Distribution resource evaluation method, device, electronic device and storage medium

Abstract

The embodiments of the present application provide a method, device, electronic device and computer-readable storage medium for evaluating distribution resources, and relate to the field of Internet information processing. The method includes: obtaining the degree of accumulation of at least one evaluation dimension of multiple distribution resources in a first historical period and a second historical period respectively; determining a first positive sample and a first negative sample from multiple distribution resources, binning multiple distribution resources, and determining the sub-evaluation results of each distribution resource in the corresponding dimension according to the proportion of the first negative sample in each bin to the total number of first negative samples; for each distribution resource, obtaining the first evaluation result of the distribution resource according to the sub-evaluation results of the distribution resource in all evaluation dimensions. The embodiments of the present application can stratify distribution resources more reasonably, and are more in line with cognition and easier to be accepted by distribution resources, laying the foundation for later warning and behavior optimization of distribution resources.

Description

Technical Field

[0001] The present application relates to the field of Internet information processing technology. Specifically, the present application relates to a distribution resource evaluation method, device, electronic device and computer-readable storage medium. Background Art

[0002] With the rapid development of Internet technology, there are more and more Internet-based applications, such as food delivery applications and shopping applications. Based on these applications, users can get the items they need without leaving home. While these applications provide convenience to users, the security of distribution resources cannot be ignored.

[0003] Related technologies usually perform calculations based on the addition and subtraction rules of fixed behaviors, or directly apply the scorecard model for scoring.

[0004] 1) For the method of adding or subtracting points based on fixed behaviors, it is easy to have difficulty in characterizing the differentiation in actual application. The overall calculated scores are either too high or too low, making it difficult to put them into actual business use.

[0005] 2) As for the scoring method of directly applying the scorecard model, since the scorecard model is binned and counted according to the distribution of the data, it is easy to have a situation that is inconsistent with the public perception when the data is sparse. For example, in general perception, the score of distribution resources with a high rate of positive reviews should be higher than the score of distribution resources with a low rate of positive reviews, but when the scorecard model is actually calculated and applied, the opposite situation will occur. Summary of the invention

[0006] The embodiments of the present application provide a distribution resource evaluation method, device, electronic device, and computer-readable storage medium, which can solve the above-mentioned problems of the prior art. The technical solution is as follows:

[0007] According to one aspect of an embodiment of the present application, a method for evaluating distribution resources is provided, the method comprising:

[0008] Obtaining accumulation levels of at least one evaluation dimension of a plurality of distribution resources in a first historical period and a second historical period, respectively; wherein the first historical period is earlier than the second historical period;

[0009] Determining a first positive sample and a first negative sample from the multiple distribution resources, the first negative sample being a distribution resource whose accumulation degree of at least one negative evaluation dimension exceeds a preset degree in a second historical period, and the first positive sample being a distribution resource other than the first negative sample among the multiple distribution resources;

[0010] For each evaluation dimension, the multiple distribution resources are binned according to the accumulation degree of the evaluation dimension of the multiple distribution resources in the first historical period, and the sub-evaluation results of each first positive sample and each first negative sample in the evaluation dimension are determined according to the proportion of the first negative samples in each bin to the total number of first negative samples;

[0011] For each distribution resource, a first evaluation result of the distribution resource is obtained according to the sub-evaluation results of the distribution resource in all evaluation dimensions.

[0012] As an optional implementation, multiple distribution resources are divided into boxes, including:

[0013] Determine a plurality of bins, each bin corresponding to an accumulation degree interval of the evaluation dimension, and the accumulation degree intervals corresponding to the bins do not have any intersection and are singular;

[0014] Determine the target sub-box where each distribution resource is located according to the accumulation degree of each distribution resource in the evaluation dimension;

[0015] The accumulation degree of the distribution resources in the evaluation dimension is located in the accumulation degree interval corresponding to the target sub-box.

[0016] As an optional implementation, determining the sub-evaluation results of each first positive sample and each first negative sample in the evaluation dimension according to the proportion of the first negative samples in each bin to the total number of first negative samples includes:

[0017] Each sub-box is traversed in turn according to a preset order. For the currently traversed sub-box, the sub-evaluation result of each distribution resource in the sub-box in the evaluation dimension is determined according to the ratio of all first negative samples in the currently traversed sub-box and the traversed sub-boxes to the total number of first negative samples.

[0018] As an optional implementation manner, obtaining the first evaluation result of the distribution resource according to the sub-evaluation results of the distribution resource in all evaluation dimensions includes:

[0019] Get the weight of each evaluation dimension;

[0020] A weighted sum is performed based on the sub-evaluation results of the distribution resource in all evaluation dimensions and the corresponding weights to obtain a first evaluation result of the distribution resource.

[0021] As an optional implementation, the step of obtaining the weight of each evaluation dimension also includes the step of determining the weight of each evaluation dimension:

[0022] Obtaining the accumulation degree of at least one evaluation dimension of multiple sample delivery resources in a third historical period and a fourth historical period respectively; wherein the third historical period is earlier than the fourth historical period;

[0023] Determine a second positive sample and a second negative sample from a plurality of sample delivery resources, wherein the second negative sample is a delivery resource for which the accumulation degree of at least one negative evaluation dimension exceeds a preset degree in a fourth historical period, and the second positive sample is a delivery resource other than the first negative sample among the plurality of sample delivery resources;

[0024] For each evaluation dimension, determine the sub-evaluation results of each second positive sample and each second negative sample in the evaluation dimension, and for each sample distribution resource, perform weighted summation according to the sub-evaluation results of the sample distribution resource in all evaluation dimensions and the corresponding initial weights to obtain the first sample evaluation result of the sample distribution resource;

[0025] The initial weight is adjusted according to the difference between the first sample evaluation result and the target value until the difference converges to obtain the weight; wherein, when the sample distribution resource is a positive sample, the target value is a first target value, and when the sample distribution resource is a negative sample, the target value is a second target value.

