process management is achieved; meanwhile, a bar code technology is adopted, and the checking personnel complete checking work through bargun scanning; and the intelligent checking, statistics, data reporting and the like can be carried out on the commodities entering and leaving the warehouse. In one embodiment, the order number, time to collect, people to collect, and the website where the people to collect, of each order contained in the manifest can be obtained from the order-in-time response schedule.
In another embodiment, the historical waybill is a dispatch waybill, and the historical waybill data includes waybill number, time of the due operation, location of the due operation, recipient address, operator, and the like. In one embodiment, the information of the consignee address, the time point of the consignment operation, the place of the consignment operation, the consignment operator, the website where the consignment operator is located and the like is obtained from the wide list of the consignment orders, then duplication removal is carried out according to the consignment orders, the latest data corresponding to each consignment order number are reserved, and historical consignment order data of the consignment orders are determined based on the latest data.
And step S120, determining target characteristics associated with preset abnormal operation according to the historical waybill data.
The preset abnormal operation is preset abnormal operation; in one embodiment, the preset abnormal operation is associated with the type of the historical waybill; for example, the historical waybill is a blanket collecting waybill, and the preset abnormal operation comprises false blanket collecting operation, specifically comprising the step of cancelling a mail; some personnel may cancel the sending piece after the receiving piece is operated for the order, so as to improve the performance data of the personnel, and the like, and the condition of false receiving can occur. If the historical waybill is a dispatching waybill, the preset abnormal operation comprises non-recipient address dispatching, and specifically comprises that the distance between a place where the operation is properly dispatched and a recipient address exceeds a threshold value; punishment such as delivery overtime is avoided by some personnel through unauthorized operation in advance, and the situation that the distance between the operation-appropriate place and the address of the receiver exceeds a threshold value can exist.
In one embodiment, the historical waybill data includes whether orders contained in the historical waybill were cancelled; presetting abnormal operations including false acquisition operations; the target features include: the number of times of sending operation is cancelled by the same person who receives and sends the mail.
In one embodiment, determining target characteristics associated with the predetermined abnormal operation based on the historical waybill data includes: after whether the orders contained in the historical waybill are cancelled or not is obtained, clustering the orders for cancelling the consignment by the dimensionality of the consignor to obtain the times of cancelling the consignor in the historical waybill data by the same consignor. In another embodiment, determining the target characteristics associated with the predetermined abnormal operation from the historical waybill data includes: and counting the historical freight note quantity corresponding to the historical orders of the cancelled sent mails to obtain the operation times of cancelling the sent mails related to the same receiving and dispatching person. Wherein, the relation of the freight note and the order comprises: a plurality of orders may be correspondingly included in one freight order. It can be understood that the number of times of the sending operation of the same consignee is too large in a certain time, and the consignee may have false collecting operation.
In the embodiment, by using the times of canceling sending after picking up the packages, the abnormal operation of false picking up which may exist in the picking up and shipping bill can be identified.
In another embodiment, the historical waybill data comprises a blanket address corresponding to the historical waybill, and a place of appropriate operation of the historical waybill; presetting abnormal operation including non-receiving party address dispatching; the target features include: the distance between the appropriate operation location and the recipient address.
Further, in one embodiment, determining a target characteristic associated with the predetermined abnormal operation based on the historical waybill data includes: and converting the receiving address into a corresponding longitude and latitude position according to a preset address longitude and latitude conversion table, wherein the distance between the longitude and latitude position of the appropriate delivery operation place and the longitude and latitude position of the receiving address is the distance between the appropriate delivery operation place and the receiving address. The appropriate operation place is usually represented by a longitude and latitude position. Calculating the distance between two latitude and longitude location points may be accomplished in any manner.
In this embodiment, the distance between the successful delivery operation location and the receiving address is used to identify the abnormal operation of non-receiving address delivery that may exist in the delivery manifest.
In this embodiment, it is necessary to identify whether the waybill has the preset abnormal operation, and it is necessary to analyze the waybill by combining the feature data related to the preset abnormal operation, that is, the target feature related to the preset abnormal operation in this embodiment. After the historical waybill data is obtained, target characteristics associated with preset abnormal operations can be determined according to the historical waybill data.
And S130, determining abnormal scores corresponding to the historical waybill data and the target features based on the local abnormal factor detection algorithm and the target features.
The Local Outlier Factor detection algorithm, also called Local Outlier Factor detection method (LOF algorithm, Local Outlier Factor), is an unsupervised Outlier detection method, and is a relatively representative algorithm in density-based Outlier detection methods. The algorithm will calculate an outlier LOF for each point in the data set.
In one embodiment, the target characteristics of each historical waybill data and the preset abnormal operation are used as characteristic values, a corresponding local abnormal factor detection model is constructed, and then the score of the preset abnormal operation corresponding to each historical waybill data output by the local abnormal factor detection model is obtained and recorded as an abnormal score in this embodiment. In one embodiment, the larger the anomaly score output by the local anomaly factor detection algorithm is, the more abnormal the input feature data are in the detection of the preset abnormal operation, and then the risk history waybills which may have the preset abnormal operation can be screened out based on the anomaly score.
