Classification method and device for an object, storage medium, and electronic device

Abstract

The present application discloses a classification method and device for objects, a storage medium, and an electronic device. It relates to the technical fields of data processing, fintech, and other related technical fields. The method includes: calculating based on the attribute information and historical behavior information of multiple objects to be classified to obtain cumulative function curves corresponding to the multiple objects to be classified under the attribute information; respectively solving the cumulative function curves according to multiple classification quantities to obtain initial classification results corresponding to each classification quantity; evaluating the initial classification results based on the attribute information and historical behavior information to obtain multiple evaluation values; determining a target classification quantity from the multiple classification quantities according to the multiple evaluation values, and determining a target classification result based on the initial classification result corresponding to the target classification quantity. Through the present application, the problem in the related art that the accuracy of classifying a target object is relatively low due to the manual classification of the target object is solved.

Description

Technical Field

[0001] The present application relates to the technical fields of data processing, fintech, and other related technical fields. Specifically, it relates to a method and apparatus for classifying objects, a storage medium, and an electronic device. Background Technique

[0002] In many fields such as the financial field, the gaming field, and the new media field, it is often necessary to bin (i.e., classify) customers according to the relationship between variables and targets. For example: In a study targeting whether a customer defaults in the credit field, given the historical product usage behavior attributes (variables) and product default flags (targets) of all customers, for each historical product usage behavior attribute, IV maximization binning is performed, so that for each historical product usage behavior attribute of each customer, the WOE value of the interval to which the corresponding historical product usage behavior attribute belongs can be obtained based on the corresponding binning result. However, since the rules and policies in each field usually show a stepwise or linear relationship, while hoping that the IV is significant enough, staff will also impose constraints such as the WOE trend of the interval, the WOE difference of the interval, and the sample difference of the interval on variable binning. Most traditional methods bin customers based on basic binning methods such as equal-frequency binning and equal-distance binning, lacking consideration of additional constraints, prone to misinterpretation risks, and time-consuming in the binning process, with low efficiency. At the same time, since the relationship between the discrimination ability and the number of bins and the relationship between endogeneity and the number of bins are inverse, there are contradictions between traditional IV indicators and WOE coding in terms of discrimination ability and endogeneity, and staff also have difficulties in selecting an appropriate number of bins.

[0003] Regarding the problem that the accuracy of classifying target objects by manual means in related technologies is relatively low, no effective solution has been proposed yet. Summary of the Invention

[0004] The main objective of the present application is to provide a method and apparatus for classifying objects, a storage medium, and an electronic device, so as to solve the problem that the accuracy of classifying target objects by manual means in related technologies is relatively low.

[0005] To achieve the above object, according to one aspect of the present application, a classification method for objects is provided. The method includes: calculating based on the attribute information of a plurality of objects to be classified and the historical behavior information of each object to be classified in the target institution to obtain a cumulative function curve corresponding to the plurality of objects to be classified under the attribute information; solving the cumulative function curve respectively according to a plurality of classification quantities to obtain initial classification results corresponding to the plurality of objects to be classified under each classification quantity; evaluating the initial classification results corresponding to the plurality of objects to be classified under each classification quantity based on the attribute information and the historical behavior information to obtain a plurality of evaluation values; determining a target classification quantity from the plurality of classification quantities according to the plurality of evaluation values, and determining a target classification result according to the initial classification result corresponding to the target classification quantity, wherein the target institution determines a processing strategy corresponding to the plurality of objects to be classified according to the target classification result.

[0006] Further, calculating based on the attribute information of a plurality of objects to be classified and the historical behavior information of each object to be classified in the target institution to obtain a cumulative function curve corresponding to the plurality of objects to be classified under the attribute information includes: determining a first value corresponding to the plurality of objects to be classified according to the attribute information; calculating a second value according to the historical behavior information of each object to be classified in the target institution; and obtaining the cumulative function curve based on the first value and the second value.

[0007] Further, solving the cumulative function curve respectively according to a plurality of classification quantities to obtain initial classification results corresponding to the plurality of objects to be classified under each classification quantity includes: determining a starting point in the cumulative function curve; for each classification quantity, solving the starting point and the second value corresponding to the historical behavior information of a plurality of objects to be classified in the cumulative function curve through a target cutting algorithm to obtain a first coordinate point, and determining a classification range based on the first coordinate point and the starting point; determining whether the classification quantity after classifying the plurality of objects to be classified under the classification range is equal to the classification quantity; if the classification quantity is not equal to the classification quantity, repeating the step of solving the first coordinate point and the second value corresponding to the historical behavior information of a plurality of objects to be classified in the cumulative function curve through the target cutting algorithm to obtain a second coordinate point until the initial classification result corresponding to the classification quantity is obtained.

[0008] Further, for each number of classifications, the target cutting algorithm is used to solve the second numerical values corresponding to the historical behavior information of multiple objects to be classified in the starting point and the cumulative function curve, and the first coordinate points obtained include: for each number of classifications, the target cutting algorithm is used to calculate the cumulative function curve based on the first sensitivity coefficient, the second sensitivity coefficient, and the starting point to obtain a candidate point set, where the first sensitivity coefficient is used to characterize the mean difference degree between the historical behavior information corresponding to each category after classification, and the second sensitivity coefficient is used to characterize the number difference degree of the objects corresponding to each category after classification; the IV index is used to calculate multiple candidate coordinate points in the candidate point set to obtain the index value corresponding to each candidate coordinate point; according to the index value corresponding to each candidate coordinate point, the first coordinate point is determined from the multiple candidate coordinate points in the candidate point set.

[0009] Further, according to the attribute information and the historical behavior information, the initial classification results corresponding to the multiple objects to be classified under each number of classifications are evaluated, and multiple evaluation values are obtained, including: encoding the attribute information of the multiple objects to be classified to obtain the encoded attribute information; calculating according to the historical behavior information, the encoded attribute information, and the attribute information to obtain the first evaluation value corresponding to the initial classification result; calculating according to the attribute information and the initial classification result to obtain the second evaluation value corresponding to the initial classification result; calculating according to the first evaluation value and the second evaluation value to obtain the multiple evaluation values.

[0010] Further, calculating according to the historical behavior information, the encoded attribute information, and the attribute information to obtain the first evaluation value corresponding to the initial classification result includes: calculating according to the encoded attribute information and the attribute information to obtain the random error term corresponding to each object to be classified; calculating according to the random error term, the encoded attribute information, and the historical behavior information to obtain a first calculation result; calculating according to the encoded attribute information and the historical behavior information to obtain a second calculation result; calculating according to the first calculation result and the second calculation result to obtain the first evaluation value.

[0011] Further, determining the target classification result based on the initial classification result corresponding to the target classification quantity includes: screening the multiple objects to be classified and the initial classification result according to a preset screening condition to obtain a first sample quantity corresponding to the screened multiple objects to be classified and a second sample quantity corresponding to the screened initial classification result; calculating according to the first sample quantity and the second sample quantity to obtain a credibility index value corresponding to the initial classification result; adjusting the initial classification result according to the credibility index value to obtain an adjusted initial classification result; and determining the adjusted initial classification result as the target classification result.

[0012] To achieve the above object, according to another aspect of the present application, there is provided a classification device for objects. The device includes: a calculation unit configured to calculate, according to the attribute information of multiple objects to be classified and the historical behavior information of each object to be classified in a target institution, a cumulative function curve corresponding to the multiple objects to be classified under the attribute information; a solution unit configured to respectively solve the cumulative function curve according to multiple classification quantities to obtain an initial classification result corresponding to the multiple objects to be classified under each classification quantity; an evaluation unit configured to evaluate the initial classification result corresponding to the multiple objects to be classified under each classification quantity according to the attribute information and the historical behavior information to obtain multiple evaluation values; and a determination unit configured to determine a target classification quantity from the multiple classification quantities according to the multiple evaluation values, and determine a target classification result according to the initial classification result corresponding to the target classification quantity, wherein the target institution determines a processing strategy corresponding to the multiple objects to be classified according to the target classification result.

[0013] Further, the calculation unit includes: a first determination module configured to determine a first numerical value corresponding to the multiple objects to be classified according to the attribute information; a first calculation module configured to calculate according to the historical behavior information of each object to be classified in the target institution to obtain a second numerical value; and a second determination module configured to obtain the cumulative function curve according to the first numerical value and the second numerical value.

[0014] Further, the solving unit includes: a third determination module for determining a starting point in the cumulative function curve; a solving module for, for each classification quantity, solving, through a target cutting algorithm, a second value corresponding to the historical behavior information of a plurality of objects to be classified in the starting point and the cumulative function curve to obtain a first coordinate point, and determining a classification range based on the first coordinate point and the starting point; a judgment module for judging whether the classification quantity after classifying the plurality of objects to be classified within the classification range is equal to the classification quantity; and an execution module for, if the classification quantity is not equal to the classification quantity, repeatedly executing the step of solving, through the target cutting algorithm, a second value corresponding to the historical behavior information of a plurality of objects to be classified in the first coordinate point and the cumulative function curve to obtain a second coordinate point until an initial classification result corresponding to the classification quantity is obtained.