[0026] As an optional implementation, the evaluation method further includes:

[0027] Obtaining accumulation levels of at least one safety dimension of each of the plurality of distribution resources in the first historical period;

[0028] For each distribution resource, obtaining a second evaluation result of the distribution resource according to the accumulation degree of at least one safety dimension of the distribution resource in the first historical period;

[0029] A comprehensive evaluation result of the distribution resource is obtained according to the first evaluation result and the second evaluation result of the distribution resource.

[0030] As an optional implementation, the evaluation method further includes:

[0031] An alarm message is sent to the distribution resource whose at least one evaluation result among the first evaluation result, the second evaluation result or the comprehensive evaluation result meets the preset conditions.

[0032] According to another aspect of an embodiment of the present application, a distribution resource evaluation device is provided, the device comprising:

[0033] An information acquisition module is used to acquire the accumulation degree of at least one evaluation dimension of multiple distribution resources in a first historical period and a second historical period respectively; wherein the first historical period is earlier than the second historical period;

[0034] A positive and negative sample determination module is used to determine a first positive sample and a first negative sample from the multiple distribution resources, wherein the first negative sample is a distribution resource whose accumulation degree of at least one negative evaluation dimension exceeds a preset degree in a second historical period, and the first positive sample is a distribution resource other than the first negative sample among the multiple distribution resources;

[0035] A sub-evaluation result module is used to, for each evaluation dimension, bin the multiple distribution resources according to the accumulation degree of the evaluation dimension of the multiple distribution resources in the first historical period, and determine the sub-evaluation results of each first positive sample and each first negative sample in the evaluation dimension according to the proportion of the first negative samples in each bin to the total number of first negative samples;

[0036] The first result acquisition module is used to obtain, for each distribution resource, a first evaluation result of the distribution resource according to the sub-evaluation results of the distribution resource in all evaluation dimensions.

[0037] According to another aspect of an embodiment of the present application, an electronic device is provided, which includes: a memory, a processor, and a computer program stored in the memory, and the processor executes the computer program to implement the distribution resource evaluation method of the above aspect.

[0038] According to another aspect of the embodiments of the present application, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the distribution resource evaluation method according to the above aspect is implemented.

[0039] The beneficial effects of the technical solution provided by the embodiment of the present application are:

[0040] By obtaining the accumulation degree of at least one evaluation dimension of multiple distribution resources in the first historical period and the second historical period respectively, and using the accumulation degree of the evaluation dimension in the later second historical period to determine the first positive sample and the negative sample, a finer granularity of data division is achieved, laying a foundation for more accurate evaluation of distribution resources. For each evaluation dimension, the accumulation degree of the evaluation dimension of the multiple distribution resources in the first historical period is used to bin the multiple distribution resources, and the sub-evaluation results of each first positive sample and each first negative sample in the evaluation dimension are determined according to the proportion of the first negative samples in each bin to the total number of first negative samples. Compared with the existing technology, the problem of missing risk correlation can be solved, and the self-evaluation result corresponding to each bin can be monotonic corresponding to the accumulation degree, thereby more reasonably stratifying the distribution resources, and because it is more in line with cognition, it is also easier to be accepted by the distribution resources, laying a foundation for later warning and behavior optimization of the distribution resources. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in describing the embodiments of the present application are briefly introduced below.

[0042] Figure 1 A schematic diagram of the evaluation results of a certain evaluation dimension determined for the relevant technology;

[0043] Figure 2 A schematic diagram of the architecture of an evaluation system for implementing configuration resources provided in an embodiment of the present application;

[0044] Figure 3 A schematic diagram of a flow chart of a distribution resource evaluation method provided in an embodiment of the present application;

[0045] Figure 4 A schematic diagram of an evaluation interface of a target application provided in an embodiment of the present application;

[0046] Figure 5 A flowchart of an evaluation method according to another embodiment of the present application;

[0047] Figure 6 A schematic diagram of the structure of a distribution resource evaluation device provided in an embodiment of the present application;

[0048] Figure 7 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0049] The embodiments of the present application are described below in conjunction with the drawings in the present application. It should be understood that the implementation methods described below in conjunction with the drawings are exemplary descriptions for explaining the technical solutions of the embodiments of the present application and do not constitute a limitation on the technical solutions of the embodiments of the present application.

[0050] It will be understood by those skilled in the art that, unless specifically stated, the singular forms "one", "an" and "the" used herein may also include plural forms. It should be further understood that the terms "including" and "comprising" used in the embodiments of the present application refer to that the corresponding features can be implemented as the presented features, information, data, steps, operations, elements and / or components, but do not exclude the implementation as other features, information, data, steps, operations, elements, components and / or combinations thereof supported by the technical field. It should be understood that when we call an element "connected" or "coupled" to another element, the one element can be directly connected or coupled to the other element, or it can refer to that the one element and the other element establish a connection relationship through an intermediate element. In addition, the "connection" or "coupling" used here may include wireless connection or wireless coupling. The term "and / or" used here indicates at least one of the items defined by the term, for example, "A and / or B" can be implemented as "A", or as "B", or as "A and B".

[0051] In order to make the objectives, technical solutions and advantages of the present application clearer, the implementation methods of the present application will be further described in detail below with reference to the accompanying drawings.

[0052] First, several terms involved in this application are introduced and explained:

[0053] 1) Delivery resources: resources required for delivery. Common delivery resources include riders, delivery drones, etc.

[0054] 2) Scorecard model: Based on the various attributes and behavioral data of the evaluation object, the credit scoring model is used to score the customer's credit, so as to determine whether there is risk, the degree of risk, etc. In scorecard modeling, binning is a name for the discretization of continuous variables.