In one embodiment, determining an anomaly score corresponding to each historical waybill data and the target feature based on a local anomaly factor detection algorithm and the target feature includes: the method comprises the steps that a local abnormal factor detection model is built based on a local abnormal factor detection algorithm by taking the operation times of the same receiving and dispatching person to cancel the consignor in all historical waybill data as a characteristic value; and acquiring the abnormal scores corresponding to the target characteristics and the historical waybill data output by the local abnormal factor detection model.
It can be understood that if a local abnormal factor detection model is constructed by taking the times of cancelling sending articles after the same receiving and dispatching person receives as a characteristic, the abnormal score output by the local abnormal factor detection model to the target characteristic value of the historical waybill data can be used for representing whether the operation of cancelling sending articles belonging to the same receiving and dispatching person in the historical waybill data is abnormal or not; it will be appreciated that the greater the anomaly score, the more anomalous the cancellation of the mail is for the recipient, and the more likely there is a risk of false acquisition.
In another embodiment, determining an anomaly score corresponding to each historical waybill data and the target feature based on a local anomaly factor detection algorithm and the target feature comprises: taking the distance between the appropriate operation place and the addressee in all historical waybill data as a characteristic value, and constructing a local abnormal factor detection model based on a local abnormal factor detection algorithm; and acquiring the abnormal scores corresponding to the target characteristics and the historical waybill data output by the local abnormal factor detection model.
It can be understood that if the local abnormal factor detection model is constructed by taking the distance between the appropriate delivery operation place and the receiving address as a characteristic, the abnormal score output by the local abnormal factor detection model to the target characteristic value of the historical waybill data can be used for representing whether the delivery of each historical waybill in the historical waybill data is abnormal or not; it can be understood that if the abnormal score is larger, the more abnormal the distance between the successful delivery operation site and the receiving address of the historical waybill is, the more possible the historical waybill is to have the risk of non-receiving address delivery; whether the operation abnormity exists can be judged by setting a threshold value subsequently.
Step S140, determining the historical waybill data with the abnormal score larger than a preset threshold value as the risk waybill data with abnormal operation.
In the embodiment, a threshold value is preset, and whether the waybill data is risk waybill data with abnormal operation is judged according to the preset threshold value; in one embodiment, the preset threshold is set in advance after analysis of a large amount of historical data, for example, after a large amount of historical data is analyzed, the selected value can obviously distinguish the waybill with abnormal operation from the waybill without abnormal operation, and the value can be set as the preset threshold; further, in one embodiment, the preset threshold may be determined by means of quantiles, for example, after calculating the abnormal scores corresponding to all the historical waybill data and the preset abnormal operation, an abnormal score is determined therefrom by means of quantiles as the preset threshold. In a specific embodiment, after all the abnormal scores are sorted from small to large, the third quartile is taken as a preset threshold. Furthermore, in another embodiment, for a scheme in which the abnormal score value is taken as the preset threshold, different k values may be taken, and the kth quartile in the sorted abnormal scores is taken, the recall rate and the precision rate corresponding to each kth quartile are respectively calculated, and one kth quartile is selected as the preset threshold by combining the recall rate and the precision rate. The specific process of determining the preset threshold will be described in detail in the following embodiments, and will not be described herein again. It will be appreciated that in other embodiments, the determination of the preset threshold may be accomplished in other ways.
In this embodiment, the abnormal scores corresponding to the historical waybill data and associated with the preset abnormal operation, which are obtained through calculation, are compared with the preset threshold, the historical waybill data corresponding to the abnormal scores greater than the preset threshold are determined as the risk waybill data in which the preset abnormal operation may exist, that is, the historical waybill in which the abnormal operation may exist, the identified risk waybill data can be subsequently pushed to relevant personnel, and the relevant personnel perform secondary confirmation to identify the waybill in which the abnormal operation actually exists in the historical waybill.
According to the risk waybill data identification method, after historical waybill data are obtained, target characteristics associated with preset abnormal operations are determined according to the historical waybill data, abnormal scores corresponding to the historical waybill are determined based on the target characteristics and a local abnormal factor detection algorithm, the abnormal scores are compared with a preset threshold value, and the historical waybill data larger than the preset threshold value are determined as the risk waybill data. When abnormal operation needs to be identified, the method utilizes characteristic information related to the abnormal operation, simultaneously outputs abnormal scores corresponding to the historical waybill data based on a local abnormal factor detection algorithm, screens out risk waybill data possibly having abnormal operation from all the historical waybill data by combining a preset threshold value, and can greatly improve the identification efficiency of the risk waybill data due to the utilization of the local abnormal factor detection algorithm.
In one embodiment, as shown in fig. 2, before determining the historical waybill data with the abnormal score larger than the preset threshold as the risk waybill data with abnormal operation, the method further includes: s210, sorting the abnormal scores in a descending order to obtain an abnormal score sorting result; step S220, determining a third quartile in the abnormal score sorting result as a preset threshold.
The abnormal scores are sorted from small to large, and the result obtained by sorting is recorded as an abnormal score sorting result in this embodiment.