[0015] Further, the solving module includes: a first calculation sub-module for, for each classification quantity, calculating, through the target cutting algorithm, the cumulative function curve based on a first sensitivity coefficient, a second sensitivity coefficient, and the starting point to obtain a candidate point set, where the first sensitivity coefficient is used to characterize the mean difference degree between the historical behavior information corresponding to each category after classification, and the second sensitivity coefficient is used to characterize the quantity difference degree of the objects corresponding to each category after classification; a second calculation sub-module for calculating, according to the IV index, a plurality of candidate coordinate points in the candidate point set to obtain an index value corresponding to each candidate coordinate point; and a determination sub-module for determining the first coordinate point from the plurality of candidate coordinate points in the candidate point set according to the index value corresponding to each candidate coordinate point.

[0016] Further, the evaluation unit includes: an encoding module for encoding the attribute information of the plurality of objects to be classified to obtain encoded attribute information; a second calculation module for calculating, according to the historical behavior information, the encoded attribute information, and the attribute information, a first evaluation value corresponding to the initial classification result; a third calculation module for calculating, according to the attribute information and the initial classification result, a second evaluation value corresponding to the initial classification result; and a fourth calculation module for calculating, according to the first evaluation value and the second evaluation value, a plurality of evaluation values.

[0017] Further, the second calculation module includes: a third calculation sub-module, configured to calculate, according to the encoded attribute information and the attribute information, a random error term corresponding to each object to be classified; a fourth calculation sub-module, configured to calculate, according to the random error term, the encoded attribute information, and the historical behavior information, a first calculation result; a fifth calculation sub-module, configured to calculate, according to the encoded attribute information and the historical behavior information, a second calculation result; and a sixth calculation sub-module, configured to calculate, according to the first calculation result and the second calculation result, the first evaluation value.

[0018] Further, the determination unit includes: a screening module, configured to screen the multiple objects to be classified and the initial classification result according to a preset screening condition, to obtain a first sample quantity corresponding to the screened multiple objects to be classified and a second sample quantity corresponding to the screened initial classification result; a fifth calculation module, configured to calculate, according to the first sample quantity and the second sample quantity, a credibility index value corresponding to the initial classification result; an adjustment module, configured to adjust the initial classification result according to the credibility index value, to obtain an adjusted initial classification result; and a fourth determination module, configured to determine the adjusted initial classification result as the target classification result.

[0019] To achieve the above object, according to one aspect of the present application, there is provided a computer-readable storage medium storing a program, wherein when the program runs, it controls a device where the storage medium is located to execute the object classification method described in any one of the above.

[0020] To achieve the above object, according to another aspect of the present application, there is further provided an electronic device including one or more processors and a memory for storing a program for the one or more processors to implement the object classification method described in any one of the above.

[0021] Through this application, the following steps are adopted: calculate based on the attribute information of multiple objects to be classified and the historical behavior information of each object to be classified in the target institution, and obtain the cumulative function curves corresponding to the multiple objects to be classified under the attribute information; solve the cumulative function curves respectively according to multiple classification quantities, and obtain the initial classification results corresponding to the multiple objects to be classified under each classification quantity; evaluate the initial classification results corresponding to the multiple objects to be classified under each classification quantity based on the attribute information and the historical behavior information, and obtain multiple evaluation values; determine the target classification quantity from the multiple classification quantities according to the multiple evaluation values, and determine the target classification result according to the initial classification result corresponding to the target classification quantity, wherein the target institution determines the processing strategies corresponding to the multiple objects to be classified according to the target classification result. Through this application, the problem in the related art that the accuracy of classifying the target object is relatively low by manually classifying the target object is solved.

[0022] In this solution, based on the attribute information corresponding to multiple objects to be classified and their historical behavior information in the target institution, the cumulative function curve corresponding to each object to be classified under the attribute information is constructed. Then, solve the cumulative function curve respectively through a preset multiple of classification quantities, and obtain the initial classification results corresponding to the multiple objects to be classified under each classification quantity. Finally, evaluate the initial classification results under each classification quantity to determine the appropriate classification quantity (i.e., the target classification quantity) and the classification result (i.e., the target classification result).

[0023] Through the attribute information and historical behavior information of multiple objects to be classified, the cumulative function curves corresponding to the multiple objects to be classified under the attribute information can be constructed quickly, so as to better capture the characteristics and change trends of the objects to be classified. At the same time, by solving the cumulative function curve respectively based on multiple classification quantities, the initial classification results corresponding to the multiple objects to be classified under each classification quantity can be obtained more intuitively, which helps to further analyze and understand the characteristics and behaviors of the objects to be classified, improves the accuracy and reliability of classification, and helps decision-makers make correct classification decisions better. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] The drawings forming a part of this application are used to provide a further understanding of this application. The schematic embodiments of this application and their descriptions are used to explain this application and do not constitute an improper limitation to this application. In the drawings:

[0025] Figure 1 is a flowchart of a method for classifying objects according to an embodiment of this application;

[0026] Figure 2 is a schematic diagram of a cumulative function curve according to an embodiment of this application;

[0027] Figure 3 It is a schematic diagram of a candidate point set provided according to an embodiment of the present application;

[0028] Figure 4 It is a schematic diagram of a classification device for an object provided according to an embodiment of the present application;

[0029] Figure 5 It is a schematic diagram of an electronic device provided according to an embodiment of the present application. Detailed implementation manners

[0030] It should be noted that, without conflict, the embodiments in the present application and the features in the embodiments may be combined with each other. The present application will be described in detail below with reference to the drawings and in combination with the embodiments.

[0031] In order to enable those skilled in the art to better understand the solution of the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without making creative efforts shall fall within the protection scope of the present application.

[0032] It should be noted that the terms "first", "second", etc. in the specification and claims of the present application and the above drawings are used to distinguish similar objects, and do not necessarily need to be used to describe a specific order or sequence. It should be understood that such data can be interchanged under appropriate circumstances for the embodiments of the present application described herein. In addition, the terms "comprising" and "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or device that includes a series of steps or units does not necessarily have to be limited to those steps or units clearly listed, but may include other steps or units not clearly listed or inherent to these processes, methods, products, or devices.

[0033] For the convenience of description, some nouns or terms related to the embodiments of the present application are described below:

[0034] WOE (Weight of Evidence): It characterizes the positive and negative sample differences of the feature variable grouping. Generally, the larger the absolute value of the WOE value, the better the prediction ability of the grouping; the smaller the absolute value of the WOE value, the worse the prediction ability of the grouping.

[0035] IV (Information Value): That is, the information value, which is used to represent the contribution degree of the feature to the target prediction, that is, the prediction ability of the feature. The larger the IV value, the stronger the prediction ability of the feature.

[0036] Sum of Squares Between Treatments or Factors (SSA): It reflects the sum of the gaps between the average levels of each group of features and the overall average level.

[0037] Constrained Optimization Binning: For each variable and the corresponding target, modelers usually need to divide the variable values of the samples into several intervals for variable discretization. In addition to maximizing indicators such as IV, staff usually hope that after a variable is discretized, it can also meet other constraint conditions. In the industry, this binning method that takes into account other constraint conditions while maximizing the indicators is generally called constrained optimization binning.

[0038] WOE Encoding: Before modeling, modelers need to perform WOE encoding, that is, according to the binning results, convert the independent variable of any sample into the WOE value of the interval to which the independent variable belongs, and use the WOE value to replace the original independent variable for modeling.

[0039] It should be noted that the relevant information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data for display, analysis data, etc.) involved in this disclosure are all information and data authorized by users or fully authorized by all parties. For example, an interface is set between this system and relevant users or institutions. Before obtaining relevant information, a request for obtaining information needs to be sent to the aforementioned users or institutions through the interface, and after receiving the consent information feedback from the aforementioned users or institutions, the relevant information is obtained.

[0040] The present invention will be described below in conjunction with preferred implementation steps. Figure 1 is a flowchart of a classification method for an object provided by an embodiment of the present application, as Figure 1 shown. The method includes the following steps:

[0041] Step S101, calculate based on the attribute information of multiple objects to be classified and the historical behavior information of each object to be classified in the target institution, and obtain the cumulative function curves corresponding to the multiple objects to be classified under the attribute information.

[0042] Optionally, the attribute information (which can also be understood as the independent variable in the classification process) includes but is not limited to: age information, occupation information, asset information, etc. of the user (i.e., each object to be classified). The historical behavior information (which can also be understood as the dependent variable in the classification process) includes but is not limited to: binary classification information such as whether there is a potential risk for the user in the target institution, or whether the user can bring benefits to the target institution. The target institution includes but is not limited to: financial institutions, Internet institutions, media institutions, game institutions, etc. Model based on the attribute information and historical behavior information to obtain the cumulative function curves corresponding to the multiple objects to be classified under the attribute information.

[0043] It should be noted that the independent variables in the classification process are numerical independent variables, including but not limited to: age information and asset information of multiple objects to be classified, etc.

[0044] It should be noted that the attribute information can be screened according to corresponding thresholds such as the missing rate and correlation (which can also be understood as the screening of independent variables), and the screened attribute information (which can also be understood as the independent variables included in the modeling process) can be obtained.

[0045] Step S102: Solve the cumulative function curves respectively according to multiple classification quantities, and obtain the initial classification results corresponding to multiple objects to be classified under each classification quantity.