[0055] An important parameter in the scorecard model is the weight of evidence (WOE), which is calculated for each binning result. Assuming good is a positive sample and bad is a negative sample, there exists: WOE i =ln(good proportion / bad proportion), where the good proportion can be determined by the ratio of the number of good evaluation objects in the i-th bin to the total number of good evaluation objects, and the bad proportion can be determined by the ratio of the number of bad evaluation objects in the i-th bin to the total number of bad evaluation objects.

[0056] In the field of distribution resource evaluation, whether a distribution resource is a positive sample or a negative sample can be determined based on the degree of negative evaluation received by the distribution resource within a certain period of time. For example, if a distribution resource receives one negative evaluation in the last month, it is considered a negative sample.

[0057] In the field of distribution resource evaluation, different bins correspond to different accumulation levels of evaluation dimensions. Generally, the accumulation level can be expressed in percentage. For example, the evaluation dimension: packaging integrity, its accumulation level is 97%, which means that 97% of the distribution process packaging integrity is qualified. Then for this evaluation dimension, if it is divided into 3 bins, bin 1 is below 95%, bin 2 is 95-98%, and bin 3 is above 98%. When the distribution resource is in bin 3, it means that the evaluation result of the distribution resource in the evaluation dimension of packaging integrity is relatively the highest. Therefore, it can be seen that the numerical change of the bin should be synchronized with the change of the size of the evaluation result WOE value.

[0058] But in general, negative reviews of delivery resources are rare. Some statistics show that less than 3% of delivery resources will receive one or more negative reviews in a month. In this case, there is a high probability that there is a risk correlation problem.

[0059] like Figure 1 As shown, it exemplarily shows a schematic diagram of the evaluation results of a certain evaluation dimension determined by the relevant technology, in which the horizontal axis represents the bins and the vertical axis represents the WOE values ​​of the corresponding bins. It can be found from the figure that the WOE values ​​do not show a monotonically decreasing trend. The WOE values ​​of bins 0 to 2 show a trend of first increasing and then decreasing, and the WOE values ​​of bins 2 to 4 show a trend of first increasing and then decreasing, and so on. The WOE values ​​of all bins have rebounded many times, so the evaluation results obtained based on the existing model do not conform to the public's perception.

[0060] The distribution resource evaluation method, device, electronic device, computer-readable storage medium and computer program product provided in this application are intended to solve the above technical problems in the prior art.

[0061] The following describes several exemplary embodiments to illustrate the technical solutions of the embodiments of the present application and the technical effects produced by the technical solutions of the present application. It should be noted that the following embodiments can refer to, draw on or combine with each other, and the same terms, similar features and similar implementation steps in different embodiments will not be described repeatedly.

[0062] Figure 2 The schematic diagram of the architecture of the evaluation system for implementing the configuration resources provided in the embodiment of the present application includes a terminal 101 and a server 102 , and the terminal 101 is communicatively connected with the server 102 via a network 103 .

[0063] In this embodiment, the terminal 101 may be any type of mobile computing device, including a mobile computer (e.g., a personal digital assistant (PDA), a laptop computer, a notebook computer, such as a tablet computer, a netbook, etc.), a mobile phone (e.g., a cellular phone), a wearable computing device (e.g., a smart watch, a head-mounted device, including smart glasses), or other types of mobile devices. In some embodiments, the terminal 101 may also be a fixed computing device, such as a desktop computer, a game console, a smart TV, etc.

[0064] The server 102 can store and run instructions that can execute the various methods described herein, and can be a single server or a server cluster or a cloud server, respectively, or any two or three of them can be the same server or the same server cluster or cloud server. It should be understood that the server mentioned herein is typically a server computer with a large amount of memory and processor resources, but other embodiments are also possible.

[0065] Examples of network 103 include a local area network (LAN), a wide area network (WAN), a personal area network (PAN), and / or a combination of communication networks such as the Internet. Each of server 102 and terminal 101 may include at least one communication interface (not shown) capable of communicating through network 103. Such a communication interface may be one or more of the following: any type of network interface (e.g., a network interface card (NIC)), a wired or wireless (such as IEEE 802.11 wireless LAN (WLAN)) wireless interface, a Worldwide Interoperability for Microwave Access (Wi-MAX) interface, an Ethernet interface, a Universal Serial Bus (USB) interface, a cellular network interface, a Bluetooth™ interface, a near field communication (NFC) interface, etc. Additional examples of communication interfaces are described elsewhere herein.

[0066] The terminal of the embodiment of the present application can report one or more evaluation dimension information of the distribution resources to the server, so that the server can execute the distribution resource evaluation method of the embodiment of the present application by collecting the information reported by the terminal and obtain the evaluation result of the distribution resources.

[0067] The present application provides a method for evaluating distribution resources. Figure 3 As shown, the method includes:

[0068] S101. Obtain the accumulation degree of at least one evaluation dimension of a plurality of distribution resources in a first historical period and a second historical period, respectively; wherein the first historical period is earlier than the second historical period.

[0069] The embodiment of the present application obtains, for each distribution resource, distribution information of the distribution resource in two historical periods, namely, a first historical period and a second historical period, wherein the first historical period is earlier than the second historical period, for example, the second historical period is from May 1 to May 31, and the first historical period is from April 1 to April 30.

[0070] The idea of ​​the embodiment of the present application is to predict the possibility of future behavior performance by the accumulation degree of each evaluation dimension. Therefore, a fixed moment is used for each distribution resource, and the moment is used as the dividing line. The period before the dividing line, that is, the performance of the first historical period (the accumulation degree of the evaluation dimension) is used as the feature, and the period after the dividing line, that is, the performance of the second historical period, is used as the label.

[0071] It should be noted that the embodiment of the present application does not specifically limit the time span of the first historical period and the second historical period, for example, it can be one month, one quarter, half a year, etc. The time span of the first historical period and the second historical period can be the same or different. The embodiment of the present application obtains the accumulation degree of the two historical periods respectively, uses the relatively fresher data of the second historical period to determine the positive and negative samples, and uses the relatively older data of the first historical period to calculate the evaluation results, so as to realize the data division of labor in a finer granularity and lay the foundation for more accurate evaluation of distribution resources.