Quantiles (quantiles), also called quantiles, refer to numerical points that divide the probability distribution range of a random variable into several equal parts, and there are commonly used median (i.e., binary), quartile, percentile, and the like. The Quartile (Quartile) is one of the quantiles in statistics, namely all the numerical values are arranged from small to large and divided into four equal parts, and the numerical values at the positions of three dividing points are the Quartile. Quartiles typically include: 1) first quartile (Q)1) The number is also called as a smaller quartile and is equal to the 25 th% of the numbers in the sample after all the numbers are arranged from small to large; 2) second quartile (Q)2) The number is also called as the median, and is equal to the 50 th% of the numbers in the sample after all the numbers are arranged from small to large; 3) third quartile (Q)3) The number is also called as the greater quartile, and is equal to the 75% of the numbers in the sample after all the numbers are arranged from small to large. The difference between the third quartile and the first quartile is also called the quartile range.
In this embodiment, after all the abnormal scores are sorted, the abnormal score corresponding to the third quartile is determined as the preset threshold. In other embodiments, the preset threshold may be determined in other manners.
In the embodiment, after the abnormal scores corresponding to the preset abnormal operation are determined based on the target characteristics of all historical waybill data, the abnormal scores are sequenced, the abnormal score corresponding to the third quartile from small to large in the abnormal scores is determined as the preset threshold, the threshold can be determined according to actual data, and the determined preset threshold changes along with the difference of the historical waybill data, so that the selected preset threshold is more consistent with the actual situation, and the risk waybill data is more accurately detected.
Further, as shown in fig. 3, in an embodiment, after determining that the historical waybill data with the abnormal score larger than the preset threshold is the risk waybill data with abnormal operation, steps S310 to S360 are further included.
Step S310, the risk waybill data is pushed to the corresponding target object, and feedback information of the target object is received.
In one embodiment, after determining the risk waybill data in the historical waybill according to the abnormal score and a preset threshold, reading an object related to the risk waybill data, recording the object as a target object, namely an object actually related to the risk waybill data, enabling the target object to perform secondary confirmation on the risk waybill data, and performing feedback information; in one embodiment, if the target object determines that the risk waybill data does have abnormal operation, a confirmed feedback message may be given; in another embodiment, if the target object determines that the risk manifest data does not have abnormal operation, the certification information for certifying that the abnormal operation does not exist, such as communication records with the client, etc., may be fed back. Further, in one embodiment, the object may represent an account number of the person being dispatched or the like.
In one embodiment, the feedback information of the target object includes the existence of abnormal operation, the absence of abnormal operation, and the like, and further, whether abnormal operation exists in the risk waybill data can be determined according to the feedback information of the target object.
In one embodiment, the process of pushing the risk manifest data to the corresponding target object may be accomplished by an operational assistant.
And step S320, determining the real abnormal freight note with abnormal operation according to the feedback information.
Because the feedback information is fed back by the object corresponding to the risk waybill data, whether the risk waybill data really has abnormal operation or not can be determined according to the feedback information, and if the risk waybill data really has abnormal operation, the risk waybill data is marked as a real abnormal waybill.
Step S330, more than two k values are taken, and the kth quartile corresponding to each k value is respectively determined.
In this embodiment, the value of the quartile is used as the candidate threshold, so 0< k < 4. In one specific embodiment, k is 2, k is 2.5, k is 3, and k is 3.5, and the corresponding k-th quartile is determined.
And step S340, determining the predicted abnormal waybill in the historical waybill data by taking each kth quartile as a candidate threshold.
And step S350, calculating the recall rate and the precision rate corresponding to each kth quartile based on the real abnormal waybill and the predicted abnormal waybill.
The abnormal freight note prediction is historical freight note data with an abnormal score larger than the kth quartile when the kth quartile is used as a candidate threshold, is equivalent to the risk freight note data in the embodiment shown in fig. 1 in concept, and is referred to as an abnormal freight note prediction for distinguishing.
Precision and recall are concepts that utilize results to evaluate model effects; in one embodiment, calculating the recall corresponding to the kth quartile comprises: and determining a first freight note quantity of the intersection of the predicted abnormal freight notes and the real abnormal freight notes and a second freight note quantity of the real abnormal freight notes, and determining the ratio of the first freight note quantity to the second freight note quantity as the recall rate. Calculating the precision of the kth quartile comprises: and determining a first freight note quantity of the intersection of the predicted abnormal freight notes and the real abnormal freight notes and a third freight note quantity of the predicted abnormal freight notes, and determining the ratio of the first freight note quantity to the third freight note quantity as an accurate rate. And calculating each kth quartile in sequence to obtain the corresponding recall rate and the corresponding precision rate.
And S360, determining a preset threshold according to the recall rate and the precision rate corresponding to each kth quartile.
In one embodiment, the determining the preset threshold according to the recall rate and the precision rate corresponding to each kth quartile comprises: and determining the product of the recall rate and the precision rate of the kth quartile as a detection effect score, and determining the kth quartile with the highest score in the detection effect scores as a preset threshold value.
In the embodiment, feedback information of the target object for the risk waybill data is combined for analysis, and the recall rate and the precision rate are respectively calculated by utilizing more than two set k values, so that a group of k-th quartiles with the best effect is selected from the more than two k values and is determined as the preset threshold, automatic iterative updating of the threshold can be realized, and the accuracy and the recall rate of the risk waybill data are further improved.
