CN111967940B

CN111967940B - Order quantity abnormity detection method and device

Info

Publication number: CN111967940B
Application number: CN202010839818.8A
Authority: CN
Inventors: 韦露娜
Original assignee: Alipay Hangzhou Information Technology Co Ltd
Current assignee: Alipay Hangzhou Information Technology Co Ltd
Priority date: 2020-08-19
Filing date: 2020-08-19
Publication date: 2023-02-21
Anticipated expiration: 2040-08-19
Also published as: CN111967940A

Abstract

The embodiment of the specification discloses an order quantity abnormity detection method, device and equipment. The method is used in the supervision field. The method comprises the following steps: acquiring historical order quantity of industries to be identified in a set historical time period; selecting a specific algorithm according to the historical order quantity to perform fitting analysis, and obtaining a historical order quantity interval of the industry to be identified under normal probability; acquiring the amount of orders to be analyzed in a set time period of the industry to be identified; if the order quantity to be analyzed is located outside the historical order quantity interval, it can be determined that the order quantity to be analyzed of the industry to be identified in the set time period is abnormal.

Description

Order quantity abnormity detection method and device

Technical Field

The present application relates to the field of computer technologies, and in particular, to a method and an apparatus for detecting an order quantity anomaly.

Background

The enterprise electronic commerce platform is a management environment established on the Internet, business activities are carried out through the platform to ensure smooth operation of businesses, the platform is different from a traditional platform, the enterprise electronic commerce platform is a virtual network space, is not limited by time and space, and has high efficiency and straight-through property.

However, the electronic commerce platform has some defects of itself while improving the transaction benefit, wherein most importantly, the virtual and false transactions on the line cannot be monitored powerfully, the false transactions not only pollute real transaction data, but also increase the decision difficulty of a monitoring organization and an enterprise. For example: some merchants want to collect some subsidies or rewards issued by the platform, intentionally generate false orders to satisfy the collection conditions, etc.

Therefore, it is desirable to provide an order quantity anomaly detection scheme to more effectively monitor platform transactions.

Disclosure of Invention

The embodiment of the specification provides an order quantity abnormity detection method and device, and aims to solve the problem that an online transaction of a platform cannot be effectively supervised due to the fact that an existing method cannot detect order quantity abnormity.

In order to solve the above technical problem, the embodiments of the present specification are implemented as follows:

the order quantity abnormality detection method provided by the embodiment of the present specification includes:

acquiring historical order quantity of industries to be identified in a set historical time period;

selecting a specific algorithm according to the historical order quantity to perform fitting analysis, and obtaining a historical order quantity interval of the industry to be identified under normal probability;

acquiring the amount of orders to be analyzed of the industry to be identified in a set time period;

judging whether the order quantity to be analyzed is located in the historical order quantity interval or not to obtain a first judgment result;

and when the first judgment result shows that the order quantity to be analyzed is positioned outside the historical order quantity interval, determining that the order quantity to be analyzed of the industry to be identified is abnormal in a set time period.

An order quantity abnormality detection device provided in an embodiment of the present specification includes:

the historical order quantity acquisition module is used for acquiring the historical order quantity of the industry to be identified in a set historical time period;

the historical order quantity interval determining module is used for selecting a specific algorithm according to the historical order quantity to perform fitting analysis so as to obtain a historical order quantity interval of the industry to be identified under normal probability;

the analysis order quantity obtaining module is used for obtaining the analysis order quantity of the industry to be identified in a set time period;

the first judging module is used for judging whether the order quantity to be analyzed is located in the historical order quantity interval or not to obtain a first judging result;

and the order quantity abnormity determining module is used for determining that the order quantity to be analyzed of the industry to be identified is abnormal in a set time period when the first judgment result shows that the order quantity to be analyzed is positioned outside the historical order quantity interval.

An order quantity anomaly detection device provided by an embodiment of the present specification includes:

at least one processor; and the number of the first and second groups,

a memory communicatively coupled to the at least one processor; wherein the content of the first and second substances,

the memory stores instructions executable by the at least one processor to cause the at least one processor to:

Embodiments of the present specification provide a computer readable medium having stored thereon computer readable instructions executable by a processor to implement an order quantity anomaly detection method.

