Method, device, electronic device and readable medium for mining abnormality of declared weight
The method and device for mining anomalies in declared weight address inefficiencies in customs clearance by determining adaptive reference weights and tolerance ranges, enhancing the accuracy and efficiency of customs declaration processing.
Patent Information
- Application Number
- CN202011566089.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2020-12-25
- Publication Date
- 2024-12-06
- Estimated Expiration
- 2040-12-25
AI Technical Summary
Existing anomaly mining methods in customs clearance fail to efficiently identify weight anomalies in customs declarations due to their focus on individual data sets, ignoring the need for adaptive reference standards and tolerance ranges, leading to inefficiencies in identifying over- or under-reported weights for tax evasion.
A method and device for mining anomalies in declared weight that determine a reference weight and adaptive tolerance range based on historical data, using statistical indicators and machine learning algorithms to identify abnormal weights and generate alarm prompts.
This approach improves the efficiency of customs clearance operations by accurately identifying abnormal weights, reducing human intervention, and minimizing errors in customs declaration processing.
Smart Images

Figure CN112598354B_ABST
Abstract
Description
Technical Field
[0001] The present disclosure relates to the field of anti-express customs clearance fraud, and in particular to a method, device, electronic device and computer-readable medium for mining anomalies in declared weight. Background Art
[0002] In the field of express customs clearance, products priced by weight, such as steel products or some industrial raw materials, will have an impact on the import tax declaration if the weight is over- or under-reported. The declaring unit is likely to falsely report the weight for the purpose of tax evasion. Customs inspection mainly solves the problem of "inconsistency between declaration and goods". Customs officers need to determine whether the declared weight of each item in the customs declaration form is abnormal before arranging customs clearance. However, the number of customs declarations per day is as high as tens of thousands, and the variety of declared items is far beyond the scope of human processing, and can only be spot-checked, which is time-consuming, labor-intensive, inefficient and inefficient.
[0003] Therefore, express customs clearance requires a method that can quickly identify which items may have weight anomalies. Weight anomalies refer to data points in a series of weight declaration data provided by different declarants for the same item that do not match other data. When a large amount of data cannot be manually marked with outliers, automatic anomaly mining becomes crucial.
[0004] Existing anomaly mining methods usually only focus on a certain data set, looking for possible outliers in it, and ignoring the outliers as noise. However, they do not conduct in-depth research on the conventional reference mode of the data set and determine the normal reference standards for different types of data sets. This obviously cannot meet the application requirements of customs declaration weight estimation.
[0005] Therefore, a new method, device, electronic device and computer-readable medium for anomaly mining of declared weight are needed.
[0006] The above information disclosed in this Background section is only for enhancement of understanding of the background of the present disclosure and therefore it may contain information that does not constitute the prior art that is already known to one of ordinary skill in the art. Summary of the invention
[0007] In view of this, the embodiments of the present disclosure provide a method, device, electronic device and computer-readable medium for mining abnormal declared weight, which can quickly mine abnormal declared weight according to reference weight and adaptive tolerance range, thereby improving the efficiency of express clearance operations.
[0008] Other features and advantages of the present disclosure will become apparent from the following detailed description, or may be learned in part by the practice of the present disclosure.
[0009] According to a first aspect of an embodiment of the present disclosure, a method for mining anomalies in declared weight is proposed, the method comprising: obtaining a target declared weight of a target declared item; determining a reference weight of the target declared item based on a historical declared weight sequence of the target declared item; determining a tolerance range corresponding to the reference weight of the target declared item based on a preset data set; if it is determined that the target declared weight of the target declared item is abnormal based on the reference weight of the target declared item and the tolerance range, generating an alarm prompt message based on the target declared item.
[0010] In an exemplary embodiment of the present disclosure, determining the reference weight of the target declared item based on the historical declared weight sequence of the target declared item includes: determining different weight values and the number of different weight values appearing in the historical declared weight sequence; determining the reference weight of the target declared item based on the different weight values and the number of different weight values.
[0011] In an exemplary embodiment of the present disclosure, determining the reference weight of the target declared item according to the different weight values and the number of the different weight values includes: if the number of the different weight values is 1, determining the different weight value as the reference weight of the target declared item; if the number of the different weight values is 2, determining the reference weight of the target declared item according to the proportion of the number of occurrences of the larger weight value and the smaller weight value among the different weight values in the historical declared weight sequence; if the number of the different weight values is greater than or equal to 3, determining the reference weight of the target declared item according to the coefficient of dispersion, the skewness coefficient, the average value of the historical declared weight sequence, the larger weight value and the smaller weight value among the different weight values.
[0012] In an exemplary embodiment of the present disclosure, determining the reference weight of the target declared item based on the dispersion coefficient, skewness coefficient, average value of the historical declared weight sequence, and the larger weight value and the smaller weight value among the different weight values of the historical declared weight sequence includes: if the dispersion coefficient is less than or equal to the dispersion coefficient threshold, determining the reference weight of the target declared item based on the average value of the historical declared weight sequence; if the dispersion coefficient is greater than the dispersion coefficient threshold, eliminating the abnormal values of the historical declared weight sequence based on the dispersion coefficient, skewness coefficient, and the maximum weight value and the minimum weight value among the different weight values of the historical declared weight sequence, so as to determine the reference weight of the target declared item based on the average value of the historical declared weight sequence after elimination.
[0013] In an exemplary embodiment of the present disclosure, determining the tolerance range corresponding to the reference weight of the target declared item according to a preset data set includes: determining the reference weight of each historical item in the preset data set; binning each historical item according to the reference weight of each historical item to obtain weight bins; determining the target weight bin of the target declared item in the weight bins according to the reference weight of the target declared item; determining the upper threshold coefficient and the lower threshold coefficient of the target weight bin; determining the tolerance range of the target declared item according to the upper threshold coefficient and the lower threshold coefficient of the target weight bin and the reference weight of the target declared item.
[0014] In an exemplary embodiment of the present disclosure, determining the upper threshold coefficient and the lower threshold coefficient of the target weight bin includes: determining a first average floating degree of each historical object according to a reference weight of each historical object and a declared weight less than the reference weight in a historical declared weight sequence of each historical object; determining a second average floating degree of each historical object according to the reference weight of each historical object and a declared weight greater than or equal to the reference weight in a historical declared weight sequence of each historical object; binning each historical object according to the first average floating degree of each historical object to obtain a first floating degree bin; binning each historical object according to the second average floating degree of each historical object to obtain a second Floating degree binning; determining the ratio of each first floating degree bin by the ratio of the number of historical items falling into the target weight bin in each first floating degree bin to the number of historical items in each first floating degree bin; determining the ratio of each second floating degree bin by the ratio of the number of historical items falling into the target weight bin in each second floating degree bin to the number of historical items in each second floating degree bin; determining the upper threshold limit coefficient of the target weight bin according to the ratio of each first floating degree bin and the right interval value of each first floating degree bin; determining the lower threshold limit coefficient of the target weight bin according to the ratio of each second floating degree bin and the right interval value of each second floating degree bin.