In one embodiment, after determining that there is a real abnormal waybill with abnormal operation according to the feedback information, the method further comprises: and recording real abnormal waybill data of each target object, and determining the score of the target object based on the real abnormal waybill data. Determining the score of the target object based on the real abnormal waybill data can be carried out according to any mode.
In an embodiment, as shown in fig. 4, the method for identifying the risk waybill data includes steps S1 to S5, and in this embodiment, the historical waybill data is taken as an example of the previous day, that is, the waybill data of the previous day is identified every day:
s1: obtaining historical waybill data
1) Obtaining the appropriate operation time point, the latitude and longitude of the appropriate operation place, the work number of the appropriate personnel and the network site where the appropriate personnel are located of the delivery order of the previous day from the wide list of the delivery order; removing duplication according to the waybill number, keeping the latest data for each waybill number, and storing the result as table 1;
2) acquiring the receiving address corresponding to each waybill number, converting the receiving address into corresponding longitude and latitude according to an address transit longitude and latitude table, and storing the result as a table 2;
3) acquiring the corresponding delivery order number, the collection time, the collection dispatching staff number and the collection dispatching staff location point of each order collected in the previous day from the order timely response detail table, and storing the result as a table 3;
4) the time when the operation of each waybill is held up and the reason for holding up is to cancel the mail is obtained from the FVP routing table, which is labeled table 4.
S2: data table integration
1) Taking the table 1 as a main table, and obtaining a basic table 5 for judging whether the freight note is dispatched by a non-receiving party address according to the freight note number left connection table 2;
2) table 3 is used as the master table, and table 4 is connected to the left according to the waybill number, and the result is stored as table 6. If the operation corresponding to a certain order number in the result table is detained and the detained reason is that the time field for cancelling the sending is empty, the order is indicated to have no sending cancelling phenomenon; if the waybill number corresponding to a certain order has a plurality of operations which are detained due to the fact that the sending is cancelled, the situation that the sending is frequently cancelled for a plurality of times is indicated;
3) and (4) screening records with the operation detention and the detention reason being non-null time fields for cancelling the sending in the table 6, and obtaining a basic table 7 for judging whether the order is falsely picked or not.
S3: abnormal operation identification model (including non-receiver address dispatch identification model and false cable-collecting identification model)
Firstly, a non-receiving address dispatching identification model:
1) according to the longitude and latitude of the place of the appropriate operation and the longitude and latitude of the receiving address in the table 5, the earth surface distance of the address of the appropriate operation and the receiving address of each waybill is calculated, and the calculation sql formula on the earth surface is as follows:
1000.0 × acos (sin ((addressee longitude × 3.1415)/180) × cos ((addressee latitude × 3.1415)/180))/6380) × 6380
2) Taking the earth surface distances of the operation fail and the receiving addresses of all the dispatch orders as characteristics, constructing an LOF local abnormal factor detection model, acquiring abnormal scores of all the dispatch orders output by the LOF model, and storing the abnormal scores and the earth surface distances of the operation fail and the receiving addresses corresponding to the orders as a table 8;
3) arranging the abnormal scores from small to large, setting the total number as n, and calculating a third quartile (Q3) of the n abnormal scores:
the position of Q3 is (n +1) × 0.75, and the score at "position of Q3" among the n abnormality scores is the third quartile of the n abnormality scores (Q3);
4) corresponding freight notes with abnormal scores of earth surface distances of the operational delivery and receiving addresses larger than the third quartile are marked as risk freight notes of non-receiving address dispatch;
5) and transmitting the freight order number marked as the non-receiving address to dispatch the risk freight order, the corresponding work number of the person who is submitted for receiving and dispatching, and the website where the person who is submitted for receiving and dispatching is located to an operation assistant.
A false acquisition identification model:
1) the times of sending the article cancel after each receiving and sending person operates the article collecting on the same day in the receiving and sending person dimension aggregation statistical table 6;
2) taking the times of cancelling the sending of the receiving and dispatching personnel after receiving the receiving and dispatching personnel on the same day as characteristics, constructing an LOF local abnormal factor detection model, acquiring abnormal scores of all receiving and dispatching orders output by the LOF model, and storing the abnormal scores and the times of cancelling the sending of the receiving and dispatching personnel after the receiving and dispatching personnel operate and receive the receiving and dispatching personnel on the same day as a table 9; the method for constructing the LOF local abnormal factor detection model specifically comprises the following steps: and (4) inputting the characteristic value in the historical waybill data as an LOF model, and performing model training fitting to obtain a local abnormal factor detection model.
3) Arranging the abnormal scores from small to large, setting the total number as n, and calculating a third quartile (Q3) of the n abnormal scores:
the position of Q3 is (n +1) × 0.75, and the score at "position of Q3" among the n abnormality scores is the third quartile of the n abnormality scores (Q3);
4) the corresponding collecting and dispatching personnel with the abnormal score of the times of canceling the sending on the day after the collecting is operated is larger than the third quartile is marked as the collecting and dispatching personnel operating the false collecting;
5) screening out order details of the collecting and dispatching personnel marked as the collecting and dispatching personnel operating the false collecting and dispatching in the table 6, wherein the order and the freight note details are marked as the false collecting and dispatching;
6) and removing the weight according to the order number, and transmitting the order number marked as false acquisition, the job number of acquisition and dispatch personnel and the website where the person who is delivered and dispatched properly to an operation assistant.