One embodiment of the present description can achieve the following advantageous effects: obtaining the historical order quantity of industries to be identified in a set historical time period; selecting a specific algorithm according to the historical order quantity to perform fitting analysis, and obtaining a historical order quantity interval of the industry to be identified under normal probability; acquiring the order quantity to be analyzed in a set time period of the industry to be identified; if the order quantity to be analyzed is located outside the historical order quantity interval, it can be determined that the order quantity to be analyzed of the industry to be identified in the set time period is abnormal. By the method, the abnormal conditions of the order quantity of each industry can be detected according to the order quantity of the industry, so that the transaction orders in the platform can be effectively monitored.

Drawings

In order to more clearly illustrate the embodiments of the present disclosure or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments described in the present application, and for those skilled in the art, other drawings can be obtained according to the drawings without any creative effort.

Fig. 1 is an overall framework diagram of an order quantity anomaly detection method provided in an embodiment of the present specification;

FIG. 2 is a flowchart of an order quantity anomaly detection method provided by an embodiment of the present specification;

fig. 3 is a schematic diagram of a normal distribution method in an order quantity anomaly detection method provided in an embodiment of the present specification;

fig. 4 is a schematic structural diagram of an order quantity abnormality detection apparatus provided in an embodiment of the present specification;

fig. 5 is a schematic structural diagram of an order quantity abnormality detection apparatus provided in an embodiment of the present specification.

Detailed Description

To make the objects, technical solutions and advantages of one or more embodiments of the present disclosure more apparent, the technical solutions of one or more embodiments of the present disclosure will be clearly and completely described below with reference to specific embodiments of the present disclosure and the accompanying drawings. It is to be understood that the embodiments described are only a few embodiments of the present specification, and not all embodiments. All other embodiments that can be derived by a person skilled in the art from the embodiments given herein without making any creative effort fall within the protection scope of one or more embodiments of the present disclosure.

The technical solutions provided by the embodiments of the present description are described in detail below with reference to the accompanying drawings.

With the rapid development of electronic commerce and the popularization of mobile terminals, online transactions using mobile devices such as mobile phones and tablets have become a life style of the current young people. With more and more merchants staying on the e-commerce platform, the online transaction order volume is larger and larger, and great difficulty exists in supervision of the e-commerce platform. Most importantly, online false transactions cannot be well identified, and not only are real transaction data polluted by the false transactions, but also the monitoring decision difficulty of authorities and enterprises is increased. For example, merchants in a trading platform generate false orders through false shopping processes or other ways to increase the order volume of merchants, and in this way, merchants can obtain better search ranking, change the reputation of stores, or draw subsidy benefits provided by the platform depending on the order volume, and the like, and in the same industry, many merchants can compete for simulation, resulting in false trading anomalies in the same industry.

The cross-border electronic commerce platform comprises merchants and consumers in the global range, and breaks through national barriers, so that international trade moves to international trade without national border. For enterprises, an open, multi-dimensional and three-dimensional multilateral trading cooperation mode constructed by cross-border electronic commerce greatly widens the path for entering the international market, and greatly promotes the optimal configuration of multilateral resources and the mutual profit and win-win between enterprises. However, it is more difficult to supervise transactions on the platform. If the trade orders on the electronic commerce platform cannot be effectively supervised, the benefits of consumers are damaged, and the users can question the electronic commerce platform; but also influences the decision of the authority on the platform, thereby influencing the development of the e-commerce platform.

Therefore, in order to solve the defects in the prior art, the scheme provides the following embodiments:

fig. 1 is an overall framework diagram of an order quantity anomaly detection method provided in an embodiment of the present specification. As shown in figure 1, the platform can acquire order quantities corresponding to various industries (industry 1, industry 2, industry 3, \8230; industry m). During analysis, the order quantity in a set historical time period (for example, within 3 months) can be obtained, specific algorithms (algorithm 1, algorithm 2, algorithm 3, algorithm 4 and the like) can be selected according to the distribution trend and the specific quantity of the order quantity, the algorithms can be specific methods for abnormality detection, and after the specific algorithms are selected for fitting analysis, whether the order quantity is abnormal or not can be detected.

According to the technical scheme provided by the embodiment of the specification, the distribution trend and the specific quantity of the order quantity in the industry are analyzed, the specific algorithm is selected for fitting analysis, the abnormal condition of the order quantity is finally determined, the abnormal condition of the order quantity corresponding to each industry can be detected, and therefore the transaction orders in the platform can be effectively monitored.