[0015] In an exemplary embodiment of the present disclosure, the method further includes: generating the preset data set based on historical declaration events with the same tax number as the target declared item; or generating the preset data set based on historical declaration events with the same brand as the target declared item.
[0016] According to a second aspect of an embodiment of the present disclosure, a device for mining anomalies in declared weight is proposed, the device comprising: a weight acquisition module, configured to acquire a target declared weight of a target declared item; a reference weight module, configured to determine a reference weight of the target declared item according to a historical declared weight sequence of the target declared item; a tolerance range module, configured to determine a tolerance range corresponding to the reference weight of the target declared item according to a preset data set; and an abnormal alarm module, configured to generate an alarm prompt message according to the target declared item if it is determined that the target declared weight of the target declared item is abnormal according to the reference weight of the target declared item and the tolerance range.
[0017] According to a third aspect of an embodiment of the present disclosure, an electronic device is proposed, comprising: one or more processors; a storage device for storing one or more programs; when the one or more programs are executed by the one or more processors, the one or more processors implement any of the above-mentioned methods for mining anomalies in declared weight.
[0018] According to a fourth aspect of an embodiment of the present disclosure, a computer-readable medium is provided, on which a computer program is stored, and when the program is executed by a processor, the method for mining anomalies in declared weight as described in any one of the above items is implemented.
[0019] According to the method, device, electronic device and computer-readable medium for abnormal mining of declared weight provided by certain embodiments of the present disclosure, the reference weight of the target declared item is determined according to the historical declared weight sequence of the target declared item, and the reference weight value of each item can be determined based on the historical declared data, and the standard library of item weight can be constructed to provide decision support for express customs clearance, avoiding the weight confusion caused by different historical weight sequences of the same item. According to the preset data set, the tolerance range corresponding to the reference weight of the target declared item is determined, and the weight characteristics of different items can be comprehensively considered. Based on the preset data set, the tolerance coefficient with adaptive characteristics for different weight intervals is constructed to generate the normal tolerance range of each item weight, avoiding the situation that when the weight difference between the declared items is large, the same threshold will have errors or cause inconsistency with the actual situation. When the target declared weight of the target declared item is determined to be abnormal according to the reference weight and tolerance range of the target declared item, an alarm prompt information is generated according to the target declared item to improve the efficiency of express customs clearance operations.
[0020] It is to be understood that the foregoing general description and the following detailed description are exemplary only and are not restrictive of the present disclosure. BRIEF DESCRIPTION OF THE DRAWINGS
[0021] The drawings herein are incorporated into the specification and constitute a part of the specification, illustrate embodiments consistent with the present disclosure, and together with the specification are used to explain the principles of the present disclosure. The drawings described below are only some embodiments of the present disclosure, and for ordinary technicians in this field, other drawings can also be obtained based on these drawings without creative work.
[0022] Figure 1 The invention is a system block diagram of a method and device for mining anomalies in declared weight according to an exemplary embodiment.
[0023] Figure 2 The present invention is a flowchart of a method for mining anomalies in reported weight according to an exemplary embodiment.
[0024] Figure 3 The present invention is a flowchart of a method for mining anomalies in reported weight according to an exemplary embodiment.
[0025] Figure 4 The present invention is a flowchart of a method for mining anomalies in reported weight according to an exemplary embodiment.
[0026] Figure 5 The present invention is a flowchart of a method for mining anomalies in reported weight according to an exemplary embodiment.
[0027] Figure 6 The present invention is a flowchart of a method for mining anomalies in reported weight according to an exemplary embodiment.
[0028] Figure 7 is a schematic diagram of binning according to an exemplary embodiment.
[0029] Figure 8 The present invention is a flowchart of a method for mining anomalies in reported weight according to an exemplary embodiment.
[0030] Fig. 9 The invention is a flowchart of a method for mining anomalies in declared weight according to another exemplary embodiment.
[0031] Fig.10 The invention is a flowchart of a method for mining anomalies in declared weight according to another exemplary embodiment.
[0032] Fig.11 The invention is a block diagram of an abnormal mining device for declaring weight according to an exemplary embodiment.
[0033] Fig.12 A block diagram schematically shows an electronic device in an exemplary embodiment of the present disclosure. DETAILED DESCRIPTION
[0034] Example embodiments will now be described more fully with reference to the accompanying drawings. However, example embodiments can be implemented in many forms and should not be construed as limited to the embodiments set forth herein; rather, these embodiments are provided so that the present invention will be comprehensive and complete and fully convey the concepts of the example embodiments to those skilled in the art. The same reference numerals in the figures represent the same or similar parts, and thus their repeated description will be omitted.
[0035] The described features, structures or characteristics may be combined in one or more embodiments in any suitable manner. In the following description, many specific details are provided to provide a full understanding of the embodiments of the present invention. However, those skilled in the art will appreciate that the technical solution of the present invention may be practiced while omitting one or more of the specific details, or other methods, components, devices, steps, etc. may be adopted. In other cases, known methods, devices, implementations or operations are not shown or described in detail to avoid blurring various aspects of the present invention.
[0036] The accompanying drawings are only schematic illustrations of the present invention, and the same reference numerals in the drawings represent the same or similar parts, so their repeated description will be omitted. Some block diagrams shown in the accompanying drawings do not necessarily correspond to physically or logically independent entities. These functional entities can be implemented in software form, or in one or more hardware modules or integrated circuits, or in different networks and / or processor devices and / or microcontroller devices.
[0037] The flowcharts shown in the accompanying drawings are only exemplary and do not necessarily include all the contents and steps, nor must they be executed in the order described. For example, some steps can be decomposed, and some steps can be combined or partially combined, so the actual execution order may change according to actual conditions.
[0038] The weight anomaly mining algorithm for customs declarations needs to target tens of thousands of different categories of items. The declared weight data characteristics of different categories of items are also different. It is necessary to use statistical indicators and machine learning algorithms to analyze the weight data of different commodities, and establish a weight warehouse for various commodities from point to surface. Considering that in actual applications, the weight filled in by the declarant has a certain error range, it is also necessary to determine a threshold range based on different classification methods such as the tax number and brand of the commodity. Only those outside this normal range will be included in the key control objects.
[0039] Exemplary embodiments of the present invention will be described in detail below with reference to the accompanying drawings.
[0040] Figure 1 The invention is a system block diagram of a method and device for mining anomalies in declared weight according to an exemplary embodiment.
[0041] The server 105 may be a server that provides various services, such as a background management server (only as an example) that provides support for the declared weight anomaly mining system operated by the user using the terminal devices 101, 102, and 103. The background management server may analyze and process the received declared weight anomaly mining request and other data, and feed back the processing results (such as reference range, tolerance range, alarm prompt information - only as an example) to the terminal device.