[0072] The delivery information of the embodiment of the present application includes the accumulation degree of at least one evaluation dimension. The evaluation dimension of the embodiment of the present application refers to the dimension in which shops or shoppers evaluate the delivery behavior of delivery resources. Generally, the evaluation dimensions may include overall evaluation, packaging conditions, delivery timeliness, communication conditions, etc. The accumulation degree of the evaluation dimension, that is, the accumulation of delivery resources in the first / second historical period. The accumulation degree of the embodiment of the present application can be a normalized result, thereby eliminating the dimensional influence between indicators. For example, a certain delivery resource received 100 evaluations of the evaluation dimension of "packaging conditions" within 3 months, 2 of which were damaged packaging. Therefore, the accumulation degree of the evaluation dimension of "packaging conditions" for this delivery resource is 98%.

[0073] S102. Determine a first positive sample and a first negative sample from a plurality of distribution resources, wherein the first negative sample is a distribution resource whose accumulation degree of at least one negative evaluation dimension exceeds a preset degree in a second historical period, and the first positive sample is a distribution resource among the plurality of distribution resources except the first negative sample.

[0074] The present application distinguishes the first positive sample from the first negative sample based on the accumulation degree of at least one evaluation dimension of each distribution resource in the second historical period.

[0075] The evaluation dimensions of the embodiment of the present application include positive evaluation dimensions and negative evaluation dimensions. When the delivery resource completes the delivery, the store manager or shopper can evaluate the delivery resource through the target application. Specifically, the target application can be an application that provides delivery services. The store can publish goods through the target application. Shoppers can purchase published goods on the target application. When shoppers purchase goods, the background of the target application can specify at least one delivery resource for delivery. After the delivery is completed, the store manager and the shopper can evaluate the delivery resource.

[0076] See also Figure 4 , which exemplarily shows a schematic diagram of the evaluation interface of the target application of the embodiment of the present application, the interface displays multiple evaluation dimension controls, when the store manager or shopper triggers one of the controls, it is considered that the store manager or shopper believes that the delivery resources meet the evaluation dimension corresponding to the control. As shown in the figure, "satisfied" is a positive evaluation dimension, "unsatisfied" is a negative evaluation dimension, "smooth communication" is a positive evaluation dimension, "unsmooth communication" is a negative evaluation dimension, etc., "intact packaging" is a positive evaluation dimension, and "damaged packaging" is a negative evaluation dimension.

[0077] In the embodiment of the present application, the distribution resources whose accumulation degree of at least one negative evaluation dimension exceeds the preset degree in the second historical period are used as the first negative samples. For example, when a distribution resource receives one or more "unsatisfactory" evaluations in the past month (the second historical period), the distribution information is used as the first negative sample. It should be understood that after all the first negative samples are determined, the distribution resources other than the first negative samples in the multiple distribution resources naturally belong to positive samples. For example, if there are 10 distribution resources, namely distribution resources 1 to 10, if distribution resources 2, 9 and 10 are determined to be the first negative samples, then distribution resources 1, 3 to 8 are the first positive samples.

[0078] S103. For each evaluation dimension, the multiple distribution resources are binned according to the accumulation degree of the evaluation dimension of the multiple distribution resources in the first historical period, and the sub-evaluation results of each first positive sample and each first negative sample in the evaluation dimension are determined according to the ratio of the first negative samples in each bin to the total number of first negative samples.

[0079] This application draws on the binning process of the existing scoring card model. For each evaluation dimension, the embodiments of this application need to perform binning. The binning process is to determine the bin where each distribution resource is located according to the accumulation degree of the evaluation dimension of each distribution resource in the first period of time. Each bin corresponds to an accumulation degree interval, that is, for one evaluation dimension, a distribution resource can only be located in a unique bin.

[0080] After obtaining the binning results, the embodiment of the present application is different from the prior art in which WOE is calculated based on the proportion of positive samples and the proportion of negative samples. Instead, only the proportion of negative samples is considered (that is, the proportion of the first negative sample in a bin to the total number of first negative samples). After verification, the embodiment of the present application can overcome the problem of lack of risk correlation in the prior art. The present application determines the sub-evaluation result of each distribution resource corresponding to the dimension based on the proportion of negative samples.

[0081] It should be noted that there are two main reasons for selecting the negative sample ratio in the embodiment of the present application:

[0082] 1. Ensure that there is still strong monotonicity after binning, so even if some data is missing, it is acceptable, which mainly solves the problem of imbalanced distribution;

[0083] 2. In actual application, the number of negative samples is actually not small. Taking a delivery resource with 500,000 daily active users as an example, even if the negative evaluation rate is 0.2%, there are 1,000 samples. In addition, the embodiment of the present application can cycle sampling according to the date, so there is no need to worry about the number of samples.

[0084] S104: For each distribution resource, obtain a first evaluation result of the distribution resource according to the sub-evaluation results of the distribution resource in all evaluation dimensions.

[0085] Since the embodiment of the present application obtains the accumulation degree of multiple evaluation dimensions, after obtaining the sub-evaluation result of each dimension corresponding to each distribution resource through step S103, the first evaluation result can be obtained by comprehensively considering the sub-evaluation results of all dimensions.

[0086] Specifically, the embodiments of the present application may accumulate the sub-evaluation results of all evaluation dimensions and use the accumulated result as the first evaluation result, or may average the sub-evaluation results of all evaluation dimensions and use the average value as the first evaluation result. The embodiments of the present application do not make specific limitations.