[0046] Optionally, loop through the potential number of bins (i.e., multiple classification quantities). For each number of bins (i.e., the classification quantity used in each loop), use the geometric cutting algorithm to solve the cumulative function curve corresponding to multiple objects to be classified under this number of bins, and obtain the initial binning results (i.e., initial classification results) corresponding to multiple objects to be classified under this number of bins.

[0047] Step S103: Evaluate the initial classification results corresponding to multiple objects to be classified under each classification quantity based on the attribute information and historical behavior information, and obtain multiple evaluation values.

[0048] Optionally, calculate the performance of the initial classification results corresponding to multiple objects to be classified under each classification quantity through comprehensive performance indicators (such as: endogeneity indicators and discrimination ability indicators, etc.) to obtain multiple evaluation values.

[0049] Step S104: Determine the target classification quantity among multiple classification quantities based on multiple evaluation values, and determine the target classification result based on the initial classification result corresponding to the target classification quantity. Among them, the target institution determines the processing strategies corresponding to multiple objects to be classified based on the target classification result.

[0050] Optionally, find the number of bins with the best performance and the corresponding initial binning results among multiple evaluation values as the target classification quantity and target classification result corresponding to multiple objects to be classified.

[0051] It should be noted that in the actual binning process of multiple objects to be classified, in addition to requiring the IV after binning of the independent variable to be as high as possible, the staff also requires considering the practical meaning and usability of the variable. For example: the staff will have the following binning constraints: 1) The WOE values of the front and back intervals show a monotonically increasing or decreasing trend; 2) The difference in WOE values between the front and back intervals should be obvious enough; 3) The difference in the number of samples in each interval should not be too large. This binning method that takes into account other constraint conditions while maximizing the target index is called constrained optimization binning in the industry.

[0052] In summary, based on the attribute information corresponding to multiple objects to be classified and their historical behavior information in the target institution, a cumulative function curve corresponding to each object to be classified under the attribute information is constructed. Then, through a preset number of classifications, the cumulative function curve is solved respectively to obtain the initial classification results corresponding to the multiple objects to be classified under each number of classifications. Finally, the initial classification results under each number of classifications are evaluated to determine the appropriate number of classifications (i.e., the target number of classifications) and the classification results (i.e., the target classification results).

[0053] Through the attribute information and historical behavior information of multiple objects to be classified, the cumulative function curves corresponding to the multiple objects to be classified under the attribute information can be constructed quickly, so as to better capture the characteristics and change trends of the objects to be classified. At the same time, by solving the cumulative function curve based on multiple numbers of classifications respectively, the initial classification results corresponding to the multiple objects to be classified under each number of classifications can be obtained more intuitively, which helps to further analyze and understand the characteristics and behaviors of the objects to be classified, improves the accuracy and reliability of classification, and helps decision-makers make correct classification decisions better.

[0054] Optionally, in the object classification method provided in the embodiment of the present application, calculating based on the attribute information of multiple objects to be classified and the historical behavior information of each object to be classified in the target institution, and obtaining the cumulative function curves corresponding to the multiple objects to be classified under the attribute information includes: determining a first value corresponding to the multiple objects to be classified according to the attribute information; calculating according to the historical behavior information of each object to be classified in the target institution to obtain a second value; and obtaining the cumulative function curve according to the first value and the second value.

[0055] In an optional embodiment, assuming that the independent variable (i.e., the attribute information) is x, the dependent variable (i.e., the historical behavior information) is y, there are n samples in the modeling (i.e., there are n customers), and the independent variable x of the t-th sample is marked as and the dependent variable y is marked as . Sort the n customers in ascending order according to the value of the independent variable x, and number the customers in turn. The customer number is , and the value of the independent variable x of the p-th customer is marked as (i.e., the first value).

[0056] Construct the cumulative dependent variable , represents the accumulation of the dependent variables from the 1st customer to the p-th customer (i.e., the second value), that is can be expressed as:

[0057]

[0058] Because , so , that is, the cumulative dependent variable is a function of the cumulative number of customers.

[0059] Using the serial number p of the customers arranged in ascending order of the independent variable x as the abscissa, and the cumulative dependent variable as the ordinate, plot the curve of the cumulative dependent variable (i.e., the cumulative function curve). When the abscissa is equal to 0, mark it as , since there is no cumulative dependent variable at this time, so , thus the origin (0, 0) can be represented as .

[0060] It should be noted that Figure 2 is a schematic diagram of the cumulative function curve provided according to the embodiments of the present application. As Figure 2 shown, assuming that the staff hopes to obtain k classification results, then take k abscissas on the horizontal axis, and the serial numbers are respectively , , …, . Among them, the last abscissa (i.e., the k-th abscissa) is fixed as n, that is . The above k points correspond to k serial numbers, that is, corresponding to k customers, and their independent variable x values are respectively , , …, . The above k abscissas correspond to the ordinates on the curve of the cumulative dependent variable respectively , , ……, . Thus, k points on the curve of the cumulative dependent variable are formed . The first point on the curve is connected to the origin , and then each point is connected to the previous point, that is is connected to , thus forming a broken line composed of k line segments. The abscissas of the two endpoints of each line segment form the upper and lower bounds of an interval. For example: the i-th line segment is the point connected to the point , so it corresponds to the i-th interval, and this interval is . Therefore, the interval corresponding to before is , and correspond to the interval , and correspond to the interval , and correspond to the interval Dividing the cumulative dependent variable curve into k line segments is equivalent to performing k - binning on the independent variable x.

[0061] It should be noted that for the i - th interval the WOE value can be expressed as:

[0062]

[0063] where represents the number of samples with the dependent variable equal to 1 in the i - th interval, represents the number of all samples with the dependent variable equal to 1, represents the number of samples with the dependent variable equal to 0 in the i - th interval, represents the number of all samples with the dependent variable equal to 0.

[0064] Requiring the mean gain of each interval after binning the independent variable to be monotonic actually means requiring the slopes of the k line segments on the "cumulative dependent variable" curve to increase or decrease one by one.

[0065] It should be noted that the constraint conditions can be concretized into geometric properties: (1) The WOE values of the front and back intervals on the cumulative function curve showing a monotonically increasing or decreasing trend is equivalent to requiring the slopes of the k line segments on the cumulative function curve to increase or decrease one by one. (2) The difference in WOE values of the front and back intervals on the cumulative function curve being sufficiently obvious is equivalent to requiring the difference in slopes of the k line segments on the cumulative function curve to be sufficiently obvious. (3) The difference in the number of samples in each interval on the cumulative function curve not being too large is equivalent to requiring the difference in the projected lengths of the slopes of the k line segments on the cumulative function curve on the abscissa not to be too large.

[0066] By comprehensively considering the attribute information and historical behavior information of multiple objects to be classified to construct the cumulative function curve, it can more comprehensively reflect the characteristics and behaviors of multiple objects to be classified, thereby improving the accuracy of classification.

[0067] Optionally, in the object classification method provided in the embodiments of the present application, solving the cumulative function curve respectively according to multiple classification quantities to obtain the initial classification results corresponding to each classification quantity for multiple objects to be classified includes: determining the starting point in the cumulative function curve; for each classification quantity, solving the second value corresponding to the historical behavior information of multiple objects to be classified in the starting point and the cumulative function curve through the target cutting algorithm to obtain the first coordinate point, and determining the classification range based on the first coordinate point and the starting point; judging whether the classification quantity after classifying multiple objects to be classified within the classification range is equal to this classification quantity; if the classification quantity is not equal to this classification quantity, then repeatedly execute the step of solving the second coordinate point by solving the second value corresponding to the historical behavior information of multiple objects to be classified in the first coordinate point and the cumulative function curve through the target cutting algorithm until the initial classification result corresponding to this classification quantity is obtained.

[0068] In an optional embodiment, the geometric cutting algorithm (i.e., the target cutting algorithm) includes three parameters, namely: the number of interval parameter k, the sensitivity coefficient δ of the mean difference between the front and rear intervals, and the sensitivity coefficient θ of the sample number difference between the front and rear intervals. Each parameter corresponds to different modeling requirements, that is, represents different meanings. The number of interval parameter k: The staff hopes that an independent variable can be divided into k bins. The sensitivity coefficient δ of the mean difference between the front and rear intervals: The staff usually hopes that after binning, the mean difference of the benefits in each interval should be obvious enough, that is, the mean difference of the benefits in each interval should be greater than , which is equivalent to that for the k line segments after geometric cutting of the cumulative function curve, the difference in slopes between the front and rear line segments should not be too small, that is, the difference in slopes between the front and rear line segments should be greater than δ. According to industry habits, the default value of δ is 0.05. The sensitivity coefficient θ of the sample number difference between the front and rear intervals: The staff usually hopes that after binning, the difference in sample numbers in each interval should not be too large, that is, the sample number of the interval with the largest sample number ÷ the sample number of the interval with the smallest sample number ≤ θ. According to industry habits, the default value of θ is 2.