[0042] The server 105 may, for example, obtain the target declared weight of the target declared item; the server 105 may, for example, determine the reference weight of the target declared item based on the historical declared weight sequence of the target declared item; the server 105 may, for example, determine the tolerance range corresponding to the reference weight of the target declared item based on a preset data set. The server 105 may, for example, generate an alarm prompt message based on the target declared item if the target declared weight of the target declared item is abnormal based on the reference weight and the tolerance range of the target declared item.
[0043] The server 105 may be a physical server, or may be composed of multiple servers, for example, a part of the server 105 may be used as the abnormal mining task submission system of the declared weight in the present disclosure, for obtaining the task to execute the abnormal mining command of the declared weight; and a part of the server 105 may also be used as the abnormal mining system of the declared weight in the present disclosure, for example, for obtaining the target declared weight of the target declared item; determining the reference weight of the target declared item according to the historical declared weight sequence of the target declared item; determining the tolerance range corresponding to the reference weight of the target declared item according to the preset data set; if the target declared weight of the target declared item is determined to be abnormal according to the reference weight of the target declared item and the tolerance range, generating an alarm prompt information according to the target declared item. .
[0044] Figure 2 1 is a flowchart of a method for mining anomalies in declared weight according to an exemplary embodiment. The method for mining anomalies in declared weight provided in the embodiment of the present disclosure can be executed by any electronic device with computing and processing capabilities, such as terminal devices 101, 102, 103 and / or server 105. In the following embodiments, the server execution method is used as an example for illustration, but the present disclosure is not limited thereto. The method 20 for mining anomalies in declared weight provided in the embodiment of the present disclosure can include steps S202 to S208.
[0045] like Figure 2 As shown, in step S202, the target declared weight of the target declared article is obtained.
[0046] In the embodiment of the present disclosure, when a new declaration request is received, the target declared weight of the target declared item can be obtained from the declaration request, wherein the target declared item may include but is not limited to data such as item name, tax number, declarant, and declared weight.
[0047] In step S204, the reference weight of the target declared item is determined according to the historical declared weight sequence of the target declared item.
[0048] In the embodiment of the present disclosure, the historical declared weight sequence of the target declared item may be a sequence of declared weights in the historical declared events where the declared items are the same as the target declared items. For example, the historical declared events may be clustered according to the declared items, and the declared weights in the historical declared events of the same type after clustering may be used to generate the historical declared weight sequence.
[0049] In order to avoid the weight reference deviation caused by repeated declaration of the same item by the same declarant, the weight data can be pre-processed. For example, when the name, brand, specification, tax number, country of origin, declaring unit, and sender information of an item are the same, if the number of declarations of the same weight of the item is > 1, it will be unified to 2, otherwise it will be 1.
[0050] In step S206, a tolerance range corresponding to the reference weight of the target declared item is determined according to the preset data set.
[0051] In an embodiment of the present disclosure, the tolerance coefficients of different weight intervals can be determined according to a preset data set, so as to determine the tolerance coefficients of the weight intervals corresponding to the reference weight of the target declared item as the tolerance range corresponding to the reference weight of the target declared item. The tolerance range is the maximum range in which the declared weight of the target declared item deviates from the reference weight. The normal reference range of the target declared weight of the target declared item can be determined according to the reference weight and tolerance range of the target declared item. For example, the reference weight is W, and the tolerance range is: [a, b], where a is the upper limit, b is the lower limit, a<0, b>0, then the normal reference range of the target declared weight can be expressed as [W+a, W+b]. The preset data set can be a data set of all declared events in the database. In an exemplary embodiment, a preset data set can also be generated based on historical declaration events with the same tax number as the target declared item; or a preset data set can be generated based on historical declaration events with the same brand as the target declared item.
[0052] In step S208, if it is determined that the target declared weight of the target declared item is abnormal according to the reference weight and the tolerance range of the target declared item, an alarm prompt information is generated according to the target declared item.
[0053] In the disclosed embodiment, when the target declared weight of the target declared item does not fall within a normal reference range determined according to the reference weight and tolerance range of the target declared item, it may be confirmed that the target declared weight of the target declared item is abnormal.
[0054] According to the method for mining the abnormality of declared weight provided by the embodiment of the present disclosure, the reference weight of the target declared item is determined according to the historical declared weight sequence of the target declared item. The reference weight value of each item can be determined based on the historical declared data, and a standard library of item weight can be constructed to provide decision support for express customs clearance, thereby avoiding the weight confusion caused by different historical weight sequences of the same item. According to the preset data set, the tolerance range corresponding to the reference weight of the target declared item is determined. The weight characteristics of different items can be comprehensively considered, and tolerance coefficients with adaptive characteristics for different weight intervals can be constructed based on the preset data set to generate a normal tolerance range for the weight of each item, thereby avoiding the situation where the same threshold value will have errors or cause inconsistency with the actual situation when the weight difference between the declared items is large. When the target declared weight of the target declared item is determined to be abnormal according to the reference weight and tolerance range of the target declared item, an alarm prompt message is generated according to the target declared item to improve the efficiency of express customs clearance operations.
[0055] Figure 3 The flowchart of a method for mining anomalies in declared weight according to an exemplary embodiment is shown. The method for mining anomalies in declared weight provided by the embodiment of the disclosure may include steps S302 to S304 when determining the reference weight of the target declared item according to the historical declared weight sequence of the target declared item.
[0056] like Figure 3 As shown, in step S302, different weight values and the number of different weight values appearing in the historical declared weight sequence are determined.
[0057] In the embodiment of the present disclosure, different weight values in the historically declared weight sequence refer to weight values with different values that appear in the historically declared weight sequence. For example, for the historically declared weight sequence: (1, 1, 3, 3, 4, 5), the different weight values are: 1, 3, 4, 5, and the number of different weight values is 4.
[0058] In step S304, the reference weight of the target declared item is determined according to the different weight values and the number of different weight values.
[0059] In the embodiment of the present disclosure, if the number of different weight values is 1, the different weight value can be determined as the reference weight of the target declared item. When the number of different weight values is 1, it means that there is only one weight value in the historical declared weight sequence, and the only different weight value can be determined as the reference weight of the target declared item. This example can be found in the attached Figure 4The branch step of the number of different weight values counts_num==1 is as follows. When the number of different weight values counts_num==1, the reference weight ref_num=num_list[0], wherein num_list is the historical declared weight sequence of the target declared item, and num_list[0] is the first element value of the historical declared weight sequence.