[0087] The distribution resource evaluation method of the embodiment of the present application obtains the accumulation degree of at least one evaluation dimension of multiple distribution resources in the first historical period and the second historical period respectively, and determines the first positive sample and the negative sample by using the accumulation degree of the evaluation dimension in the later second historical period, thereby realizing data division in a finer granularity and laying a foundation for more accurate evaluation of distribution resources. For each evaluation dimension, the accumulation degree of the evaluation dimension of the multiple distribution resources in the first historical period is used to divide the multiple distribution resources into boxes, and the sub-evaluation results of each first positive sample and each first negative sample in the evaluation dimension are determined according to the proportion of the first negative samples in each box to the total number of the first negative samples. Compared with the prior art, the problem of missing risk correlation can be solved, and the self-evaluation result corresponding to each box can be monotonic with respect to the accumulation degree, thereby more reasonably stratifying the distribution resources. Moreover, since it is more in line with cognition, it is also more easily accepted by the distribution resources, laying a foundation for later warning and behavior optimization of the distribution resources.

[0088] Based on the above embodiments, as an optional embodiment, binning is performed on the accumulation degree of the evaluation dimension of each first positive sample and each first negative sample in the first historical period, including:

[0089] S201, determining a plurality of sub-boxes, each sub-box corresponding to an accumulation degree interval of the evaluation dimension, and the accumulation degree intervals corresponding to the sub-boxes have no intersection and are singular;

[0090] S202: Determine the target sub-box where each distribution resource is located according to the accumulation degree of each distribution resource in the evaluation dimension.

[0091] The embodiment of the present application can specifically determine multiple sub-boxes according to the distribution of the accumulation degree of all delivery resources in the evaluation dimension, for example,

[0092] There is no intersection between the accumulation degree intervals corresponding to the various sub-boxes in the embodiment of the present application, so that one distribution resource is prevented from being placed in multiple sub-boxes, and there is a singleness between the accumulation degree intervals corresponding to the sub-boxes, that is, the accumulation degree intervals corresponding to the sub-boxes are either getting smaller and smaller, that is, the first accumulation degree interval includes the maximum value of the accumulation degree, and the last accumulation degree interval includes the minimum value of the accumulation degree. Of course, the accumulation degree intervals corresponding to the sub-boxes can also be getting larger and larger. The specific number of sub-boxes corresponding to the embodiment of the present application is not limited and can be set according to actual conditions.

[0093] For example, if there are five distribution resources (distribution resources 1 to 5) in a certain evaluation dimension, and their accumulation levels are 0.01%, 0.05%, 0.12%, 0.08% and 0.14% respectively, three sub-boxes can be set. If they are set in a monotonically increasing manner, the accumulation level interval of sub-box 1 is [0.01%, 0.05%), the accumulation level interval of sub-box 2 is [0.05%, 0.10%), and the accumulation level interval of sub-box 3 is [0.10%, 0.15%). Accordingly, distribution resource 1 is in sub-box 1, distribution resources 2 and 4 are in sub-box 2, and distribution resources 3 and 5 are in sub-box 3.

[0094] Based on the above embodiments, as an optional embodiment, according to the proportion of the first negative samples in each bin to the total number of the first negative samples, the sub-evaluation results of each first positive sample and each first negative sample in the evaluation dimension are determined, including:

[0095] Each sub-box is traversed in turn according to a preset order. For the currently traversed sub-box, the sub-evaluation result of each distribution resource in the sub-box in the evaluation dimension is determined according to the ratio of all first negative samples in the currently traversed sub-box and the traversed sub-boxes to the total number of first negative samples.

[0096] Since the accumulation degree intervals corresponding to the bins in the embodiment of the present application are monotonic, when calculating the sub-evaluation results of the distribution resources in each bin, the sub-evaluation results corresponding to each bin can be monotonic by integrating and traversing the negative sample ratios in the bins. The embodiment of the present application can well solve the problem of risk correlation loss caused by data sparsity while having distinguishability.

[0097] Specifically, the following formula can be used to calculate the sub-evaluation result WOE of the distribution resource in the i-th sub-box: i :

[0098]

[0099] Among them, B k represents the number of the first negative samples in the kth bin, B T It is the total number of first negative samples in all configured resources.

[0100] The above formula is due to It must not be greater than 1, so the value of WOE is always less than 0, and the later the sub-box number is, the greater the sub-evaluation result of the distribution resource in the sub-box is. The embodiment of the present application does not limit the specific formula for calculating the sub-evaluation result, and other monotonic functions can also be used for calculation.

[0101] Based on the above embodiments, as an optional embodiment, obtaining a first evaluation result of the distribution resource according to the sub-evaluation results of the distribution resource in all evaluation dimensions includes:

[0102] Get the weight of each evaluation dimension;

[0103] A weighted sum is performed based on the sub-evaluation results of the distribution resource in all evaluation dimensions and the corresponding weights to obtain a first evaluation result of the distribution resource.

[0104] The embodiment of the present application can obtain the weight of each evaluation dimension by machine learning, so as to more accurately determine the first evaluation result of the distribution resource. Specifically, the step of determining the weight of each evaluation dimension in the embodiment of the present application includes:

[0105] S301. Obtain the accumulation degree of at least one evaluation dimension of multiple sample distribution resources in a third historical period and a fourth historical period respectively; the third historical period is earlier than the fourth historical period.

[0106] S302. Determine a second positive sample and a second negative sample from a plurality of sample distribution resources, wherein the second negative sample is a distribution resource in which the accumulation degree of at least one negative evaluation dimension exceeds a preset degree in a fourth historical period, and the second positive sample is a distribution resource among the plurality of sample distribution resources except the first negative sample.

[0107] Step S302 is consistent with the concept of the above embodiments, and determines each second positive sample and each second negative sample from the sample distribution resources based on the accumulation degree of the negative evaluation dimension in the fourth historical period.

[0108] S303. For each evaluation dimension, determine the sub-evaluation results of each second positive sample and each second negative sample in the evaluation dimension. For each sample distribution resource, perform weighted summation according to the sub-evaluation results of the sample distribution resource in all evaluation dimensions and the corresponding initial weights to obtain the first sample evaluation result of the sample distribution resource.