[0069] For each classification quantity (for example: assuming that k is initially set to 6, then each classification quantity is respectively: 6, 5, 4, 3, and 2), solve the origin of the cumulative function (i.e., the starting point) and the second value corresponding to the historical behavior information of multiple objects to be classified through the geometric cutting algorithm based on the above parameters respectively to obtain the first coordinate point, use the range between the first coordinate point and the origin as the first classification range, and obtain the initial classification results corresponding to multiple objects to be classified under the attribute information according to the first classification range.

[0070] It should be noted that if (i - 1) groups have been successfully obtained, that is, the first (i - 1) line segments have been successfully separated on the cumulative dependent variable curve, but the number of classifications obtained is not equal to the current number of classifications (i.e., the current k value). Now, it is desired to obtain the i-th segmentation point, so as to separate the i-th line segment and then obtain the i-th group. First, draw a straight line passing through the point (i.e., the first coordinate point) and (i.e., the second coordinate point) . And draw a straight line passing through the point and with a slope of (δ is the sensitivity coefficient of the mean difference between the front and back intervals, represents the slope of the (i - 1)-th line segment on the cumulative function curve) . On the cumulative dependent variable curve, the points below the straight line and above the straight line and with the abscissa between (θ is the sensitivity coefficient of the difference in the number of samples between the front and back intervals, k is the number of each classification, and n is the number of samples) and (θ is the sensitivity coefficient of the difference in the number of samples between the front and back intervals, k is the number of each classification, and n is the number of samples) are the potential feasible point sets. Sort the points in the feasible point sets by priority, and select the point with the highest priority as the i-th segmentation point, and its coordinates are . Connect and to form a line segment, and the slope of this line segment can be obtained as . If the feasible point set is empty, it means that this recursive call is invalid, and it is necessary to return to the previous step. In the feasible point set formed in the previous step, select the point with a priority one level lower than as the (i - 1)-th segmentation point, and then repeat the above operations. The geometric cutting algorithm will finally end after completing k-bin binning, that is, it ends when finding the (k - 1)-th segmentation point, and outputs all k segmentation points and k intervals (i.e., the initial classification results). Since the geometric cutting algorithm will return to the previous step when it cannot continue to split, in extreme cases, the algorithm will continuously return until it returns to the feasible point set used to determine the first segmentation point. If no suitable i-th segmentation point can be found even after all the feasible point sets used to determine the first segmentation point have been explored, it means that the classification fails, and the algorithm will output an empty set.

[0071] It should be noted that first, assume that the classification trend of multiple objects to be classified finally is monotonically increasing. If the result output when assuming that the classification trend of multiple objects to be classified finally is monotonically increasing is an empty set, then assume that the classification trend of multiple objects to be classified finally is monotonically decreasing, and solve the initial classification result corresponding to the current number of classifications according to the above method.

[0072] It should be noted that the solution objective of the geometric cutting algorithm is as follows: (1) To achieve a monotonicity trend after binning, that is, after binning is completed, the woe values of each interval increase or decrease monotonically. (2) To ensure that after binning is completed, the woe value differences of each interval are sufficiently obvious, that is, the woe value of the high-risk customer group is as high as possible, while the woe value of the low-risk customer group is as low as possible, and at the same time, the differences in the number of samples included in each interval are not too large.

[0073] By continuously repeating the execution of the geometric cutting algorithm, it is possible to ensure obtaining the initial classification result corresponding to the current number of classifications, improving the accuracy and reliability of classification, and ensuring that each object to be classified is correctly classified into the corresponding group.

[0074] Optionally, in the object classification method provided in the embodiments of the present application, for each number of classifications, the target cutting algorithm is used to solve the second numerical values corresponding to the historical behavior information of multiple objects to be classified in the starting point and the cumulative function curve, and the obtained first coordinate points include: for each number of classifications, the target cutting algorithm is used to calculate the cumulative function curve based on the first sensitivity coefficient, the second sensitivity coefficient, and the starting point to obtain a candidate point set, where the first sensitivity coefficient is used to characterize the mean difference degree between the historical behavior information corresponding to each category after classification, and the second sensitivity coefficient is used to characterize the quantity difference degree of the objects corresponding to each category after classification; the IV index is used to calculate multiple candidate coordinate points in the candidate point set to obtain the index value corresponding to each candidate coordinate point; based on the index value corresponding to each candidate coordinate point, the first coordinate point is determined from the multiple candidate coordinate points in the candidate point set.

[0075] In an alternative embodiment, for each number of classifications, if (i - 1) groups have been successfully obtained, that is, the first (i - 1) line segments have been successfully separated on the cumulative dependent variable curve, but the obtained number of classifications is not equal to the current number of classifications (i.e., the current k value), and now it is desired to obtain the i-th segmentation point to separate the i-th line segment and thus obtain the i-th group. First, draw a straight line (i.e., the first coordinate point) and (i.e., the second coordinate point). And draw a straight line passing through the point and with a slope of (δ is the sensitivity coefficient of the mean difference between the front and rear intervals, represents the slope of the (i - 1)-th line segment on the cumulative function curve . On the cumulative dependent variable curve, it is located below the straight line and above the straight line , and the abscissa is between and (where θ is the sensitive coefficient of the difference in the number of samples between the front and back intervals, k is the number of each classification, and n is the number of samples) and The points between (where θ is the sensitive coefficient of the difference in the number of samples between the front and back intervals, k is the number of each classification, and n is the number of samples) are the potential feasible point sets. Among them, the mean difference sensitivity coefficient δ between the front and back intervals is the first sensitivity coefficient, and the sensitive coefficient θ of the difference in the number of samples between the front and back intervals is the second sensitivity coefficient.

[0076] It should be noted that Figure 3 is a schematic diagram of the candidate point set provided by the embodiment of the present application. As Figure 3 shown, if the number of points in the feasible point set is greater than 1, then among the feasible point sets, it is necessary to prioritize the potentially feasible points (i.e., candidate coordinate points). Assume that a certain point in the feasible point set is , and this point is the i-th segmentation point, which can form the potentially i-th interval. Calculate the IV value of the mean return of the first k intervals corresponding to this point, denoted as , and sort according to . The higher the point, the higher the priority.

[0077] By comprehensively evaluating the historical behavior information of multiple objects to be classified through the geometric cutting algorithm, the optimal first coordinate point is found, improving the accuracy and efficiency of classification.

[0078] Optionally, in the object classification method provided by the embodiment of the present application, according to the attribute information and historical behavior information, the initial classification results corresponding to multiple objects to be classified under each classification quantity are evaluated to obtain multiple evaluation values, including: encoding the attribute information of multiple objects to be classified to obtain the encoded attribute information; calculating according to the historical behavior information, the encoded attribute information, and the attribute information to obtain the first evaluation value corresponding to the initial classification result; calculating according to the attribute information and the initial classification result to obtain the second evaluation value corresponding to the initial classification result; calculating according to the first evaluation value and the second evaluation value to obtain multiple evaluation values.

[0079] In an alternative embodiment, there is a contradiction between the traditional IV index and the WOE encoding. There is a causal relationship between the traditional WOE encoding and the dependent variable, that is, the statistical endogeneity problem. When the number of bins is larger, the IV value of the variable is of course higher, but at the same time, the potential endogeneity problem of the WOE encoding will become more obvious; when the number of bins is smaller, the endogeneity problem of the WOE encoding will be alleviated, but the discrimination ability of the variable will be weakened, that is, the IV value will decline. Therefore, the essence of the contradiction between the IV index and the WOE encoding is that the relationship between the discrimination ability and the number of bins, and the relationship between the endogeneity and the number of bins are inverse. To solve the related contradictions, it is necessary to evaluate the initial classification results to obtain multiple evaluations, and then find the optimal number of bins (i.e., the target classification quantity) according to the multiple evaluation values.

[0080] If there are m initial classification results corresponding to each classification quantity for multiple objects to be classified (i.e., m bins are made for the independent variable), then for any one of the n samples (i.e., n objects to be classified), if its independent variable (i.e., attribute information) belongs to the i-th interval , then its encoded variable (i.e., encoded attribute information) will be assigned , and the dependent variable is (i.e., historical behavior information), which can be expressed as follows:

[0081]

[0082] According to the encoded variable , independent variable and dependent variable for calculation, the endogeneity index values of the initial classification results corresponding to each classification quantity for multiple objects to be classified can be obtained (i.e., the first evaluation value). The larger this index value, the less there is endogeneity of mutual causation, that is, the better the endogeneity problem is alleviated.

[0083] Mark the m initial classification results as respectively, and construct the difference index corresponding to the initial classification results, and the expression of

[0084]

[0085] is as follows: where represents the difference index corresponding to the initial classification results, represents the number of all samples with the dependent variable equal to 1, represents the number of all samples with the dependent variable equal to 0, SSA is the sum of squared between-group deviations in statistics, and the meanings of each component of this index are as follows:

[0086] After simplifying the difference index, it can be known that follows a chi-square distribution, that is:

[0087]

[0088] Therefore, the difference index values (i.e., the second evaluation value) of the initial classification results corresponding to each classification quantity for multiple objects to be classified can be obtained, and its expression is as follows:

[0089]

[0090] The larger the value of this index, the more obvious the difference between the final samples, that is, the better the discrimination ability.