[0060] If the number of different weight values is 2, the reference weight of the target declared item is determined based on the proportion of the number of times the larger weight value and the smaller weight value appear in the historical declared weight sequence. Figure 4 In the branch step of different weight values counts_num==2, there are 2 different weight values in the historical declared weight sequence, assuming they are W0 (the smaller weight value of the 2 different weight values) and W1 (the larger weight value). Compare the proportion of the number of occurrences of the 2 different weight values in the historical declared weight sequence (that is, the proportion of the number of declarations of different weight values). When the difference in the proportion of the number of occurrences is greater than the proportion threshold, the different weight value with the larger proportion of the number of occurrences is the reference weight. For example, about 20% of the values in the historical declared weight sequence are W0 (smaller weight value), 80% are W1 (larger weight value), and the proportion threshold is 40% in percentage form. Since 80%-20%>40%, the reference weight is W1 (larger weight value).
[0061] For another example, when the difference in the proportion of the occurrences of two different weight values is less than or equal to the proportion threshold, and W0 is less than half of W1, W1 (the larger weight value) is the reference weight. In other cases, the average value of the weight sequence can be used as the reference weight.
[0062] If the number of different weight values is greater than or equal to 3, the reference weight of the target declared item is determined based on the dispersion coefficient, skewness coefficient, average value of the historical declared weight sequence, and the larger and smaller weight values of the different weight values. Figure 5 The branching step of different weight values counts_num>=3, at this time, there are multiple different values in the weight sequence, and the statistical characteristics of the weight sequence are calculated: dispersion coefficient cv, skew coefficient skew, minimum weight value q0, and maximum weight value q4. When the dispersion coefficient is less than or equal to the dispersion coefficient threshold, that is, Figure 5 For the branch step where cv≤0.1 in the above formula, it can be considered that the weight distribution in the historical declared weight sequence is concentrated, and the reference weight of the target declared item can be determined based on the average value of the historical declared weight sequence.
[0063] When the dispersion coefficient is greater than the dispersion coefficient threshold, that is, Figure 5In the branching step where cv>0.1, that is, when there are extreme values in the historical declared weight sequence or the data distribution of the historical declared weight sequence is relatively dispersed, the outliers in the historical declared weight sequence can be eliminated according to the dispersion coefficient, skewness coefficient, and the maximum weight value and minimum weight value of the historical declared weight sequence. For example, the maximum inter-class variance algorithm (OTSU algorithm) is used to mine outliers in the historical declared weight sequence. After the outliers are eliminated from the historical declared weight sequence by combining statistical indicators such as the skewness coefficient and the ratio of the maximum weight value to the minimum weight value, the average value of the historical declared weight sequence after elimination is used as the reference weight.
[0064] For example Figure 5 As shown, when the discrete coefficient cv satisfies 0.1≤cv≤0.3, it is considered that the numerical distribution of the values is relatively concentrated, but there are still extreme values. In this case, when the minimum weight value q0 is less than half of the maximum weight value q4 / 2, the reference weight can be obtained by the following steps: S501, use the otsu algorithm to calculate the segmentation threshold thres_otsu of the historical declared weight sequence; S502, if the skew coefficient skew<0, then the weight values greater than or equal to thres_otsu in the historical declared weight sequence num_list are divided into inlier_list, and the weight values less than thres_otsu are divided into outlier_list; if the skew coefficient skew≥0, then the weight values less than or equal to t in num_list are divided into outlier_list. The weight value of hres_otsu is divided into inlier_list, and the weight value greater than thres_otsu is divided into outlier_list; S503, if the number of different weight values in outlier_list is greater than 1 and greater than or equal to the number of different weight values in inlier_list, and the sequence length of outlier_list is greater than or equal to 3 / 4 of the sequence length of inlier_list, then swap outlier_list and inlier_list; S504, if inlier_list has only two different weight values, call Figure 4 The reference weight is obtained by the get_weight_ref_num_with_2_element method in S505; otherwise, the reference weight is equal to the average value of inlier_list.
[0065] exist Figure 5 In the branch step where the discrete coefficient cv satisfies 0.1≤cv≤0.3, if the minimum weight value q0 is greater than or equal to half of the maximum weight value q4 / 2, the reference weight is equal to the average value of the historical declared weight sequence num_list.
[0066] exist Figure 5When the discrete coefficient cv is greater than 0.3 in the branching step, it is considered that the numerical distribution of the data is relatively dispersed or there are extreme values. In this case, the reference weight can be obtained through the following steps: S506, use the otsu algorithm to calculate the segmentation threshold thres_otsu of the historical declared weight sequence; S507, if the skew coefficient skew<0, the weight values in the historical declared weight sequence num_list that are greater than or equal to thres_otsu are divided into inlier_list, and the weight values that are less than thres_otsu are divided into outlier_list; if the skew coefficient skew≥0, the weight values in num_list that are less than or equal to thres_otsu are divided into inlier_list, and the weight values that are greater than thres_otsu are divided into outlier_list; S508, if the number of different weight values in outlier_list is greater than 1 and greater than or equal to the number of different weight values in inlier_list, and the sequence length of outlier_list is greater than or equal to 3 / 4 of the sequence length of inlier_list, then the outlier_list and inlier_list are exchanged; S509, if inlier_list has only two different weight values, call Figure 4 Get the reference weight using the get_weight_ref_num_with_2_element method in; S510, otherwise calculate the cv value of inlier_list and update skew and cv; S511, if the updated cv>0.3, repeat steps S506-S510, otherwise the reference weight is equal to the average value of inlier_list.
[0067] Figure 6 The flowchart of a method for mining anomalies in declared weight according to an exemplary embodiment is shown. The method for mining anomalies in declared weight provided by the embodiment of the disclosure may include steps S602 to S610 when determining the tolerance range corresponding to the reference weight of the target declared item according to a preset data set.
[0068] In step S602, the reference weight of each historical item in the preset data set is determined.
[0069] In the embodiment of the present disclosure, the reference weight of each historical item can be determined by Figure 3 , Figure 4 and Figure 5 Similar steps are not repeated here.
[0070] In step S604, each historical item is binned according to its reference weight to obtain weight bins.
[0071] In the embodiment of the present disclosure, an example diagram of the binning result can be shown as follows: Figure 7 As shown. Figure 7 In the example of binning results shown, the reference weight is in [0, 3) (units such as Figure 7 Historical items D, E, and J within the reference weight range of [kg) are grouped into one weight bin, historical items C, H, and J within the reference weight range of [3, 5) are grouped into one weight bin, and historical items A, B, G, and F within the reference weight range of [5, 8) are grouped into one weight bin.
[0072] In step S606, a target weight bin of the target declared item in the weight bin is determined according to the reference weight of the target declared item.
[0073] In the embodiment of the present disclosure, the weight bin corresponding to the weight range within which the reference weight of the target declared article falls can be determined as the target weight bin. Figure 7 For example, assuming that the reference weight of the target declared item is 3.8 kg, it falls into the weight bin in the range of [3, 5).