S4: operational assistant task verification
1) The operation assistant marks the events which are marked as false acquisition and non-acquisition address dispatch in the last month for feedback and evidence submission every month;
2) the network node supervisor completes the checking of the false behavior of the event marked as false acquisition and non-receiving address dispatch in the previous month and the confirmation of whether the event is false or not in the operation assistant by combining the feedback of the receiving and dispatching personnel, the final route of the freight note, the feedback of the client and the like;
in addition, in another embodiment, the website administrator may also add new events of false acquisition and non-acquisition address dispatch, and submit the corresponding invoice number, operation acquisition and dispatch staff number, false type and related evidence that the behavior of false acquisition and non-acquisition address dispatch exists in the website in the last month but is not mentioned in the operation assistant.
In another embodiment, the confirmation returned by the operating assistant is a false solicitation, a list of non-recipient address dispatch orders (feedback information above), which may be transmitted to other departments for calculating payrolls, etc.
S5: false behavior recognition model threshold adjustment
Firstly, a non-receiving address dispatching identification model:
1) according to the table 8 output by the LOF model in the previous month, the abnormal scores are arranged from small to large, the total number is set as n, and the Kth quartile of the n abnormal scores is sequentially calculated (K candidates contain [ 2,2.5,3.3.5 ]):
the position of the kth quartile is (n +1) × K/4, and the score at the "position of the kth quartile" among the n abnormal scores is the kth quartile of the n abnormal scores;
2) the corresponding freight notes with the earth surface distance abnormal score larger than the Kth quartile of the operational fail and the receiving address are marked as the non-receiving address dispatch, and the freight note set marked as the non-receiving address dispatch is Nlabel；
3) The waybill which considers the returned confirmation in the operation assistant as the non-receiving address to dispatchThe waybill mark of the non-receiving address dispatch newly added by each network node supervisor is a real non-receiving address dispatch waybill, and the set of the waybill of the real non-receiving address dispatch is Ntrue；
4) Calculating the recall rate and the accuracy rate, and calculating the detection effect score taking the Kth quartile as a threshold value:
detecting an effect score, namely recall rate and accuracy rate;
5) repeating the steps 1) -4) for each K candidate item, and outputting the K value with the highest detection effect score as Kbest；
6) The abnormal threshold value of the earth surface distance for operating the proper delivery and receiving addresses in the non-receiving address delivery identification model is changed into KthbestA quartile number.
A false acquisition identification model:
1) according to the table 9 output by the LOF model in the previous month, the abnormal scores are arranged from small to large, the total number is set as n, and the kth quartile of the n abnormal scores is sequentially calculated (the K candidates contain [ 2,2.5,3.3.5 ]):
2) the position of the kth quartile is (n +1) × K/4, and the score of the "position where the kth quartile is a point number" in the n abnormal scores is the kth quartile of the n abnormal scores;
3) the corresponding collecting and dispatching personnel with the abnormal times of cancelling the sending of the receiving and dispatching personnel after the receiving and dispatching personnel operates the receiving and dispatching on the same day and the abnormal score of the times of cancelling the sending of the receiving and dispatching personnel is larger than the Kth quartile are marked as the collecting and dispatching personnel operating the false receiving and dispatching;
4) screening out order details of the collecting and dispatching personnel in the previous month table 6 as objects marked as operation false acquisition, wherein the order and the freight note details are marked as false acquisition;
5) the waybill number is deduplicated, and the set of these waybill marked as false acquisition is Nlabel；
6) The waybill which is returned from the operation assistant and really considers the false cable receipts and the waybill which is newly added to the main pipe of each network point are marked as a real false cable collecting waybill, and the set of the real false cable collecting waybill is Ntrue；
7) Calculating the recall rate and the accuracy rate, and calculating the detection effect score taking the Kth quartile as a threshold value:
detecting an effect score as recall rate precision rate;
8) repeating the steps 1) -7) for each K candidate item, and outputting the K value with the highest detection effect score as Kbest；
9) The abnormal threshold value of the times of cancelling the sending of the receiving member after each receiving and sending member operates the receiving member on the same day in the false receiving identification model is changed into KthbestQuartile number.
According to the risk waybill data identification method in the embodiment, a local abnormal factor detection model is built by combining a local abnormal factor detection algorithm (the abnormal degree of each point is judged according to the local density of all points), and intelligent identification of illegal false operation at a receiving and dispatching end is carried out by utilizing the model; closed loop iteration is carried out on the local abnormal factor detection model by combining a mechanism fed back by service personnel, so that the model can be more flexibly adjusted to be an identification threshold value suitable for the current reasonable illegal false operation; the illegal false operation phenomena of false cable collection at a receiving and dispatching end and non-receiving address dispatching are reduced, the customer experience is improved, the customer complaints and the company cost are reduced, and the overall benefit of a company is improved.
It should be understood that, although the steps in the flowcharts involved in the above embodiments are shown in order as indicated by the arrows, the steps are not necessarily performed in order as indicated by the arrows. The steps are not limited to being performed in the exact order illustrated and, unless explicitly stated herein, may be performed in other orders. Moreover, at least a part of the steps in each flowchart involved in the above embodiments may include multiple steps or multiple stages, which are not necessarily performed at the same time, but may be performed at different times, and the order of performing the steps or stages is not necessarily sequential, but may be performed alternately or alternately with other steps or at least a part of the steps or stages in other steps.