[0109] S304. Adjust the initial weight according to the difference between the first sample evaluation result and the target value until the difference converges to obtain the weight; wherein, when the sample distribution resource is a positive sample, the target value is a first target value, and when the sample distribution resource is a negative sample, the target value is a second target value.

[0110] Since the positive and negative samples indicate that the distribution resources have more negative evaluations in the fourth historical period, the present application also hopes that the difference between the sub-evaluation results of the positive and negative samples is significant when calculating the sub-evaluation results of the positive and negative samples. Therefore, two target values ​​with large differences are set for the second positive sample and the second negative sample respectively. In this way, for the first sample evaluation result obtained after weighted summation using the initial weights, if the first sample evaluation result is a positive sample, the initial weights are continuously adjusted through iterative operations so that the first sample evaluation result is closer to the first target value. For the first sample evaluation result of the negative sample, the initial weights are continuously adjusted through iterative operations so that the first sample evaluation result is closer to the second target value.

[0111] For example, there are three delivery resources, and the accumulation degree of the two evaluation dimensions (good review rate and bad review rate) of each delivery resource is as follows:

[0112] Delivery resource 1 (negative sample), label: 0, features: negative review rate 0.1, positive review rate 0.1;

[0113] Delivery resource 2 (positive sample), label: 1, features: negative review rate 0.01, positive review rate 0.2;

[0114] Delivery resource 3 (positive sample), label: 1, features: negative review rate 0.02, positive review rate 0.3;

[0115] Assuming that the bins have been divided, the samples after binning and mapping the WOE values ​​are as follows (the bad rating is the sub-evaluation result of the bad rating rate, and the good rating is the sub-evaluation result of the bad and good rating rate):

[0116] Delivery resource 1 (negative sample), label: 0, mapping features: bad score -0.3, good score 0.01;

[0117] Delivery resource 2 (positive sample), label: 1, mapping features: bad score -0.01, good score 0.2;

[0118] Delivery resource 3 (positive sample), label: 1, mapping features: bad score -0.1, good score 0.5;

[0119] For distribution resource 1, the first sample evaluation result is: sigmoid(-0.3*a+0.01*b)

[0120] For distribution resource 2, the first sample evaluation result is: sigmoid(-0.01*a+0.2*b)

[0121] For distribution resource 3, the first sample evaluation result is: sigmoid(-0.1*a+0.5*b)

[0122] At this time, the values ​​of a and b are uncertain. We need to make the first sample evaluation result of distribution resource 1 closer to 0 (the second target value), and make the first sample evaluation results of distribution resources 2 and 3 closer to 1 (the first target value). In this way, there is a learning process, and then we can get the values ​​of a and b, which are weights.

[0123] Based on the above embodiments, as an optional embodiment, the embodiment of the present application also includes:

[0124] S401, obtaining the accumulation degree of at least one security dimension of the plurality of distribution resources in the first historical period;

[0125] S402: for each distribution resource, obtaining a second evaluation result of the distribution resource according to the accumulation degree of at least one security dimension of the distribution resource in the first historical period;

[0126] S403: Obtain a comprehensive evaluation result of the distribution resource according to the first evaluation result and the second evaluation result of the distribution resource.

[0127] The security dimension of the embodiment of the present application can be the dimension of security incidents reported by shops, shoppers and / or public institutions. Common security dimensions include public security incidents (such as threats, insults, physical conflicts, etc.) and medical incidents (traffic accidents, medical treatment caused by physical conflicts, etc.). The embodiment of the present application can obtain a second evaluation result by statistically analyzing the accumulation degree of various security dimensions and setting different weights for different security dimensions. Further combined with the above-mentioned first evaluation result, a comprehensive evaluation result of the distribution resources can be obtained.

[0128] See also Figure 5 , which exemplarily shows a flow chart of an evaluation method of another embodiment of the present application. As shown in the figure, the embodiment of the present application mainly involves obtaining two types of evaluation results, namely the first evaluation result and the second evaluation result. For the first evaluation result, by obtaining the accumulation degree of multiple distribution resources in the first historical period and the second historical period, and multiple evaluation dimensions, the sub-evaluation result of each distribution resource in each scoring dimension is determined based on the improved scorecard model of the present application (i.e., corresponding to one evaluation dimension, only based on the proportion of negative samples in each sub-box), and the first evaluation result is obtained by combining the sub-evaluation results of all scoring dimensions. For the second evaluation result, the accumulation degree of each security dimension is determined by public security incidents, medical incidents, etc. reported from shops, shoppers or public institutions, and different weights are set in combination with different security dimensions to obtain the second evaluation result. The embodiment of the present application also supports some special bonus items (for example, the distribution resources appear to be brave and righteous), so the first evaluation result, the second evaluation result and the special bonus items are combined to obtain a comprehensive evaluation result.

[0129] Based on the above embodiments, as an optional embodiment, it also includes:

[0130] An alarm message is sent to the distribution resource whose at least one evaluation result among the first evaluation result, the second evaluation result or the comprehensive evaluation result meets the preset conditions.

[0131] The present application embodiment provides a distribution resource evaluation device, such as Figure 6 As shown, the distribution resource evaluation device may include: an information acquisition module 601, a positive and negative sample determination module 602, a sub-evaluation result module 603 and a first result acquisition module 604, wherein:

[0132] Information acquisition module 601, applying to acquire the accumulation degree of at least one evaluation dimension of multiple distribution resources in a first historical period and a second historical period respectively; wherein the first historical period is earlier than the second historical period;

[0133] The positive and negative sample determination module 602 is applied to determine a first positive sample and a first negative sample from the multiple distribution resources, wherein the first negative sample is a distribution resource whose accumulation degree of at least one negative evaluation dimension exceeds a preset degree in a second historical period, and the first positive sample is a distribution resource other than the first negative sample among the multiple distribution resources;

[0134] The sub-evaluation result module 603 is used to, for each evaluation dimension, bin the multiple distribution resources according to the accumulation degree of the evaluation dimension of the multiple distribution resources in the first historical period, and determine the sub-evaluation results of each first positive sample and each first negative sample in the evaluation dimension according to the proportion of the first negative samples in each bin to the total number of first negative samples;

[0135] The first result acquisition module 604 is used to obtain, for each distribution resource, a first evaluation result of the distribution resource according to the sub-evaluation results of the distribution resource in all evaluation dimensions.