[0074] In step S608, the upper threshold coefficient and the lower threshold coefficient of the target weight bin are determined.
[0075] In the disclosed embodiment, the upper threshold coefficient and the lower threshold coefficient corresponding to the target weight bin can be calculated in advance. For another example, the upper threshold coefficient and the lower threshold coefficient corresponding to each weight bin can be calculated so that when the reference weight of the target declared article falls into any weight bin, the upper threshold coefficient and the lower threshold coefficient corresponding to the weight bin can be determined.
[0076] In step S610, the tolerance range of the target declared items is determined according to the upper threshold coefficient and the lower threshold coefficient of the target weight bin and the reference weight of the target declared items.
[0077] In the present disclosure, it is assumed that the upper threshold coefficient is expressed as α high , the lower threshold coefficient is expressed as α low , the reference weight of the target declared item is expressed as W ref , then the tolerance range of the target declared item can be expressed as: [W ref *(1-α low ), W ref *(1+α high )].
[0078] Figure 8 The flowchart of a method for mining anomalies in declared weight according to an exemplary embodiment is shown. The method for mining anomalies in declared weight provided by the embodiment of the disclosure may include steps S802 to S810 when determining the upper threshold coefficient and the lower threshold coefficient of the target weight bin.
[0079] In step S802, a first average floating degree of each historical item is determined according to the reference weight of each historical item and the declared weight less than the reference weight in the historical declared weight sequence of each historical item.
[0080] In the disclosed embodiment, the historical declared weight sequence of historical items may be a sequence of declared weights of declared items in historical declaration events that are the same as the historical items. The first average floating degree of the historical items is the absolute value of the average floating degree of each element in the historical declared weight sequence that is less than the reference weight and the reference weight of the historical items. For example, for the historical declared weight sequence (s1, s2, s3, s4) = (1, 2, 3, 4), assuming that its reference weight is w = 2.5, the floating degree fi of each element in the historical declared weight sequence can be calculated as f i =|s i -W| / w, i=1, 2, 3, 4. The first average floating degree g1 of the historical item can be calculated as g1(∑ si<w f i ) / k, where k is the number of elements in the historical declared weight sequence that are less than the reference weight. For example, in the historical declared weight sequence (1, 2, 3, 4), since w = 2.5, its first average floating degree
[0081] In step S804, a second average floating degree of each historical item is determined according to the reference weight of each historical item and the declared weight greater than or equal to the reference weight in the historical declared weight sequence of each historical item.
[0082] In the embodiment of the present disclosure, the second average floating degree of the historical item is the absolute value of the average floating degree of each element in the historical declared weight sequence that is greater than or equal to the reference weight and the reference weight of the historical item. For example, for the historical declared weight sequence (s1, s2, s3, s4) = (1, 2, 3, 4), assuming that its reference weight w = 2.5, the second average floating degree f2 of the historical item can be calculated as f2 = (∑ si≥w f i ) / k, where k is the number of elements in the historical declared weight sequence that are greater than or equal to the reference weight. For example, in the historical declared weight sequence (1, 2, 3, 4), since w = 2.5, its second average fluctuation degree is
[0083] In step S806, each historical item is binned according to the first average floating degree of each historical item to obtain first floating degree bins.
[0084] In the embodiment of the present disclosure, the schematic diagram of the first floating degree binning can be as follows: Figure 7 As shown. Among them, historical items A, B, C, D, and E whose first average floating degree is within the range of (0, 0.05] are located in the same first floating degree bin. Similarly, F, G, and H are located in the same first floating degree bin, and I and J are located in the same first floating degree bin.
[0085] In step S808, each historical item is binned according to the second average floating degree of each historical item to obtain second floating degree bins.
[0086] In the disclosed embodiment, the second average floating degree binning result is similar to the first average floating degree binning result, which will not be described in detail here.
[0087] In step S810, the ratio of the number of historical items falling into the target weight bin in each first floating degree bin to the number of historical items in each first floating degree bin is determined as the proportion of each first floating degree bin.
[0088] In the disclosed embodiment, assuming that the target weight bin is (3, 5], taking the first floating degree bin as (0, 0.05] as an example, which includes historical items A, B, C, D, and E, among which the historical items falling into the target weight bin are A and B, then the proportion pct1 of the first floating degree bin (0, 0.05] is calculated as: 2 / 5=0.4. Similarly, the proportions of the first floating degree bins such as (0.05, 0.1] and (0.1, 0.15] can be calculated.
[0089] In step S812, the ratio of the number of historical items falling into the target weight bin in each second floating degree bin to the number of historical items in each second floating degree bin is determined as the proportion of each second floating degree bin.
[0090] In the disclosed embodiment, the calculation method of the ratio of the second floating degree bin is similar to the calculation method of the ratio of the first floating degree bin, which will not be described in detail here.
[0091] In step S814, the threshold upper limit coefficient of the target weight bin is determined according to the proportion of each first floating degree bin and the right interval value of each first floating degree bin.
[0092] In the embodiment of the present disclosure, the proportion of each first floating degree bin can be generated as a proportion vector, and the right interval of each first floating degree bin can be generated as a right interval vector, then the threshold upper limit coefficient of the target weight bin can be calculated as the inner product of the proportion vector and the right interval vector. The calculation method of the threshold upper limit coefficient is expressed as: α1 = ∑pct 1i *i1. Among them, i1 is the right interval of the first floating degree bin, pct 1iis the ratio of the corresponding first floating degree bins. In other embodiments, the upper threshold coefficient may be set to 2α1 based on experience.
[0093] In step S816, the lower threshold coefficient of the target weight bin is determined according to the proportion of each second floating degree bin and the right interval value of each second floating degree bin.
[0094] In the embodiment of the present disclosure, the calculation method of the lower threshold coefficient can be similar to that of the upper threshold coefficient. For example, the calculation method of the lower threshold coefficient is expressed as: α2 = ∑ pct 2i *i2. i2 is the right interval of the second floating level bin, pct 2i is the ratio of the corresponding second floating degree bins. In other embodiments, the upper threshold coefficient may be set to 2α2 based on experience.
[0095] The changes of the upper and lower threshold coefficients with the reference weight bins can be shown in Table 1.
[0096] Table 1 Changes of upper and lower threshold coefficients with weight binning
[0097] W(weight by box) <![CDATA[α low (lower threshold coefficient)]]> <![CDATA[α high (Upper threshold coefficient)<!-- 9 --> ]]> 0.5 0.25 0.261 1 0.204 0.214 1.5 0.145 0.158 2 0.141 0.151 3 0.158 0.14 4 0.125 0.131 5 0.1 0.1 10 0.1 0.1 20 0.1 0.1 30 0.1 0.1 40 0.1 0.1
[0098] Among them, weight bin 0.5 refers to the weight bin of (0, 0.5], and weight bin 1 refers to the weight bin of (0.5, 1].