Next, an order quantity abnormality detection method provided in the embodiments of the specification will be specifically described with reference to the accompanying drawings:

fig. 2 is a flowchart of an order quantity abnormality detection method provided in an embodiment of the present specification. From the program perspective, the execution subject of the flow may be a program installed in the application server or an application client. The execution subject in this embodiment may be a server or an application server corresponding to the e-commerce platform.

As shown in fig. 2, the process may include the following steps:

step 202: and acquiring the historical order quantity of industries to be identified in a set historical time period.

It should be noted that the set historical time period may be a time period set by the platform according to an actual application scenario, for example: 10/month 1 of 2019 to 11/month 1 of 2019.

Order data corresponding to all merchants entering the platform can be acquired in the platform, and order quantity corresponding to each industry is analyzed. When the abnormality is identified, the order quantity of all merchants in the platform can be obtained, and after the categories are classified according to industries, the order quantity abnormality of each industry is detected respectively. Of course, the industry to be identified may be determined first, then the order data of the industry is acquired, and the order quantity corresponding to the industry is determined after the order data is analyzed, so as to perform the detection.

Step 204: and selecting a specific algorithm to perform fitting analysis according to the historical order quantity to obtain a historical order quantity interval of the industry to be identified under normal probability.

It should be noted that the specific algorithm herein may refer to an algorithm for performing anomaly detection. For example: a density-based clustering algorithm, a normal Distribution algorithm, an isolated Forest (Isolation Forest) algorithm, a Generalized Pareto Distribution (GPD) algorithm, and the like.

Fitting may be understood as connecting a series of points on a plane with a smooth curve. Because of the myriad possibilities for this curve, there are a variety of fitting methods. If the function to be determined is linear, it may be referred to as linear fitting or linear regression, otherwise it may be referred to as non-linear fitting or non-linear regression. The expression may also be a piecewise function, in which case it may be referred to as a spline fit.

The basic idea of anomaly detection is as follows: if a small probability event occurs, the data is considered to be abnormal, and the abnormal detection is also a mode two-classification method, but the two types of data are seriously unbalanced, and the abnormal data is obviously less than the normal data.

The normal probability may represent a probability threshold set according to a one-month scenario of actual application in the above-described specific algorithm.

After a specific algorithm is selected to perform fitting analysis on the historical order quantity, the order quantity interval under normal probability can be determined. For example: assuming that the analyzed data is order quantity of a certain month in the industry A, and determining that the order quantity interval under the normal probability is 100-10000 after analysis.

Of course, in practical application, the order data of a certain industry within one year can be obtained, the order quantity of each month is determined, the order quantity of the whole year is subjected to fitting analysis according to the month, and after the order quantity interval of each month is obtained, the corresponding order quantity interval with the average normal probability of each month is determined. Specifically, a specific algorithm is selected according to the historical order quantity to perform fitting analysis, and a historical order quantity interval of the industry to be identified under normal probability is obtained.

Step 206: acquiring the amount of orders to be analyzed of the industry to be identified in a set time period; the time length of the set time period is equal to the time length of the historical time period.

It should be noted that the time length of the set time period may be equal to or similar to the length of the historical time period. For example: if the historical time period is one month, the set time period should also be about one month. Such as: the industry to be identified is industry A, the historical time period is 3/1/2019-4/1/2019, and when the order amount of the industry to be identified is analyzed, the set time period can be 3/1/2020-4/1/2020. Of course, the time length may be equal to the time length of the historical time period from 5/10/2020/10 to 6/10/2020/6.

Step 208: and judging whether the order quantity to be analyzed is located in the historical order quantity interval or not to obtain a first judgment result.

Step 210: and when the first judgment result shows that the order quantity to be analyzed is positioned outside the historical order quantity interval, determining that the order quantity to be analyzed of the industry to be identified is abnormal in a set time period.

When judging whether the order quantity of the industry to be identified is abnormal, judging whether the order quantity to be analyzed is in a historical order quantity interval, and if the order quantity to be analyzed is in the historical order quantity interval, determining that the order quantity of the industry is not abnormal. If the order quantity to be analyzed is outside the historical order quantity interval, the industry to be identified can be considered to have order quantity abnormity. For example: along the use example, the analyzed data is assumed to be the order quantity of the industry A in a certain month, the order quantity interval under the normal probability is determined to be 100-10000 after analysis, and when the order quantity to be analyzed is 10 ten thousand, the fact that the order quantity to be analyzed in the industry is abnormal in the set time period can be determined.