[0136] The device of the embodiments of the present application can execute the method provided by the embodiments of the present application, and the implementation principles are similar. The actions performed by each module in the device of each embodiment of the present application correspond to the steps in the method of each embodiment of the present application. For the detailed functional description of each module of the device, please refer to the description in the corresponding method shown in the previous text, which will not be repeated here.

[0137] In an embodiment of the present application, an electronic device is provided, including a memory, a processor, and a computer program stored in the memory, wherein the processor executes the computer program to implement the steps of the distribution resource evaluation method. Compared with the related art, the following can be achieved: by obtaining the accumulation degree of at least one evaluation dimension of multiple distribution resources in a first historical period and a second historical period, respectively, and using the accumulation degree of the evaluation dimension in the later second historical period to determine the first positive sample and the negative sample, a finer granularity of data division is achieved, which lays a foundation for more accurate evaluation of distribution resources; for each evaluation dimension, the accumulation degree of the evaluation dimension of the multiple distribution resources in the first historical period is used to bin the multiple distribution resources; according to the proportion of the first negative samples in each bin to the total number of the first negative samples, the sub-evaluation results of each first positive sample and each first negative sample in the evaluation dimension are determined; compared with the prior art, the problem of missing risk correlation can be solved; the self-evaluation result corresponding to each bin can be monotonic corresponding to the accumulation degree, thereby more reasonably stratifying the distribution resources; and because it is more in line with cognition, it is also easier to be accepted by the distribution resources, which lays a foundation for later warning and behavior optimization of the distribution resources.

[0138] In an alternative embodiment, an electronic device is provided, such as Figure 7 As shown, Figure 7 The electronic device 4000 shown includes: a processor 4001 and a memory 4003. The processor 4001 and the memory 4003 are connected, such as through a bus 4002. Optionally, the electronic device 4000 may also include a transceiver 4004, which may be used for data interaction between the electronic device and other electronic devices, such as data transmission and / or data reception. It should be noted that in actual applications, the transceiver 4004 is not limited to one, and the structure of the electronic device 4000 does not constitute a limitation on the embodiments of the present application.

[0139] Processor 4001 may be a CPU (Central Processing Unit), a general-purpose processor, a DSP (Digital Signal Processor), an ASIC (Application Specific Integrated Circuit), an FPGA (Field Programmable Gate Array) or other programmable logic devices, transistor logic devices, hardware components or any combination thereof. It may implement or execute various exemplary logic blocks, modules and circuits described in conjunction with the disclosure of this application. Processor 4001 may also be a combination that implements computing functions, such as a combination of one or more microprocessors, a combination of a DSP and a microprocessor, etc.

[0140] The bus 4002 may include a path to transmit information between the above components. The bus 4002 may be a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus. The bus 4002 may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 7 Only one thick line is used in the diagram, but this does not mean that there is only one bus or only one type of bus.

[0141] The memory 4003 may be a ROM (Read Only Memory) or other types of static storage devices that can store static information and instructions, a RAM (Random Access Memory) or other types of dynamic storage devices that can store information and instructions, or an EEPROM (Electrically Erasable Programmable Read Only Memory), a CD-ROM (Compact Disc Read Only Memory) or other optical disk storage, optical disk storage (including compressed optical disk, laser disk, optical disk, digital versatile disk, Blu-ray disk, etc.), magnetic disk storage media, other magnetic storage devices, or any other medium that can be used to carry or store computer programs and can be read by a computer, without limitation herein.

[0142] The memory 4003 is used to store the computer program for executing the embodiment of the present application, and the execution is controlled by the processor 4001. The processor 4001 is used to execute the computer program stored in the memory 4003 to implement the steps shown in the above method embodiment.

[0143] An embodiment of the present application provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps and corresponding contents of the aforementioned method embodiment can be implemented.

[0144] The embodiment of the present application also provides a computer program product, including a computer program, which can implement the steps and corresponding contents of the aforementioned method embodiment when executed by a processor.

[0145] The terms "first", "second", "third", "fourth", "1", "2", etc. (if any) in the specification and claims of this application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the numbers used in this way can be interchanged where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than that shown or described in the drawings.

[0146] It should be understood that, although each operation step is indicated by arrows in the flowchart of the embodiment of the present application, the implementation order of these steps is not limited to the order indicated by the arrows. Unless clearly stated herein, in some implementation scenarios of the embodiment of the present application, the implementation steps in each flowchart can be performed in other orders according to demand. In addition, some or all of the steps in each flowchart may include multiple sub-steps or multiple stages based on actual implementation scenarios. Some or all of these sub-steps or stages may be executed at the same time, and each sub-step or stage in these sub-steps or stages may also be executed at different times respectively. In different scenarios of execution time, the execution order of these sub-steps or stages may be flexibly configured according to demand, and the embodiment of the present application does not limit this.

[0147] The above is only an optional implementation method for some implementation scenarios of the present application. It should be pointed out that for ordinary technicians in this technical field, without departing from the technical concept of the solution of the present application, other similar implementation methods based on the technical ideas of the present application are also within the protection scope of the embodiments of the present application.