[0099] In other embodiments, when the preset data set is generated based on the historical declaration event with the same tax number as the target declared item, the historical items can be first classified according to the tax number, and then the upper threshold coefficient and the lower threshold coefficient corresponding to different tax numbers can be calculated according to the above steps. For the average floating degree of the tax number category with a large number of missing values, the average floating value of all samples can be used instead. For the missing values in the low weight interval of the data set of each tax number, 0.3 can be used to fill, and the missing values in the high weight interval can be filled with 0.1.
[0100] After analysis, the reference weight thresholds of the scheme are as follows: 1_α, 4_α, 6_α, 9_α, 10_α, 11_α, 22_α, 27_α, and others represent preset data sets of different tax numbers.
[0101] Table 2 Changes of threshold lower limit coefficient with weight binning
[0102] Weight binning 1_α 4_α 6_α 9_α 10_α 11_α 22_α 27_α other (0,0.5] 0.254 0.221 0.6 0.237 0.233 0.247 0.314 0.269 0.25 (0.5,1] 0.209 0.208 0.12 0.183 0.2 0.159 0.15 0.214 0.204 (1,1.5] 0.15 0.1 0.2 0.145 0.1 0.214 0.167 0.109 0.145 (1.5,2] 0.149 0.1 0.4 0.125 0.1 0.113 0.1 0.2 0.141 (2,3] 0.167 0.2 0.1 0.1 0.1 0.2 0.111 0.1 0.158 (3,4] 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.3 0.125 (4,5] 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 (5,10] 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 (10,20] 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 (20,30] 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 (30,40] 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1
[0103] Table 3 Changes of upper threshold coefficient with weight binning
[0104]
[0105]
[0106] In other embodiments, when the preset data set is generated based on the historical declaration event of the same brand as the target declared item, the historical items can be first classified by brand, and then the upper threshold coefficient and the lower threshold coefficient corresponding to different brands can be calculated according to the above steps. For data with empty brands, "other" can be used to fill it and treated as a category. Assuming 0.1≤α≤0.3, there are 3834 commodity brands in the sample, and the lower threshold coefficient and the upper threshold coefficient are calculated according to the floating degree of the weight sequence of each brand.
[0107] In another exemplary embodiment of the present disclosure, in the method for mining the anomaly of declared weight, the express customs declaration data of the second half of a certain period can be analyzed, and the weight sequence data of each historical item can be obtained after preprocessing. The reference weight is calculated based on the historical declared weight sequence of each historical item, and the weight standard library of the item is established. Some results are shown in Table 4.
[0108] Table 4 Item weight standard library (partial)
[0109]
[0110]
[0111] After determining the reference weight of the item, the upper and lower threshold coefficients of the commodity are determined using a scheme based on the tax number, and the normal weight range of each item is calculated. Some results are shown in Table 4. After obtaining the target declared weight of the target declared item, the following steps can be performed: 1) Arrange the historical declared weight sequence W (W1, W2, ..., W n ), sort them in ascending order, and perform the weight data preprocessing mentioned above. 2) Use the ostu algorithm to remove the historical declared weight sequence W (W1, W2, ..., W n ) to obtain the new normal weight sequence W inlier (W1,W2,…,W m ), calculate the weight sequence W inlier The number of times different weight values appear in count, weight sequence W inlier 3) If count = 1, the weight threshold is [0.7W ref , 1.3W ref, the threshold of 0.3 is due to the fact that the weight fluctuation range of most items is between 0 and 0.3. Due to the need to relax the threshold, the threshold is set to 0.3. 4) If count = 2, that is, the non-repeated values in the weight sequence are W1 and W2, calculate the ratio pct_value_W1 of W1 to W2, calculate the proportion pct_count_W1 of the number of times W1 appears in the weight sequence, and the tolerance range can be calculated according to the judgment conditions in the appendix Fig. 9 5) If count > 2, calculate the coefficient of variation cv, the first quartile q1, and the third quartile q3 of the weight sequence W inlier (W1, W2,..., W m ), and calculate the tolerance range according to the judgment conditions in the appendix Fig.10 .
[0112] In Fig. 9 the illustrated embodiment, in step S902, for the weight sequence W inlier (W1, W2), the ratio of W1 to W2 is pct_value_W1, and the proportion of the number of times W1 appears in the weight sequence is pct_count_W1.
[0113] In step S904, if pct_count_W1 ≥ 0.6, the tolerance range is [min(W1, (1 - α low )W ref ), (1 + α high )W ref .
[0114] In step S1006, if pct_count_W1 ≤ 0.4, or 0.4 < pct_count_W1 < 0.6 and pct_value_W1 < 0.5, the tolerance range is [(1 - α 1ow )W ref , max(W2, (1 + α high )W ref )].
[0115] In step S1008, if 0.4 < pct_count_W1 < 0.6 and pct_value_W1 ≥ 0.5, the tolerance range is [min(W1, (1 - α low )W ref ), max(W2, (1 + α high )W ref )].
[0116] In Fig.10 the illustrated embodiment, in step S1002, for the weight sequence W inlier (W1, W2,..., W m), the coefficient of variation is cv, the first quartile is q1, and the third quartile is q3.
[0117] In step S1004, if cv ≤ 0.1, the tolerance range is [min(W1, (1 - α low )W ref ), max(W m , (1 + α high )W ref )].
[0118] In step S1006, if cv > 0.3, the tolerance range is [(1 - α low )W ref , (1 + α high )W ref .
[0119] In step S1008, if 0.1 < cv ≤ 0.3, the tolerance range is [min(q1, (1 - α low )W ref ), max(q3, (1 + α high )W ref )].
[0120] The method for abnormal mining of declared weight in the embodiments of the present disclosure determines the reference weight of an item and the normal weight fluctuation range of the item by analyzing the declared data of historical customs declarations, and establishes the information of the item weight standard library, so as to integrate the declared information of items accumulated in historical customs declarations. And according to the information of the weight standard library, combined with the abnormal mining technology, it analyzes whether there are abnormalities in the declared weight sequences of various items, which items are the key categories with abnormalities, and which need to be key controlled, etc., so as to better improve the efficiency of express clearance operations.
[0121] Introducing the concept and technology of abnormal mining into the field of anti-express clearance fraud, and researching the method for abnormal mining of express weight. It can discover goods that may be falsely reported in advance after the declaration form is filled out and before the manual customs inspection is executed, effectively solving the problems of low efficiency and random inspection of manual inspection, and is an important measure to ensure express clearance.
[0122] The embodiments of the present disclosure propose a method for abnormal mining of declared weight, which determines the weight reference value of each item from the historical declared weight of express deliveries, constructs an item weight standard library, and provides decision support for express clearance. An adaptive weight threshold calculation method is proposed to avoid errors or situations inconsistent with the actual situation when using the same threshold when the weight differences between declared items are large.