[0091] By multiplying the endogeneity index value and the difference index value, a comprehensive index value (i.e., multiple evaluation values) corresponding to the initial classification results of multiple objects to be classified under each classification quantity can be obtained. The larger the value of this index, the more the number of bins can weigh and relieve endogeneity and improve the advantages and disadvantages of discrimination ability.

[0092] Among them, the comprehensive index value has the following expression:

[0093]

[0094] By multiplying and calculating the endogeneity index value and the difference index value, the evaluation values of the initial classification results corresponding to multiple objects to be classified under each classification quantity can be obtained quickly and accurately.

[0095] Optionally, in the object classification method provided in the embodiments of the present application, calculating based on historical behavior information, encoded attribute information, and attribute information to obtain a first evaluation value corresponding to the initial classification result includes: calculating based on the encoded attribute information and the attribute information to obtain a random error term corresponding to each object to be classified; calculating based on the random error term, the encoded attribute information, and the historical behavior information to obtain a first calculation result; calculating based on the encoded attribute information and the historical behavior information to obtain a second calculation result; calculating based on the first calculation result and the second calculation result to obtain a first evaluation value.

[0096] In an alternative embodiment, construct a unary linear equation about the encoded variable and the original independent variable , and the expression is as follows:

[0097]

[0098] Among them, is the disturbance term (i.e., the random error term), and η and λ are fixed parameters.

[0099] After statistical operations, the estimated values of the parameters η and λ can be obtained. Let the estimated values be and . Then, according to the unary linear equation, the estimated quantity of the disturbance term of each sample can be obtained, and the expression is as follows:

[0100]

[0101] Construct the encoded variables separately With and without disturbance terms , and the dependent variable for the simple linear equation.

[0102] The encoded variable In the case of the existence of disturbance terms , and the dependent variable The expression of the simple linear equation is as follows:

[0103]

[0104] where α, β, ρ, and ζ are fixed parameters, represents the encoded variable, represents the disturbance term.

[0105] It can then be expressed as:

[0106]

[0107] where represents the probability that the dependent variable equals 1, represents the probability that the dependent variable equals 0, α, β, and ρ are fixed parameters, represents the encoded variable, represents the disturbance term.

[0108] Thus, the log-likelihood function , The expression is as follows:

[0109]

[0110] The maximum value of the log-likelihood function can then be obtained (i.e., the first calculation result).

[0111] The encoded variable In the case of the non-existence of disturbance terms , and the dependent variable The expression of the simple linear equation is as follows:

[0112]

[0113] where α, β, and ζ are fixed parameters, represents the encoded variable.

[0114] It can then be expressed as:

[0115]

[0116] wherein, represents the probability that the dependent variable equals 1, represents the probability that the dependent variable equals 0, α and β are fixed parameters, represents the encoded variable.

[0117] Thus, the log-likelihood function can be constructed, and the expression of

[0118]

[0119] is as follows: The maximum value of the log-likelihood function (i.e., the second calculation result) can be obtained.

[0120] According to the Taylor expansion and the law of large numbers, it can be known that when the amount of data is large enough, the difference between the maximum values of the above two log-likelihood functions follows a chi-square distribution, and the expression is as follows:

[0121]

[0122] Therefore, by calculating based on the difference between the maximum values of the above two log-likelihood functions, the value of the endogeneity index corresponding to the initial classification result (i.e., the first index value) can be obtained.

[0123] wherein, the expression of the value of the endogeneity index is as follows:

[0124]

[0125] By calculating through the historical behavior information, the encoded attribute information, and the attribute information, the first evaluation value corresponding to the initial classification result can be obtained quickly and accurately.

[0126] Optionally, in the classification method of the object provided in the embodiment of the present application, determining the target classification result based on the initial classification result corresponding to the target classification number includes: screening the multiple objects to be classified and the initial classification result according to the preset screening conditions to obtain the first sample quantity corresponding to the multiple screened objects to be classified and the second sample quantity corresponding to the screened initial classification result; calculating based on the first sample quantity and the second sample quantity to obtain the credibility index value corresponding to the initial classification result; adjusting the initial classification result according to the credibility index value to obtain the adjusted initial classification result; and determining the adjusted initial classification result as the target classification result.

[0127] In an optional embodiment, the maximum evaluation value is selected from multiple evaluation values, and the number of classifications corresponding to this evaluation value is determined as the optimal number of bins (i.e., the target number of classifications). For the initial classification result corresponding to the target number of classifications, WOE encoding correction is performed. The WOE encoding correction process is as follows:

[0128] First, according to the WOE values of (i.e., the target number of classifications) intervals calculate the average WOE value of intervals, and the expression of

[0129]

[0130] For n samples, obtain the maximum value max(x) and the minimum value min(x) of the dependent variable in the samples, and calculate the sampling interval interval(x) for the samples in the ideal state. The expression is as follows:

[0131]

[0132] For n samples, for each sample, calculate the minimum value of the difference between the independent variables of other samples and this sample , and the expression is as follows:

[0133]

[0134] Calculate the number of samples less than or equal to interval(x) among n samples (i.e., the first sample quantity), and the expression is as follows:

[0135]

[0136] It should be noted that the samples with a moderate distance from other samples are simply referred to as uniform points. Then indicates that there are uniform points among n samples. Thus reflects the density of the spatial point group. Assuming that the of the overall samples is an unbiased estimate of the true parameter, this means that in the current overall space, there are denser spatial point groups and sparser spatial point groups. The density of the spatial point group also means that if true random sampling is achieved, the density of the obtained sampling point group should be similar to the density of the spatial point group; otherwise, the sampling is considered to be biased.

[0137] For each interval For each sample within it, calculate the minimum value of the difference between the independent variables of other samples and this sample within this interval , and its expression is as follows:

[0138]

[0139] For each interval , calculate within this interval the number of samples less than or equal to interval(x) (i.e., the second sample quantity), and its expression is as follows:

[0140]

[0141] It should be noted that reflects the density situation within this interval, so reflects the overall density situation inferred according to the density situation of this interval.

[0142] According to the geometric meaning of the Poisson distribution, for each interval , calculate the credibility index value , and its expression is as follows:

[0143]

[0144] Among them, represents the rounding operation on , represents the continuous multiplication of all natural numbers from 1 to , Similarly, represents the number of samples in the sample less than or equal to interval(x), represents the number of samples within the interval less than or equal to interval(x).

[0145] Among them, the simplified expression is as follows:

[0146]

[0147] Among them, represents the rounding operation on , represents the continuous multiplication of all natural numbers from 1 to , Similarly, represents from to the continuous multiplication of all natural numbers, Similarly, represents the number of samples in the sample The number of samples less than or equal to interval(x), indicating within the interval the number of samples less than or equal to interval(x).

[0148] It should be noted that the geometric meaning of the Poisson distribution indicates that given the selection of n samples from the overall space, the expected value of the number of uniform points is , and then, the number of uniform point samples selected from n samples should follow the Poisson distribution. At the same time, reflects the density of the spatial point group. If true random sampling is achieved, the density of the sampling point group should be similar to the density of the spatial point group; otherwise, the sampling is considered biased.

[0149] Regarding each bin of the binned results as resampling, focusing on a certain interval, the density inferred based on this interval is , and thus the probability value can be obtained. This probability value is divided by the peak value of the Poisson distribution , and finally the obtained represents the similarity between the density of this interval and the density of the overall space. The larger it is, the more it indicates that the sampling method in this interval follows the point group distribution law of the overall space density and is credible.

[0150] Therefore, intervals can calculate number of , and output the corrected woe coding value for each interval , in order to obtain the target classification result, The expression of which is as follows:

[0151]

[0152] It should be noted that traditional classification results rely on WOE coding, but traditional WOE coding may have extreme values due to sampling bias. For example: If the density of the overall space is uniform, a certain interval is (0, 100], and there are 100 samples falling into this interval, but 90 samples are between (0, 10], and the remaining samples are scattered between (10, 100], then the WOE value of the entire interval is not very credible because there are a large number of blanks between (10, 100], which may be caused by improper sampling, resulting in the sample information between (10, 100] not being fully incorporated into the coding value. The WOE value of such an interval is extremely likely to have extreme values. Therefore, it is necessary to correct and adjust the WOE coding.

[0153] Through the credibility index value, the initial classification result can be adjusted quickly and accurately, and then the target classification result can be obtained efficiently.

[0154] The classification method of the object provided by the embodiment of the present application calculates based on the attribute information of multiple objects to be classified and the historical behavior information of each object to be classified in the target institution, and obtains the cumulative function curves corresponding to the multiple objects to be classified under the attribute information; solves the cumulative function curves respectively according to multiple classification quantities, and obtains the initial classification results corresponding to the multiple objects to be classified under each classification quantity; evaluates the initial classification results corresponding to the multiple objects to be classified under each classification quantity according to the attribute information and the historical behavior information, and obtains multiple evaluation values; determines the target classification quantity from the multiple classification quantities according to the multiple evaluation values, and determines the target classification result according to the initial classification result corresponding to the target classification quantity, wherein the target institution determines the processing strategies corresponding to the multiple objects to be classified according to the target classification result, and solves the problem that in the related art, the target object is classified manually, resulting in a relatively low accuracy of classifying the target object.