It should be understood that the order of some steps in the method described in one or more embodiments of the present disclosure may be interchanged according to actual needs, or some steps may be omitted or deleted.

The method of fig. 2, by obtaining the historical order quantity of the industry to be identified within a set historical time period; selecting a specific algorithm according to the historical order quantity to perform fitting analysis to obtain a historical order quantity interval of the industry to be identified under normal probability; acquiring the amount of orders to be analyzed in a set time period of the industry to be identified; if the order quantity to be analyzed is located outside the historical order quantity interval, it can be determined that the order quantity to be analyzed of the industry to be identified in the set time period is abnormal. By the method, the authenticity of the trade orders in the industry can be detected according to the order quantity of the industry, so that the effective supervision of the trade orders in the platform is realized.

Based on the method of fig. 2, the present specification also provides some specific embodiments of the method, which are described below.

Optionally, before selecting a specific algorithm according to the historical order quantity for fitting analysis, the method may further include:

judging whether the historical order quantity is larger than or equal to a detection threshold value or not to obtain a second judgment result;

when the second judgment result shows that the historical order quantity is larger than or equal to the detection threshold value, carrying out abnormity detection on the historical order quantity;

and when the second judgment result shows that the historical order quantity is smaller than the detection threshold value, not performing abnormity detection on the historical order quantity.

It should be noted that, in the actual anomaly detection process, if the data amount is too small, the detection accuracy is low, and at this time, the detection significance is not great. Therefore, before the abnormality detection is performed, the acquired data volume can be evaluated, and when the data volume of the industry meets the detection threshold, the abnormality detection is performed on the order volume of the industry. The detection threshold here may be defined according to the actual application scenario.

By the method, the order quantity for carrying out the abnormal detection can meet the detection condition, so that the accuracy of the detection result is ensured.

In practical application, when abnormal detection is performed on the order quantity of the industry, a proper specific algorithm needs to be selected for fitting analysis, so that the order quantity interval under normal probability is determined. Wherein, selecting a specific algorithm can be selected based on the following three ways:

the method comprises the steps of selecting a specific algorithm to determine a historical order quantity interval according to the distribution trend of the historical order quantity.

Optionally, the selecting a specific algorithm according to the historical order quantity to perform fitting analysis to obtain a historical order quantity interval of the industry to be identified under a normal probability may specifically include:

and selecting a specific algorithm to perform fitting analysis according to the distribution trend of the historical order quantity to obtain a historical order quantity interval of the industry to be identified under the normal probability corresponding to the specific algorithm.

It should be noted that the distribution trend in the above steps may represent a specific distribution algorithm satisfied after the data is fitted. The centralized trend of data distribution can be determined from the distribution trend of the data, and the degree of the data approaching to or gathering from the central value of the data can be reflected; the dispersion degree of distribution can be determined from the distribution trend of the data, and the trend of each data far away from the central value of the data can be reflected; the shape of the data distribution and the like can also be determined from the data distribution tendency.

At this time, the order quantity obeys the distribution of which algorithm, and the obeyed algorithm can be adopted to perform fitting analysis on the order quantity, so that the order quantity interval under normal probability is determined.

Selecting a specific algorithm according to the distribution trend of the historical order quantity to perform fitting analysis, and obtaining a historical order quantity interval of the industry to be identified under a normal probability corresponding to the specific algorithm, wherein the specific algorithm specifically comprises the following steps:

performing normality check on the historical order quantity, and judging whether the historical order quantity obeys normal distribution or not to obtain a third judgment result;

and when the third judgment result shows that the historical order quantity obeys normal distribution, determining the historical order quantity interval of the industry to be identified under normal probability according to the four-sigma principle of the normal distribution.

It should be noted that the normality check in the above steps may mean generating a normal probability graph according to the obtained order quantity and performing hypothesis testing to check whether the observed value complies with normal distribution. For the normality test, the original hypothesis is H0: the data obeys normal distribution; let us assume H1: data do not follow a normal distribution.

Normal distribution (also known as Gaussian distribution) is a very important probability distribution in the fields of mathematics, physics, engineering, etc. If the random variable obeys the probability distribution of one position parameter and one scale parameter, the probability distribution is recorded as: the probability density function is that the mathematical expected value or the expected value of the normal distribution is equal to the position parameter, and the position of the distribution is determined; the square of the square or standard deviation of its variance is equal to the scale parameter, determining the magnitude of the distribution.