[0123] It should be clearly understood that the present disclosure describes how to form and use specific examples, but the principles of the present disclosure are not limited to any details of these examples. On the contrary, based on the teachings of the contents disclosed in the present disclosure, these principles can be applied to many other embodiments.
[0124] Those skilled in the art will appreciate that all or part of the steps for implementing the above embodiments are implemented as a computer program executed by a central processing unit (CPU). When the computer program is executed by the central processing unit (CPU), the above functions defined by the above method provided by the present disclosure are executed. The program may be stored in a computer-readable storage medium, which may be a read-only memory, a disk or an optical disk, etc.
[0125] In addition, it should be noted that the above figures are only schematic illustrations of the processes included in the method according to the exemplary embodiments of the present disclosure, and are not intended to be limiting. It is easy to understand that the processes shown in the above figures do not indicate or limit the time sequence of these processes. In addition, it is also easy to understand that these processes can be performed synchronously or asynchronously, for example, in multiple modules.
[0126] The following are embodiments of the device disclosed herein, which can be used to execute the method embodiments disclosed herein. For details not disclosed in the device embodiments disclosed herein, please refer to the method embodiments disclosed herein.
[0127] Fig.11 is a block diagram of an abnormal excavation device for declaring weight according to an exemplary embodiment. Fig.11 The device 1100 for mining anomalies in declared weight provided in the embodiment of the present disclosure may include: a weight acquisition module 1102 , a reference weight module 1104 , a tolerance range module 1106 and an anomaly alarm module 1108 .
[0128] In the device 1100 for mining anomalies in declared weight, the weight acquisition module 1102 may be configured to acquire a target declared weight of a target declared item.
[0129] The reference weight module 1104 may be configured to determine a reference weight of the target declared item according to a historical declared weight sequence of the target declared item.
[0130] The tolerance range module 1106 may be configured to determine a tolerance range corresponding to a reference weight of a target declared item according to a preset data set.
[0131] The abnormal alarm module 1108 may be configured to generate an alarm prompt message according to the target declared item if it is determined that the target declared weight of the target declared item is abnormal according to the reference weight and tolerance range of the target declared item.
[0132] According to the device for mining the abnormality of declared weight provided by the embodiment of the present disclosure, the reference weight of the target declared item is determined according to the historical declared weight sequence of the target declared item. The reference weight value of each item can be determined based on the historical declared data, and a standard library of item weight can be constructed to provide decision support for express customs clearance, thereby avoiding the weight confusion caused by different historical weight sequences of the same item. The tolerance range corresponding to the reference weight of the target declared item is determined according to the preset data set, and the weight characteristics of different items can be comprehensively considered. The tolerance coefficient with adaptive characteristics for different weight intervals is constructed based on the preset data set to generate a normal tolerance range for the weight of each item, thereby avoiding the situation that when the weight difference between the declared items is large, the same threshold will have errors or cause inconsistency with the actual situation. When the target declared weight of the target declared item is determined to be abnormal according to the reference weight and tolerance range of the target declared item, an alarm prompt information is generated according to the target declared item to improve the efficiency of express customs clearance operations.
[0133] Refer to the following Fig.12 The electronic device 1200 according to this embodiment of the present invention is described. Fig.12 The electronic device 1200 shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present invention.
[0134] like Fig.12 As shown, the electronic device 1200 is in the form of a general computing device. The components of the electronic device 1200 may include but are not limited to: at least one processing unit 1210, at least one storage unit 1220, and a bus 1230 connecting different system components (including the storage unit 1220 and the processing unit 1210).
[0135] The storage unit stores program codes, which can be executed by the processing unit 1210, so that the processing unit 1210 performs the steps according to various exemplary embodiments of the present invention described in the above “Exemplary Method” section of this specification. For example, the processing unit 1210 can perform the following steps: Figure 1 In step S202 shown in the figure, the target declared weight of the target declared item is obtained; step S204: the reference weight of the target declared item is determined according to the historical declared weight sequence of the target declared item; step S206: the tolerance range corresponding to the reference weight of the target declared item is determined according to the preset data set; step S208: if the target declared weight of the target declared item is abnormal according to the reference weight and the tolerance range of the target declared item, an alarm prompt information is generated according to the target declared item.
[0136] The storage unit 1220 may include a readable medium in the form of a volatile storage unit, such as a random access storage unit (RAM) 12201 and / or a cache storage unit 12202 , and may further include a read-only storage unit (ROM) 12203 .
[0137] The storage unit 1220 may also include a program / utility 12204 having a set (at least one) of program modules 12205, such program modules 12205 including but not limited to: an operating system, one or more application programs, other program modules, and program data, each of which or some combination may include an implementation of a network environment.
[0138] Bus 1230 may represent one or more of several types of bus structures, including a memory unit bus or memory unit controller, a peripheral bus, an accelerated graphics port, a processing unit, or a local bus using any of a variety of bus architectures.
[0139] The electronic device 1200 may also communicate with one or more external devices 1300 (e.g., keyboards, pointing devices, Bluetooth devices, etc.), may also communicate with one or more devices that enable a user to interact with the electronic device 1200, and / or communicate with any device that enables the electronic device 1200 to communicate with one or more other computing devices (e.g., routers, modems, etc.). Such communication may be performed via an input / output (I / O) interface 1250. In addition, the electronic device 1200 may also communicate with one or more networks (e.g., local area networks (LANs), wide area networks (WANs), and / or public networks, such as the Internet) via a network adapter 1260. As shown, the network adapter 1260 communicates with other modules of the electronic device 1200 via a bus 1230. It should be understood that, although not shown in the figure, other hardware and / or software modules may be used in conjunction with the electronic device 1200, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems.
[0140] Through the description of the above implementation, it is easy for those skilled in the art to understand that the example implementation described here can be implemented by software, or by software combined with necessary hardware. Therefore, the technical solution according to the implementation of the present disclosure can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (which can be a CD-ROM, a USB flash drive, a mobile hard disk, etc.) or on a network, including several instructions to enable a computing device (which can be a personal computer, a server, a terminal device, or a network device, etc.) to execute the method according to the implementation of the present disclosure.
[0141] In an exemplary embodiment of the present disclosure, a computer-readable storage medium is also provided, on which a program product capable of implementing the above method of the present specification is stored. In some possible implementations, various aspects of the present invention can also be implemented in the form of a program product, which includes a program code, and when the program product is run on a terminal device, the program code is used to enable the terminal device to execute the steps according to various exemplary embodiments of the present invention described in the above "Exemplary Method" section of the present specification.
[0142] The program product may use any combination of one or more readable media. The readable medium may be a readable signal medium or a readable storage medium. The readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples (non-exhaustive list) of readable storage media include: an electrical connection with one or more wires, a portable disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above.