[0155] In summary, in this solution, based on the attribute information corresponding to multiple objects to be classified and their historical behavior information in the target institution, the cumulative function curves corresponding to each object to be classified under the attribute information are constructed. Then, through a preset number of classification quantities, the cumulative function curves are solved respectively to obtain the initial classification results corresponding to the multiple objects to be classified under each classification quantity. Finally, the initial classification results under each classification quantity are evaluated to determine the appropriate classification quantity (i.e., the target classification quantity) and the classification result (i.e., the target classification result).

[0156] Through the attribute information and historical behavior information of multiple objects to be classified, the cumulative function curves corresponding to the multiple objects to be classified under the attribute information can be constructed quickly, so as to better capture the characteristics and change trends of the objects to be classified. At the same time, by solving the cumulative function curves respectively based on multiple classification quantities, the initial classification results corresponding to the multiple objects to be classified under each classification quantity can be obtained more intuitively, which helps to further analyze and understand the characteristics and behaviors of the objects to be classified, improves the accuracy and reliability of classification, and helps the decision maker to make a correct classification decision better.

[0157] It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although the logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in a different order than here.

[0158] The embodiments of the present application also provide an object classification device. It should be noted that the object classification device in the embodiments of the present application can be used to execute the object classification method provided in the embodiments of the present application. The following introduces the object classification device provided in the embodiments of the present application.

[0159] Figure 4 It is a schematic diagram of the object classification device according to the embodiments of the present application. As Figure 4 shown, the device includes: a calculation unit 401, a solution unit 402, an evaluation unit 403, and a determination unit 404.

[0160] The calculation unit 401 is configured to calculate based on the attribute information of multiple objects to be classified and the historical behavior information of each object to be classified in the target institution, and obtain the cumulative function curves corresponding to the multiple objects to be classified under the attribute information;

[0161] The solution unit 402 is configured to solve the cumulative function curves respectively according to multiple classification quantities, and obtain the initial classification results corresponding to the multiple objects to be classified under each classification quantity;

[0162] The evaluation unit 403 is configured to evaluate the initial classification results corresponding to the multiple objects to be classified under each classification quantity based on the attribute information and the historical behavior information, and obtain multiple evaluation values;

[0163] The determination unit 404 is configured to determine the target classification quantity from multiple classification quantities based on multiple evaluation values, and determine the target classification result according to the initial classification result corresponding to the target classification quantity, wherein the target institution determines the processing strategies corresponding to the multiple objects to be classified according to the target classification result.

[0164] For the object classification device provided in the embodiments of the present application, the calculation unit 401 calculates based on the attribute information of multiple objects to be classified and the historical behavior information of each object to be classified in the target institution, and obtains the cumulative function curves corresponding to the multiple objects to be classified under the attribute information; the solution unit 402 solves the cumulative function curves respectively according to multiple classification quantities, and obtains the initial classification results corresponding to the multiple objects to be classified under each classification quantity; the evaluation unit 403 evaluates the initial classification results corresponding to the multiple objects to be classified under each classification quantity based on the attribute information and the historical behavior information, and obtains multiple evaluation values; the determination unit 404 determines the target classification quantity from multiple classification quantities based on multiple evaluation values, and determines the target classification result according to the initial classification result corresponding to the target classification quantity, wherein the target institution determines the processing strategies corresponding to the multiple objects to be classified according to the target classification result, which solves the problem in the related art that the accuracy of classifying the target object is relatively low by manually classifying the target object.

[0165] In this solution, based on the attribute information of multiple objects to be classified and their historical behavior information in the target institution, a cumulative function curve corresponding to each object to be classified under the attribute information is constructed. Then, through a preset number of classification quantities, the cumulative function curve is solved respectively to obtain the initial classification results corresponding to each object to be classified under each classification quantity. Finally, the initial classification results under each classification quantity are evaluated to determine the appropriate classification quantity (i.e., the target classification quantity) and the classification result (i.e., the target classification result).

[0166] Through the attribute information and historical behavior information of multiple objects to be classified, the cumulative function curves corresponding to multiple objects to be classified under the attribute information can be quickly constructed, so as to better capture the characteristics and change trends of the objects to be classified. At the same time, by solving the cumulative function curve based on multiple classification quantities respectively, the initial classification results corresponding to each object to be classified under each classification quantity can be obtained more intuitively, which helps to further analyze and understand the characteristics and behaviors of the objects to be classified, improves the accuracy and reliability of classification, and helps the decision maker make a correct classification decision better.

[0167] Optionally, in the object classification device provided in the embodiment of the present application, the calculation unit includes: a first determination module for determining a first value corresponding to multiple objects to be classified according to the attribute information; a first calculation module for calculating according to the historical behavior information of each object to be classified in the target institution to obtain a second value; a second determination module for obtaining the cumulative function curve according to the first value and the second value.

[0168] Optionally, in the object classification device provided in the embodiment of the present application, the solution unit includes: a third determination module for determining the starting point in the cumulative function curve; a solution module for, for each classification quantity, solving the second value corresponding to the historical behavior information of multiple objects to be classified in the starting point and the cumulative function curve through a target cutting algorithm to obtain a first coordinate point, and determining the classification range according to the first coordinate point and the starting point; a judgment module for judging whether the classification quantity after classifying multiple objects to be classified under the classification range is equal to the classification quantity; an execution module for, if the classification quantity is not equal to the classification quantity, repeatedly executing the step of solving the second value corresponding to the historical behavior information of multiple objects to be classified in the first coordinate point and the cumulative function curve through the target cutting algorithm to obtain a second coordinate point until the initial classification result corresponding to the classification quantity is obtained.

[0169] Optionally, in the object classification device provided in the embodiments of the present application, the solution module includes: a first calculation sub-module, configured to, for each classification quantity, calculate a cumulative function curve based on a first sensitivity coefficient, a second sensitivity coefficient, and a starting point through a target cutting algorithm to obtain a candidate point set, where the first sensitivity coefficient is used to characterize the mean difference degree between historical behavior information corresponding to each category after classification, and the second sensitivity coefficient is used to characterize the quantity difference degree of objects corresponding to each category after classification; a second calculation sub-module, configured to calculate, according to the IV index, multiple candidate coordinate points in the candidate point set to obtain an index value corresponding to each candidate coordinate point; a determination sub-module, configured to determine a first coordinate point from the multiple candidate coordinate points in the candidate point set according to the index value corresponding to each candidate coordinate point.

[0170] Optionally, in the object classification device provided in the embodiments of the present application, the evaluation unit includes: an encoding module, configured to encode the attribute information of multiple objects to be classified to obtain encoded attribute information; a second calculation module, configured to calculate, according to historical behavior information, the encoded attribute information, and the attribute information, to obtain a first evaluation value corresponding to the initial classification result; a third calculation module, configured to calculate, according to the attribute information and the initial classification result, to obtain a second evaluation value corresponding to the initial classification result; a fourth calculation module, configured to calculate, according to the first evaluation value and the second evaluation value, to obtain multiple evaluation values.

[0171] Optionally, in the object classification device provided in the embodiments of the present application, the second calculation module includes: a third calculation sub-module, configured to calculate, according to the encoded attribute information and the attribute information, to obtain a random error term corresponding to each object to be classified; a fourth calculation sub-module, configured to calculate, according to the random error term, the encoded attribute information, and the historical behavior information, to obtain a first calculation result; a fifth calculation sub-module, configured to calculate, according to the encoded attribute information and the historical behavior information, to obtain a second calculation result; a sixth calculation sub-module, configured to calculate, according to the first calculation result and the second calculation result, to obtain a first evaluation value.

[0172] Optionally, in the object classification device provided in the embodiments of the present application, the determination unit includes: a screening module, configured to screen multiple objects to be classified and the initial classification result according to a preset screening condition to obtain a first sample quantity corresponding to the screened multiple objects to be classified and a second sample quantity corresponding to the screened initial classification result; a fifth calculation module, configured to calculate, according to the first sample quantity and the second sample quantity, to obtain a credibility index value corresponding to the initial classification result; an adjustment module, configured to adjust the initial classification result according to the credibility index value to obtain an adjusted initial classification result; a fourth determination module, configured to determine the adjusted initial classification result as the target classification result.

[0173] The classification device of an object includes a processor and a memory. The above-mentioned computing unit 401, solving unit 402, evaluation unit 403, determination unit 404, etc. are all stored in the memory as program units, and the processor executes the above program units stored in the memory to achieve accurate classification of the object.

[0174] The processor contains a kernel, and the kernel retrieves the corresponding program unit from the memory. One or more kernels can be set, and accurate classification of the object can be achieved by adjusting the kernel parameters.

[0175] The memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of, for example, read-only memory (ROM) or flash memory (flash RAM). The memory includes at least one storage chip.

[0176] An embodiment of the present invention provides a computer-readable storage medium with a program stored thereon. When the program is executed by a processor, it implements a method for classifying an object.

[0177] An embodiment of the present invention provides a processor for running a program. When the program runs, it executes a method for classifying an object.