In one embodiment, as shown in fig. 5, there is provided a risk manifest data identification apparatus, including: an acquisition module 510, a feature determination module 520, an anomaly score determination module 530, and a recognition module 540, wherein:
an obtaining module 510, configured to obtain historical waybill data;
a characteristic determination module 520, configured to determine a target characteristic associated with a preset abnormal operation according to the historical waybill data;
an abnormal score determining module 530, configured to determine, based on the local abnormal factor detection algorithm and the target feature, an abnormal score corresponding to each historical waybill data and the target feature;
and the identification module 540 is used for determining the historical waybill data with the abnormal score larger than the preset threshold as the risk waybill data with abnormal operation.
After the historical waybill data is obtained, the risk waybill data identification device determines target characteristics associated with preset abnormal operation according to the historical waybill data, determines abnormal scores corresponding to the historical waybill based on the target characteristics and a local abnormal factor detection algorithm, compares the abnormal scores with a preset threshold value, and determines the historical waybill data larger than the preset threshold value as the risk waybill data. When abnormal operation needs to be identified, the device outputs the abnormal scores corresponding to the historical waybill data based on the local abnormal factor detection algorithm by utilizing the characteristic information related to the abnormal operation, and screens out the risk waybill data possibly having the abnormal operation from all the historical waybill data by combining the preset threshold value.
In one embodiment, as shown in fig. 6, the risk waybill data identification device further includes a sorting module 610 and a threshold determination module 620, where:
the sorting module 610 is configured to sort the abnormal scores in a descending order to obtain an abnormal score sorting result;
and a threshold determining module 620, configured to determine a third quartile in the abnormal score sorting result as a preset threshold.
In one embodiment, as shown in fig. 7, the risk waybill data identification device further includes: the system comprises a feedback information receiving module 710, a real abnormal freight note determining module 720, a calculating module 730, a predicted abnormal freight note module 740 and a threshold value determining module 750.
The feedback information receiving module 710 is configured to push the risk manifest data to a corresponding target object and receive feedback information of the target object;
a real abnormal waybill determining module 720, configured to determine, according to the feedback information, a real abnormal waybill with abnormal operation;
a calculating module 730, configured to take more than two k values, and determine a kth quartile corresponding to each k value respectively;
the abnormal freight note prediction module 740 is configured to determine abnormal freight notes to be predicted in the historical freight note data by using the kth quartile as a candidate threshold;
the calculating module 730 is used for calculating the recall rate and the precision rate corresponding to each kth quartile based on the real abnormal waybill and the predicted abnormal waybill;
and a threshold determining module 750, configured to determine a preset threshold according to the recall rate and the precision rate corresponding to each kth quartile.
In one embodiment, the historical waybill data includes whether orders contained in the historical waybill were cancelled; presetting abnormal operations including false acquisition; the target features include: the number of times of sending operation is cancelled by the same person who receives and sends the mail.
In one embodiment, the historical waybill data comprises a blanket address corresponding to the historical waybill, a place of appropriate operation of the historical waybill; presetting abnormal operation including non-receiving party address dispatching; the target features include: the distance between the appropriate operation location and the recipient address.
For specific limitations of the risk waybill data identification device, reference may be made to the above limitations on the risk waybill data identification method, which is not described herein again. The modules in the risk waybill data identification device can be wholly or partially implemented by software, hardware and a combination thereof. The modules can be embedded in a hardware form or independent from a processor in the computer device, and can also be stored in a memory in the computer device in a software form, so that the processor can call and execute operations corresponding to the modules.
In one embodiment, a computer device is provided, which may be a terminal, and its internal structure diagram may be as shown in fig. 8. The computer device includes a processor, a memory, a communication interface, a display screen, and an input device connected by a system bus. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device comprises a nonvolatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of an operating system and computer programs in the non-volatile storage medium. The communication interface of the computer device is used for carrying out wired or wireless communication with an external terminal, and the wireless communication can be realized through WIFI, an operator network, NFC (near field communication) or other technologies. The computer program is executed by a processor to implement a method of risk manifest data identification. The display screen of the computer equipment can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer equipment can be a touch layer covered on the display screen, a key, a track ball or a touch pad arranged on the shell of the computer equipment, an external keyboard, a touch pad or a mouse and the like.
Those skilled in the art will appreciate that the architecture shown in fig. 8 is merely a block diagram of some of the structures associated with the disclosed aspects and is not intended to limit the computing devices to which the disclosed aspects apply, as particular computing devices may include more or less components than those shown, or may combine certain components, or have a different arrangement of components.
In one embodiment, a computer device is provided, comprising a memory and a processor, the memory having a computer program stored therein, the processor implementing the following steps when executing the computer program:
acquiring historical waybill data; determining target characteristics associated with preset abnormal operation according to historical waybill data; determining abnormal scores corresponding to the historical waybill data and the target features based on a local abnormal factor detection algorithm and the target features; and determining the historical waybill data with the abnormal score larger than a preset threshold value as the risk waybill data with abnormal operation.