Claims

1. A distribution resource evaluation method, characterized in that: include: Obtaining accumulation levels of at least one evaluation dimension of a plurality of distribution resources in a first historical period and a second historical period, respectively; wherein the first historical period is earlier than the second historical period; Determining a first positive sample and a first negative sample from the multiple distribution resources, the first negative sample being a distribution resource whose accumulation degree of at least one negative evaluation dimension exceeds a preset degree in a second historical period, and the first positive sample being a distribution resource other than the first negative sample among the multiple distribution resources; For each evaluation dimension, the multiple distribution resources are binned according to the accumulation degree of the evaluation dimension of the multiple distribution resources in the first historical period, and the sub-evaluation results of each first positive sample and each first negative sample in the evaluation dimension are determined according to the proportion of the first negative samples in each bin to the total number of the first negative samples; For each distribution resource, obtaining a first evaluation result of the distribution resource according to the sub-evaluation results of the distribution resource in all evaluation dimensions; The step of performing box processing on the plurality of distribution resources includes: Determine a plurality of bins, each bin corresponding to an accumulation degree interval of the evaluation dimension, and the accumulation degree intervals corresponding to the bins do not have any intersection and are singular; Determine the target sub-box where each distribution resource is located according to the accumulation degree of each distribution resource in the evaluation dimension; The accumulation degree of the distribution resources in the evaluation dimension is located in the accumulation degree interval corresponding to the target sub-box.

2. The evaluation method according to claim 1, characterized in that: According to the proportion of the first negative samples in each bin to the total number of the first negative samples, the sub-evaluation results of each first positive sample and each first negative sample in the evaluation dimension are determined, including: Each sub-box is traversed in turn according to a preset order. For the currently traversed sub-box, the sub-evaluation result of each distribution resource in the sub-box in the evaluation dimension is determined according to the ratio of all first negative samples in the currently traversed sub-box and the traversed sub-boxes to the total number of first negative samples.

3. The evaluation method according to any one of claims 1 to 2, characterized in that: The obtaining, according to the sub-evaluation results of the distribution resource in all evaluation dimensions, a first evaluation result of the distribution resource comprises: Get the weight of each evaluation dimension; A weighted sum is performed based on the sub-evaluation results of the distribution resource in all evaluation dimensions and the corresponding weights to obtain a first evaluation result of the distribution resource.

4. The evaluation method according to claim 3, characterized in that: The step of obtaining the weight of each evaluation dimension also includes the step of determining the weight of each evaluation dimension: Obtaining the accumulation degree of at least one evaluation dimension of multiple sample delivery resources in a third historical period and a fourth historical period respectively; wherein the third historical period is earlier than the fourth historical period; Determine a second positive sample and a second negative sample from a plurality of sample delivery resources, wherein the second negative sample is a delivery resource for which the accumulation degree of at least one negative evaluation dimension exceeds a preset degree in a fourth historical period, and the second positive sample is a delivery resource other than the first negative sample among the plurality of sample delivery resources; For each evaluation dimension, determine the sub-evaluation results of each second positive sample and each second negative sample in the evaluation dimension, and for each sample distribution resource, perform weighted summation according to the sub-evaluation results of the sample distribution resource in all evaluation dimensions and the corresponding initial weights to obtain the first sample evaluation result of the sample distribution resource; The initial weight is adjusted according to the difference between the first sample evaluation result and the target value until the difference converges to obtain the weight; wherein, when the sample distribution resource is a positive sample, the target value is a first target value, and when the sample distribution resource is a negative sample, the target value is a second target value.

5. The evaluation method according to claim 1, characterized in that: Also includes: Obtaining accumulation levels of at least one safety dimension of each of the plurality of distribution resources in the first historical period; For each distribution resource, obtaining a second evaluation result of the distribution resource according to the accumulation degree of at least one safety dimension of the distribution resource in the first historical period; A comprehensive evaluation result of the distribution resource is obtained according to the first evaluation result and the second evaluation result of the distribution resource.

6. The evaluation method according to claim 5, characterized in that: Also includes: An alarm message is sent to the distribution resource whose at least one evaluation result among the first evaluation result, the second evaluation result or the comprehensive evaluation result meets the preset conditions.

7. A distribution resource evaluation device, characterized in that: include: An information acquisition module is used to acquire the accumulation degree of at least one evaluation dimension of multiple distribution resources in a first historical period and a second historical period respectively; wherein the first historical period is earlier than the second historical period; A positive and negative sample determination module is used to determine a first positive sample and a first negative sample from the multiple distribution resources, wherein the first negative sample is a distribution resource whose accumulation degree of at least one negative evaluation dimension exceeds a preset degree in a second historical period, and the first positive sample is a distribution resource other than the first negative sample among the multiple distribution resources; A sub-evaluation result module is used to, for each evaluation dimension, bin the multiple distribution resources according to the accumulation degree of the evaluation dimension of the multiple distribution resources in the first historical period, and determine the sub-evaluation results of each first positive sample and each first negative sample in the evaluation dimension according to the proportion of the first negative samples in each bin to the total number of first negative samples; A first result acquisition module, configured to obtain, for each distribution resource, a first evaluation result of the distribution resource according to the sub-evaluation results of the distribution resource in all evaluation dimensions; The sub-assessment result module performs binning processing on the multiple distribution resources, including: Determine a plurality of bins, each bin corresponding to an accumulation degree interval of the evaluation dimension, and the accumulation degree intervals corresponding to the bins do not have any intersection and are singular; Determine the target sub-box where each distribution resource is located according to the accumulation degree of each distribution resource in the evaluation dimension; The accumulation degree of the distribution resources in the evaluation dimension is located in the accumulation degree interval corresponding to the target sub-box.

8. An electronic device comprising a memory, a processor and a computer program stored in the memory, characterized in that: The processor executes the computer program to implement the steps of the distribution resource evaluation method according to any one of claims 1 to 6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the distribution resource evaluation method according to any one of claims 1 to 6 are implemented.

Images