[0178] As Figure 5 shown, an embodiment of the present invention provides an electronic device. The device includes a processor, a memory, and a program stored on the memory and executable on the processor. When the processor executes the program, the following steps are implemented: calculating based on the attribute information of multiple objects to be classified and the historical behavior information of each object to be classified in a target institution to obtain cumulative function curves corresponding to the multiple objects to be classified under the attribute information; solving the cumulative function curves respectively according to multiple classification quantities to obtain initial classification results corresponding to the multiple objects to be classified under each classification quantity; evaluating the initial classification results corresponding to the multiple objects to be classified under each classification quantity based on the attribute information and the historical behavior information to obtain multiple evaluation values; determining a target classification quantity from the multiple classification quantities based on the multiple evaluation values, and determining a target classification result based on the initial classification result corresponding to the target classification quantity, where the target institution determines a processing strategy corresponding to the multiple objects to be classified based on the target classification result.

[0179] Optionally, calculating based on the attribute information of multiple objects to be classified and the historical behavior information of each object to be classified in a target institution to obtain cumulative function curves corresponding to the multiple objects to be classified under the attribute information includes: determining a first value corresponding to the multiple objects to be classified based on the attribute information; calculating a second value based on the historical behavior information of each object to be classified in the target institution; and obtaining the cumulative function curve based on the first value and the second value.

[0180] Optionally, the cumulative function curves are solved respectively according to multiple classification quantities, and the initial classification results corresponding to each classification quantity for multiple objects to be classified are obtained, including: determining the starting point in the cumulative function curve; for each classification quantity, solving the first coordinate point by using a target cutting algorithm for the starting point and the second numerical values corresponding to the historical behavior information of multiple objects to be classified in the cumulative function curve, and determining the classification range according to the first coordinate point and the starting point; judging whether the classification quantity after classifying multiple objects to be classified within the classification range is equal to this classification quantity; if the classification quantity is not equal to this classification quantity, repeat the step of solving the second coordinate point by using the target cutting algorithm for the first coordinate point and the second numerical values corresponding to the historical behavior information of multiple objects to be classified in the cumulative function curve until the initial classification result corresponding to this classification quantity is obtained.

[0181] Optionally, for each classification quantity, solving the first coordinate point by using a target cutting algorithm for the starting point and the second numerical values corresponding to the historical behavior information of multiple objects to be classified in the cumulative function curve includes: for each classification quantity, calculating the cumulative function curve by using the target cutting algorithm based on the first sensitivity coefficient, the second sensitivity coefficient and the starting point to obtain a candidate point set, where the first sensitivity coefficient is used to characterize the mean difference degree between the historical behavior information corresponding to each category after classification, and the second sensitivity coefficient is used to characterize the quantity difference degree of the objects corresponding to each category after classification; calculating the index value corresponding to each candidate coordinate point in the candidate point set according to the IV index; determining the first coordinate point from multiple candidate coordinate points in the candidate point set according to the index value corresponding to each candidate coordinate point.

[0182] Optionally, according to the attribute information and the historical behavior information, evaluating the initial classification results corresponding to each classification quantity for multiple objects to be classified to obtain multiple evaluation values, including: encoding the attribute information of multiple objects to be classified to obtain the encoded attribute information; calculating according to the historical behavior information, the encoded attribute information and the attribute information to obtain the first evaluation value corresponding to the initial classification result; calculating according to the attribute information and the initial classification result to obtain the second evaluation value corresponding to the initial classification result; calculating according to the first evaluation value and the second evaluation value to obtain multiple evaluation values.

[0183] Optionally, calculating a first evaluation value corresponding to an initial classification result based on historical behavior information, encoded attribute information, and attribute information includes: calculating, based on the encoded attribute information and the attribute information, a random error term corresponding to each object to be classified; calculating a first calculation result based on the random error term, the encoded attribute information, and the historical behavior information; calculating a second calculation result based on the encoded attribute information and the historical behavior information; and calculating the first evaluation value based on the first calculation result and the second calculation result.

[0184] Optionally, determining a target classification result based on an initial classification result corresponding to a target classification quantity includes: screening multiple objects to be classified and the initial classification result according to a preset screening condition to obtain a first sample quantity corresponding to the screened multiple objects to be classified and a second sample quantity corresponding to the screened initial classification result; calculating a credibility index value corresponding to the initial classification result based on the first sample quantity and the second sample quantity; adjusting the initial classification result according to the credibility index value to obtain an adjusted initial classification result; and determining the adjusted initial classification result as the target classification result.

[0185] The devices in this article can be servers, PCs, PADs, mobile phones, etc.

[0186] This application also provides a computer program product, which, when executed on a data processing device, is adapted to execute a program initialized with the following method steps: calculating, based on the attribute information of multiple objects to be classified and the historical behavior information of each object to be classified in a target institution, a cumulative function curve corresponding to the multiple objects to be classified under the attribute information; respectively solving the cumulative function curve according to multiple classification quantities to obtain an initial classification result corresponding to each object to be classified under each classification quantity; evaluating the initial classification result corresponding to each object to be classified under each classification quantity based on the attribute information and the historical behavior information to obtain multiple evaluation values; determining a target classification quantity from the multiple classification quantities based on the multiple evaluation values, and determining a target classification result based on the initial classification result corresponding to the target classification quantity, wherein the target institution determines a processing strategy corresponding to the multiple objects to be classified according to the target classification result.

[0187] Optionally, calculating, based on the attribute information of multiple objects to be classified and the historical behavior information of each object to be classified in a target institution, a cumulative function curve corresponding to the multiple objects to be classified under the attribute information includes: determining a first numerical value corresponding to the multiple objects to be classified based on the attribute information; calculating a second numerical value based on the historical behavior information of each object to be classified in the target institution; and obtaining the cumulative function curve based on the first numerical value and the second numerical value.

[0188] Optionally, the cumulative function curves are solved respectively according to multiple classification quantities, and the initial classification results corresponding to each classification quantity for multiple objects to be classified are obtained, including: determining the starting point in the cumulative function curve; for each classification quantity, solving the first coordinate point by using a target cutting algorithm for the starting point and the second numerical values corresponding to the historical behavior information of multiple objects to be classified in the cumulative function curve, and determining the classification range according to the first coordinate point and the starting point; judging whether the classification quantity after classifying multiple objects to be classified within the classification range is equal to this classification quantity; if the classification quantity is not equal to this classification quantity, then repeat the step of solving the second coordinate point by using the target cutting algorithm for the first coordinate point and the second numerical values corresponding to the historical behavior information of multiple objects to be classified in the cumulative function curve until the initial classification result corresponding to this classification quantity is obtained.

[0189] Optionally, for each classification quantity, solving the first coordinate point by using a target cutting algorithm for the starting point and the second numerical values corresponding to the historical behavior information of multiple objects to be classified in the cumulative function curve includes: for each classification quantity, calculating the cumulative function curve by using the target cutting algorithm based on the first sensitivity coefficient, the second sensitivity coefficient and the starting point to obtain a candidate point set, where the first sensitivity coefficient is used to characterize the mean difference degree between the historical behavior information corresponding to each category after classification, and the second sensitivity coefficient is used to characterize the quantity difference degree of the objects corresponding to each category after classification; calculating each candidate coordinate point in the candidate point set according to the IV index to obtain the index value corresponding to each candidate coordinate point; determining the first coordinate point from multiple candidate coordinate points in the candidate point set according to the index value corresponding to each candidate coordinate point.

[0190] Optionally, evaluating the initial classification results corresponding to each classification quantity for multiple objects to be classified according to the attribute information and the historical behavior information to obtain multiple evaluation values includes: encoding the attribute information of multiple objects to be classified to obtain the encoded attribute information; calculating according to the historical behavior information, the encoded attribute information and the attribute information to obtain the first evaluation value corresponding to the initial classification result; calculating according to the attribute information and the initial classification result to obtain the second evaluation value corresponding to the initial classification result; calculating according to the first evaluation value and the second evaluation value to obtain multiple evaluation values.

[0191] Optionally, calculating a first evaluation value corresponding to an initial classification result based on historical behavior information, encoded attribute information, and attribute information includes: calculating, based on the encoded attribute information and the attribute information, a random error term corresponding to each object to be classified; calculating a first calculation result based on the random error term, the encoded attribute information, and the historical behavior information; calculating a second calculation result based on the encoded attribute information and the historical behavior information; and calculating the first evaluation value based on the first calculation result and the second calculation result.

[0192] Optionally, determining a target classification result based on an initial classification result corresponding to a target classification number includes: screening multiple objects to be classified and the initial classification result according to a preset screening condition to obtain a first sample number corresponding to the screened multiple objects to be classified and a second sample number corresponding to the screened initial classification result; calculating a credibility index value corresponding to the initial classification result based on the first sample number and the second sample number; adjusting the initial classification result according to the credibility index value to obtain an adjusted initial classification result; and determining the adjusted initial classification result as the target classification result.

[0193] Those skilled in the art should understand that the embodiments of the present application can be provided as a method, a system, or a computer program product. Therefore, the present application can adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0194] The present application is described with reference to the flowcharts and / or block diagrams of methods, apparatuses (systems), and computer program products according to embodiments of the present application. It should be understood that each process and / or block in the flowcharts and / or block diagrams, and the combination of processes and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, such that the instructions executed by the processor of the computer or other programmable data processing devices generate means for implementing the functions specified in Figure 1 one process or multiple processes and / or blocks Figure 1 one block or multiple blocks.