In one embodiment, the processor, when executing the computer program, further performs the steps of:
sorting the abnormal scores in a descending order to obtain an abnormal score sorting result; and determining a third quartile in the abnormal score sorting result as a preset threshold.
In one embodiment, the processor when executing the computer program further performs the steps of:
pushing the risk waybill data to a corresponding target object, and receiving feedback information of the target object; determining a real abnormal freight note with abnormal operation according to the feedback information; taking more than two k values, and respectively determining a kth quartile corresponding to each k value; respectively taking each kth quartile as a candidate threshold value, and determining a predicted abnormal freight note in each historical freight note data; calculating the recall rate and the precision rate corresponding to each kth quartile based on the real abnormal waybill and the predicted abnormal waybill; and determining a preset threshold according to the recall rate and the precision rate corresponding to each kth quartile.
In another embodiment, a computer-readable storage medium is provided, on which a computer program is stored, which computer program, when executed by a processor, performs the steps of:
acquiring historical waybill data; determining target characteristics associated with preset abnormal operation according to historical waybill data; determining abnormal scores corresponding to the historical waybill data and the target features based on a local abnormal factor detection algorithm and the target features; and determining the historical waybill data with the abnormal score larger than a preset threshold value as the risk waybill data with abnormal operation.
In one embodiment, the computer program when executed by the processor further performs the steps of:
sorting the abnormal scores in a sequence from small to large to obtain an abnormal score sorting result; and determining a third quartile in the abnormal score sorting result as a preset threshold.
In one embodiment, the computer program when executed by the processor further performs the steps of:
pushing the risk waybill data to a corresponding target object, and receiving feedback information of the target object; determining a real abnormal freight note with abnormal operation according to the feedback information; taking more than two k values, and respectively determining the kth quartile corresponding to each k value; respectively taking each kth quartile as a candidate threshold value, and determining a predicted abnormal freight note in each historical freight note data; calculating the recall rate and the precision rate corresponding to each kth quartile based on the real abnormal waybill and the predicted abnormal waybill; and determining a preset threshold according to the recall rate and the precision rate corresponding to each kth quartile.
It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above can be implemented by hardware related to instructions of a computer program, which can be stored in a non-volatile computer-readable storage medium, and when executed, the computer program can include the processes of the embodiments of the methods described above. Any reference to memory, storage, database or other medium used in the embodiments provided herein can include at least one of non-volatile and volatile memory. Non-volatile Memory may include Read-Only Memory (ROM), magnetic tape, floppy disk, flash Memory, optical storage, or the like. Volatile Memory can include Random Access Memory (RAM) or external cache Memory. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM), among others.
The technical features of the above embodiments can be arbitrarily combined, and for the sake of brevity, all possible combinations of the technical features in the above embodiments are not described, but should be considered as the scope of the present specification as long as there is no contradiction between the combinations of the technical features.
The above examples only express several embodiments of the present application, and the description thereof is more specific and detailed, but not construed as limiting the scope of the invention. It should be noted that, for a person skilled in the art, several variations and modifications can be made without departing from the concept of the present application, which falls within the scope of protection of the present application. Therefore, the protection scope of the present patent shall be subject to the appended claims.
Claims (10)
1. A risk waybill data identification method is characterized by comprising the following steps:
acquiring historical waybill data;
determining target characteristics associated with preset abnormal operation according to the historical waybill data;
determining an abnormal score corresponding to each historical waybill data and the target feature based on a local abnormal factor detection algorithm and the target feature;
and determining the historical waybill data with the abnormal score larger than a preset threshold value as the risk waybill data with abnormal operation.
2. The method of claim 1, further comprising, prior to determining historical waybill data having an anomaly score greater than a preset threshold as risky waybill data for abnormal operation:
sequencing all the abnormal scores in a descending order to obtain an abnormal score sequencing result;
and determining a third quartile in the abnormal score sorting result as the preset threshold.
3. The method of claim 2, wherein after determining historical waybill data having an anomaly score greater than a preset threshold as risky waybill data for abnormal operation, further comprising:
pushing the risk waybill data to a corresponding target object, and acquiring feedback information of the target object;
determining a real abnormal freight note with abnormal operation according to the feedback information;
taking more than two k values, and respectively determining a kth quartile corresponding to each k value;
respectively taking each kth quartile as a candidate threshold value, and determining a predicted abnormal freight note in each historical freight note data;
calculating a recall rate and a precision rate corresponding to each kth quartile based on the real abnormal waybill and the predicted abnormal waybill;
and determining the preset threshold according to the recall rate and the precision rate corresponding to each kth quartile.
4. The method of claim 1, wherein:
the historical waybill data comprises whether orders contained in the historical waybill are cancelled or not;
the preset abnormal operation comprises false acquisition;
the target features include: the number of times of sending operation is cancelled by the same person who receives and sends the mail.
5. The method of claim 4, wherein determining an anomaly score for each of the historical waybill data corresponding to the target feature based on a local anomaly factor detection algorithm and the target feature comprises:
taking the operation times of the same receiving and dispatching personnel in all the historical waybill data for cancelling the consignor as a characteristic value, and constructing a local abnormal factor detection model based on a local abnormal factor detection algorithm;
and acquiring an abnormal score corresponding to the target characteristic and each historical waybill data output by the local abnormal factor detection model.