[0143] Computer readable signal media may include data signals propagated in baseband or as part of a carrier wave, in which readable program code is carried. Such propagated data signals may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. Readable signal media may also be any readable medium other than a readable storage medium, which may send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, apparatus, or device.
[0144] The program code embodied on the readable medium may be transmitted using any appropriate medium, including but not limited to wireless, wired, optical cable, RF, etc., or any suitable combination of the foregoing.
[0145] Program code for performing the operations of the present invention may be written in any combination of one or more programming languages, including object-oriented programming languages such as Java, C++, etc., and conventional procedural programming languages such as "C" or similar programming languages. The program code may be executed entirely on the user computing device, partially on the user device, as a separate software package, partially on the user computing device and partially on a remote computing device, or entirely on a remote computing device or server. In cases involving a remote computing device, the remote computing device may be connected to the user computing device through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computing device (e.g., through the Internet using an Internet service provider).
[0146] In addition, the above-mentioned figures are only schematic illustrations of the processes included in the method according to an exemplary embodiment of the present invention, and are not intended to be limiting. It is easy to understand that the processes shown in the above-mentioned figures do not indicate or limit the time sequence of these processes. In addition, it is also easy to understand that these processes can be performed synchronously or asynchronously, for example, in multiple modules.
[0147] Those skilled in the art will readily appreciate other embodiments of the present disclosure after considering the specification and practicing the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of the present disclosure that follow the general principles of the present disclosure and include common knowledge or customary techniques in the art that are not disclosed in the present disclosure. The specification and examples are to be considered exemplary only, and the true scope and concept of the present disclosure are indicated by the claims.
Claims
1. A method for mining anomalies in declared weight, characterized in that: include: Get the target declared weight of the target declared items; Determine the reference weight of the target declared item according to the historical declared weight sequence of the target declared item; Determine the tolerance range corresponding to the reference weight of the target declared item according to the preset data set; If it is determined that the target declared weight of the target declared item is abnormal according to the reference weight of the target declared item and the tolerance range, an alarm prompt information is generated according to the target declared item; Among them, determining the tolerance range corresponding to the reference weight of the target declared item according to the preset data set includes: determining the tolerance coefficients of different weight intervals according to the preset data set, so as to determine the tolerance coefficient of the weight interval corresponding to the reference weight of the target declared item as the tolerance range corresponding to the reference weight of the target declared item.
2. The method according to claim 1, characterized in that Determining the reference weight of the target declared item according to the historical declared weight sequence of the target declared item includes: Determine the different weight values and the number of different weight values that appear in the historical declared weight sequence; The reference weight of the target declared item is determined according to the different weight values and the number of the different weight values.
3. The method according to claim 2, characterized in that Determining the reference weight of the target declared article according to the different weight values and the number of the different weight values includes: If the number of the different weight values is 1, determining the different weight values as the reference weights of the target declared items; If the number of the different weight values is 2, the reference weight of the target declared item is determined according to the proportion of the number of occurrences of the larger weight value and the smaller weight value among the different weight values in the historical declared weight sequence; If the number of different weight values is greater than or equal to 3, the reference weight of the target declared item is determined based on the coefficient of dispersion, the coefficient of skewness, the average value of the historical declared weight sequence, the larger weight value and the smaller weight value among the different weight values.
4. The method according to claim 3, characterized in that Determining the reference weight of the target declared article according to the dispersion coefficient, the skewness coefficient, the average value of the historical declared weight sequence, the larger weight value and the smaller weight value among the different weight values includes: If the dispersion coefficient is less than or equal to the dispersion coefficient threshold, the reference weight of the target declared item is determined according to the average value of the historical declared weight sequence; If the coefficient of dispersion is greater than the coefficient of dispersion threshold, the abnormal values of the historical declared weight sequence are eliminated according to the coefficient of dispersion, the skewness coefficient, the maximum weight value and the minimum weight value among the different weight values of the historical declared weight sequence, so as to determine the reference weight of the target declared item according to the average value of the historical declared weight sequence after elimination.
5. The method according to claim 1, characterized in that The tolerance range corresponding to the reference weight of the target declared item determined according to the preset data set includes: Determining a reference weight of each historical item in the preset data set; Divide the historical objects into boxes according to the reference weights of the historical objects to obtain weight boxes; Determine the target weight bin of the target declared item in the weight bin according to the reference weight of the target declared item; Determine an upper threshold coefficient and a lower threshold coefficient of the target weight bin; The tolerance range of the target declared item is determined according to the upper threshold coefficient and the lower threshold coefficient of the target weight bin and the reference weight of the target declared item.
6. The method according to claim 5, characterized in that Determining the upper threshold coefficient and the lower threshold coefficient of the target weight bin includes: Determine a first average floating degree of each historical object according to a reference weight of each historical object and a declared weight in a historical declared weight sequence of each historical object that is smaller than the reference weight; Determine a second average floating degree of each historical object according to a reference weight of each historical object and a declared weight in a sequence of historical declared weights of each historical object that is greater than or equal to the reference weight; Binning the historical items according to the first average floating degree of the historical items to obtain first floating degree bins; Binning the historical items according to the second average floating degree of the historical items to obtain second floating degree bins; Determine the ratio of each first floating degree bin by the ratio of the number of historical items falling into the target weight bin to the number of historical items in each first floating degree bin; Determine the ratio of each second floating degree bin by the ratio of the number of historical items falling into the target weight bin to the number of historical items in each second floating degree bin; Determine the upper threshold coefficient of the target weight bin according to the ratio of each first floating degree bin and the right interval value of each first floating degree bin; The lower threshold coefficient of the target weight bin is determined according to the proportion of each second floating degree bin and the right interval value of each second floating degree bin.
7. The method according to any one of claims 1 or 5, characterized in that: Also includes: Generate the preset data set according to the historical declaration events with the same tax number as the target declared item; or The preset data set is generated according to historical declaration events of the same brand as the target declared article.
8. An abnormal excavation device with declared weight, characterized in that: include: A weight acquisition module configured to acquire a target declared weight of a target declared item; A reference weight module, configured to determine a reference weight of the target declared item according to a historical declared weight sequence of the target declared item; A tolerance range module, configured to determine a tolerance range corresponding to a reference weight of the target declared item according to a preset data set; An abnormal alarm module, configured to generate an alarm prompt information according to the target declared item if it is determined that the target declared weight of the target declared item is abnormal according to the reference weight of the target declared item and the tolerance range; Among them, the tolerance range module is also used to: determine the tolerance coefficients of different weight intervals according to a preset data set, so as to determine the tolerance coefficient of the weight interval corresponding to the reference weight of the target declared item as the tolerance range corresponding to the reference weight of the target declared item.
9. An electronic device, characterized in that: include: one or more processors; A storage device for storing one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors implement the method according to any one of claims 1 to 7.
10. A computer readable medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the method according to any one of claims 1 to 7 is implemented.