[0195] These computer program instructions can also be stored in a computer-readable memory capable of guiding a computer or other programmable data processing device to work in a specific manner, such that the instructions stored in the computer-readable memory generate a manufactured article including an instruction device, and the instruction device implements the functions in the processFigure 1 one or more processes and / or blocks Figure 1 functions specified in one or more blocks

[0196] These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus, causing a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process, so that the instructions executed on the computer or other programmable apparatus provide steps for implementing the functions specified in one or more processes Figure 1 one or more processes and / or blocks Figure 1 steps for implementing the functions specified in one or more blocks

[0197] In a typical configuration, a computing device includes one or more processors (CPUs), an input / output interface, a network interface, and memory

[0198] The memory may include non-permanent memory in the form of computer-readable media, random access memory (RAM) and / or non-volatile memory such as read-only memory (ROM) or flash memory (flash RAM). Memory is an example of computer-readable media

[0199] Computer-readable media includes both permanent and non-permanent, removable and non-removable media implemented by any method or technology for storing information. The information may be computer-readable instructions, data structures, program modules, or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, compact disc read-only memory (CD-ROM), digital versatile discs (DVD) or other optical storage, magnetic cassettes, magnetic tape disk storage or other magnetic storage devices, or any other non-transmission media that can be used to store information accessible by a computing device. As defined herein, computer-readable media does not include transitory computer-readable media such as modulated data signals and carrier waves

[0200] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0201] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems or computer program products. Therefore, the present application may take the form of a complete hardware embodiment, a complete software embodiment or an embodiment combining software and hardware. Moreover, the present application may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes.

[0202] The above are only embodiments of the present application and are not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included within the scope of the claims of the present application.

Claims

1. A method for classifying an object, characterized in that: include: Calculating based on the attribute information of multiple objects to be classified and the historical behavior information of each object to be classified in the target organization, obtaining the cumulative function curve corresponding to the multiple objects to be classified under the attribute information; Solving the cumulative function curves respectively according to multiple classification numbers to obtain initial classification results corresponding to the multiple objects to be classified under each classification number; According to the attribute information and the historical behavior information, the initial classification results corresponding to the multiple objects to be classified under each classification quantity are evaluated to obtain multiple evaluation values; Determining a target classification number from the multiple classification numbers according to the multiple evaluation values, and determining a target classification result according to an initial classification result corresponding to the target classification number, wherein the target mechanism determines a processing strategy corresponding to the multiple objects to be classified according to the target classification result; Wherein, performing calculations based on the attribute information of a plurality of objects to be classified and the historical behavior information of each object to be classified in the target organization to obtain the cumulative function curve corresponding to the plurality of objects to be classified under the attribute information includes: determining the first values ​​corresponding to the plurality of objects to be classified based on the attribute information; performing calculations based on the historical behavior information of each object to be classified in the target organization to obtain a second value; and obtaining the cumulative function curve based on the first value and the second value; Solving the cumulative function curve according to multiple classification quantities respectively to obtain the initial classification results corresponding to the multiple objects to be classified under each classification quantity includes: determining a starting point in the cumulative function curve; for each classification quantity, solving the second numerical value corresponding to the starting point and the historical behavior information of the multiple objects to be classified in the cumulative function curve by a target cutting algorithm to obtain a first coordinate point, and determining a classification range according to the first coordinate point and the starting point; judging whether the number of classifications after classifying the multiple objects to be classified under the classification range is equal to the number of classifications; if the number of classifications is not equal to the number of classifications, repeatedly executing the step of solving the first coordinate point and the second numerical value corresponding to the historical behavior information of the multiple objects to be classified in the cumulative function curve by the target cutting algorithm to obtain a second coordinate point, until the initial classification result corresponding to the classification quantity is obtained; The object to be classified is a user of the target organization, the target organization is an Internet organization, or a media organization, or a game organization, the attribute information includes age information and occupation information, and the historical behavior information includes whether the user has potential risks in the target organization; Wherein, for each number of classifications, the target cutting algorithm is used to solve the second numerical value corresponding to the historical behavior information of the plurality of objects to be classified in the starting point and the cumulative function curve, and the first coordinate point is obtained including: For each number of classifications, the target cutting algorithm is used to calculate the cumulative function curve based on the first sensitivity coefficient, the second sensitivity coefficient and the starting point to obtain a candidate point set, wherein the first sensitivity coefficient is used to characterize the degree of difference in the mean values ​​of the historical behavior information corresponding to each category after classification, and the second sensitivity coefficient is used to characterize the degree of difference in the number of objects corresponding to each category after classification; Calculating multiple candidate coordinate points in the candidate point set according to the IV index to obtain an index value corresponding to each candidate coordinate point; Determine the first coordinate point from a plurality of candidate coordinate points in the candidate point set according to an index value corresponding to each candidate coordinate point; Based on the attribute information and the historical behavior information, the initial classification results corresponding to the multiple objects to be classified under each classification quantity are evaluated to obtain multiple evaluation values ​​including: Encoding the attribute information of the plurality of objects to be classified to obtain encoded attribute information; Calculating according to the historical behavior information, the encoded attribute information and the attribute information to obtain a first evaluation value corresponding to the initial classification result; Calculating according to the attribute information and the initial classification result to obtain a second evaluation value corresponding to the initial classification result; The plurality of evaluation values ​​are obtained by performing calculations according to the first evaluation value and the second evaluation value.

2. The method according to claim 1, characterized in that Calculating according to the historical behavior information, the encoded attribute information, and the attribute information to obtain a first evaluation value corresponding to the initial classification result includes: Calculating according to the encoded attribute information and the attribute information to obtain a random error term corresponding to each object to be classified; Performing calculation according to the random error term, the encoded attribute information and the historical behavior information to obtain a first calculation result; Performing calculation based on the encoded attribute information and the historical behavior information to obtain a second calculation result; Calculation is performed according to the first calculation result and the second calculation result to obtain the first evaluation value.

3. The method according to claim 1, characterized in that Determining the target classification results according to the initial classification results corresponding to the target classification quantity includes: Screening the multiple objects to be classified and the initial classification results according to a preset screening condition to obtain a first sample quantity corresponding to the screened multiple objects to be classified and a second sample quantity corresponding to the screened initial classification results; Calculating according to the first sample quantity and the second sample quantity to obtain a credibility index value corresponding to the initial classification result; Adjusting the initial classification result according to the credibility index value to obtain an adjusted initial classification result; The adjusted initial classification result is determined as the target classification result.

4. A device for classifying objects, characterized in that: The classification device is used to perform the object classification method according to claim 1, comprising: A calculation unit, used to calculate according to the attribute information of the plurality of objects to be classified and the historical behavior information of each object to be classified in the target organization, to obtain the cumulative function curve corresponding to the plurality of objects to be classified under the attribute information; A solving unit, used for solving the cumulative function curve respectively according to a plurality of classification quantities to obtain initial classification results corresponding to the plurality of objects to be classified under each classification quantity; An evaluation unit, configured to evaluate the initial classification results corresponding to each number of classifications of the plurality of objects to be classified according to the attribute information and the historical behavior information, to obtain a plurality of evaluation values; A determination unit, configured to determine a target classification number from among the multiple classification numbers according to the multiple evaluation values, and determine a target classification result according to an initial classification result corresponding to the target classification number, wherein the target mechanism determines a processing strategy corresponding to the multiple objects to be classified according to the target classification result; The object to be classified is a user of the target organization, the target organization is an Internet organization, or a media organization, or a game organization, the attribute information includes age information and occupation information, and the historical behavior information includes whether the user has potential risks in the target organization; The solution module includes: a first calculation submodule, for calculating the cumulative function curve for each classification quantity through the target cutting algorithm based on the first sensitivity coefficient, the second sensitivity coefficient and the starting point to obtain a candidate point set, wherein the first sensitivity coefficient is used to characterize the degree of difference in the mean between the historical behavior information corresponding to each category after classification, and the second sensitivity coefficient is used to characterize the degree of difference in the number of objects corresponding to each category after classification; a second calculation submodule, for calculating multiple candidate coordinate points in the candidate point set according to the IV index to obtain the index value corresponding to each candidate coordinate point; a determination submodule, for determining the first coordinate point from multiple candidate coordinate points in the candidate point set according to the index value corresponding to each candidate coordinate point; The evaluation unit includes: an encoding module, which is used to encode the attribute information of the multiple objects to be classified to obtain the encoded attribute information; a second calculation module, which is used to perform calculations based on the historical behavior information, the encoded attribute information and the attribute information to obtain a first evaluation value corresponding to the initial classification result; a third calculation module, which is used to perform calculations based on the attribute information and the initial classification result to obtain a second evaluation value corresponding to the initial classification result; and a fourth calculation module, which is used to perform calculations based on the first evaluation value and the second evaluation value to obtain the multiple evaluation values.

5. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a stored program, wherein when the program is executed, the storage medium is controlled to execute the object classification method according to any one of claims 1 to 3 on a device.

6. An electronic device, characterized in that: It includes one or more processors and a memory, wherein the memory is used to store one or more programs, wherein when the one or more programs are executed by the one or more processors, the one or more processors implement the object classification method described in any one of claims 1 to 3.

Images