6. The method of claim 1, wherein:
the historical waybill data comprises a collecting address corresponding to the historical waybill and a successful delivery operation place of the historical waybill;
the preset abnormal operation comprises non-recipient address dispatching;
the target features include: the distance between the appropriate operation location and the recipient address.
7. The method of claim 6, wherein determining an anomaly score for each of the historical waybill data corresponding to the target feature based on a local anomaly factor detection algorithm and the target feature comprises:
taking the distance between the appropriate operation place and the addressee in all the historical waybill data as a characteristic value, and constructing a local abnormal factor detection model based on a local abnormal factor detection algorithm;
and acquiring an abnormal score corresponding to the target characteristic and each historical waybill data output by the local abnormal factor detection model.
8. An apparatus for identifying risk manifest data, the apparatus comprising:
the acquisition module is used for acquiring historical waybill data;
the characteristic determining module is used for determining target characteristics associated with preset abnormal operation according to the historical waybill data;
an abnormal score determining module, configured to determine an abnormal score corresponding to the target feature for each piece of historical waybill data based on a local abnormal factor detection algorithm and the target feature;
and the identification module is used for determining the historical waybill data with the abnormal score larger than the preset threshold value as the risk waybill data with abnormal operation.
9. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that the processor, when executing the computer program, implements the steps of the method of any of claims 1 to 7.
10. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the steps of the method of any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011423686.7A CN114612029A (en) | 2020-12-08 | 2020-12-08 | Risk waybill data identification method and device, computer equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011423686.7A CN114612029A (en) | 2020-12-08 | 2020-12-08 | Risk waybill data identification method and device, computer equipment and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114612029A true CN114612029A (en) | 2022-06-10 |
Family
ID=81856739
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011423686.7A Pending CN114612029A (en) | 2020-12-08 | 2020-12-08 | Risk waybill data identification method and device, computer equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114612029A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115018596A (en) * | 2022-08-03 | 2022-09-06 | 浙江口碑网络技术有限公司 | False positioning identification and model training method, device, equipment and medium |
| CN115239242A (en) * | 2022-07-25 | 2022-10-25 | 武汉奥恒翱贸易有限公司 | International trade shipping bill intelligent management method based on digital twin |
-
2020
- 2020-12-08 CN CN202011423686.7A patent/CN114612029A/en active Pending
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115239242A (en) * | 2022-07-25 | 2022-10-25 | 武汉奥恒翱贸易有限公司 | International trade shipping bill intelligent management method based on digital twin |
| CN115239242B (en) * | 2022-07-25 | 2023-07-28 | 广东马泰找船科技有限公司 | International trade shipping bill intelligent management method based on digital twinning |
| CN115018596A (en) * | 2022-08-03 | 2022-09-06 | 浙江口碑网络技术有限公司 | False positioning identification and model training method, device, equipment and medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US8131607B2 (en) | Device and method of planning and managing real-time postal delivery work | |
| US5043908A (en) | Mail delivery system with arrival monitoring | |
| US7509228B2 (en) | Systems and methods for forecasting container density | |
| CA2600835C (en) | Systems and methods for dynamically updating a dispatch plan | |
| US11373536B1 (en) | System and method for customer and/or container discovery based on GPS drive path and parcel data analysis for a waste / recycling service vehicle | |
| CN102282447A (en) | Systems and methods for freight tracking and monitoring | |
| CN109389270B (en) | Logistics object determination method and device and machine readable medium | |
| CN114612029A (en) | Risk waybill data identification method and device, computer equipment and storage medium | |
| CN109583806A (en) | A kind of vehicle scheduling pickup method and system based on after single weight under adjustment | |
| CN115310896B (en) | Logistics information management method based on big data | |
| US8798785B2 (en) | System and process for real time monitoring of mail and print jobs and providing real time verification of mail piece completion | |
| WO2023180581A1 (en) | Digital system for forecasting a future damage or loss impact on cargo or cargo logistics services and automated allocating of a damage cover and method thereof | |
| Sarvareddy et al. | Evaluation of two modeling methods for generating heavy-truck trips at an intermodal facility by using vessel freight data | |
| CN114693039A (en) | Abnormal express identification method and device, computer equipment and storage medium | |
| CN115759514A (en) | Cold chain distribution vehicle scheduling management method and device | |
| CN112529509A (en) | Method for identifying malicious freight notes in transportation platform | |
| CN112132521A (en) | Dangerous and useless traceability management system | |
| CN114548848A (en) | Abnormal waybill identification method and device, computer equipment and storage medium | |
| US20230368114A1 (en) | Systems and methods for logistics facility management | |
| US11810061B2 (en) | Pre-trip inspection prediction and PTI reduction systems, processes and methods of use | |
| CN114912873A (en) | Dispatch stagnation exception prompting method, device, equipment and storage medium | |
| CN115018107A (en) | Express item timeliness prediction method and device, computer equipment and storage medium | |
| CN116955853A (en) | Method, device, equipment and storage medium for predicting and inquiring order signing time | |
| CN118134371A (en) | Method, device, equipment and storage medium for generating departure certificate of cargo transportation | |
| CN113139770A (en) | Express delivery routing data analysis method, device, equipment and storage medium |
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 |