A normal distribution is a distribution of continuous random variables with two parameters, μ and σ ^2, the first parameter μ being the mean of the random variables that follow the normal distribution, and the second parameter σ ^2 being the variance of this random variable, so the normal distribution is denoted as N (μ, σ ^ 2). The probability law of the random variables following normal distribution is that the probability of taking values adjacent to mu is large, and the probability of taking values farther away from mu is smaller; the smaller σ, the more concentrated the distribution is near μ, and the larger σ, the more dispersed the distribution. The normally distributed density function is characterized in that: with respect to μ symmetry, a maximum is reached at μ, a value of 0 at positive (negative) infinity, and an inflection point at μ ± σ. The shape of the image is high in the middle and low on two sides, and the image is a bell-shaped curve located above the x axis. When μ =0, σ ^2=1, it is called a standard normal distribution, and is denoted as N (0,1). When the mu-dimensional random vector has similar probability rules, the random vector is called to follow the multidimensional normal distribution.

The formula for a normal distribution is as follows:

where μ denotes a normal mean value and may describe a central position of the data distribution, and σ denotes a normal variance and may describe a degree of dispersion of the data distribution.

Optionally, the determining the historical order quantity interval of the industry to be identified under the normal probability according to the four-sigma principle of the normal distribution may specifically include:

determining a normal mean value and a normal variance corresponding to the historical order quantity according to the historical order quantity;

determining a four-sigma value according to the normal mean and the normal variance;

and determining the order quantity interval of the industry to be identified within the normal probability according to the four sigma value.

In the above steps, the four-sigma value in the normal distribution method is used to determine the order quantity interval of the industry to be identified within the normal probability, which can be described with reference to fig. 2:

fig. 3 is a schematic diagram of a normal distribution method in an order quantity abnormality detection method provided in an embodiment of the present specification.

As shown in fig. 3, the normal distribution curve has a normal mean μ and a normal variance σ, and is referred to as a standard normal distribution when μ =0 and σ = 1.

Wherein, when describing, sigma can be expressed by sigma (sigma), and within the range of ± sigma, the corresponding probability value is 68.27%; within the range of +/-2 sigma, the corresponding probability value is 95.45 percent; within the range of +/-3 sigma, the corresponding probability value is 99.73 percent; within the range of +/-4 sigma, the corresponding probability value is 99.9937 percent; within the range of +/-5 sigma, the corresponding probability value is 99.99943 percent; within 6 σ, the corresponding probability value is 99.9999998%.

In this embodiment, a historical order quantity interval is determined by using a four-sigma principle, and since the normal probability value is 99.9937% in the range of ± 4 σ, after the order quantity interval of μ ± 4 σ is calculated, the order quantity to be analyzed which finally falls outside the interval is determined to be an abnormal order quantity. Of course, in the actual application process, the interval of the historical order quantity may also be determined by using a five-sigma principle, a six-sigma principle, and the like, and in the actual selection process, the selection may be performed according to a specific application scenario and a required accuracy rate. And drawing a normal distribution curve corresponding to the historical order quantity of the industry to be identified according to the specific quantity of the historical order quantity, the normal mean value and the normal variance.

Optionally, the performing a normality check on the historical order quantity, determining whether the historical order quantity complies with a normal distribution, and after obtaining a third determination result, the method may further include:

when the third judgment result shows that the historical order quantity does not comply with normal distribution, verifying whether the historical order quantity complies with pareto distribution or not;

and if the historical order quantity obeys the pareto distribution, determining the historical order quantity interval of the industry to be identified under normal probability by adopting a pareto distribution method.

It should be noted that Pareto distributions (Pareto distributions) in the above method are power law distributions found from a large number of real-world phenomena. It is also called Power Law Distribution (Power Law Distribution), and the essence of the pareto Distribution is a positive feedback mechanism (positive feedback loop), which results when events are no longer independent and the generation of one event affects the generation of itself and other homogeneous events.

A pareto distribution is understood to be a thick-tailed distribution. For a random variable X, X _m Is the minimum value x can take. The probability distribution function of X is specifically as follows:

this results in a coefficient α which is positive, called shape parameter, or tail index. x is a radical of a fluorine atom _m At 1, the smaller alpha, the more pronounced the thick tail feature. When x → ∞ the pareto distribution has an order of

The tail of (2).

After the order quantity pareto distribution method is analyzed, corresponding quantiles can be determined, and corresponding order quantity intervals under normal probability are determined and calculated according to the quantile intervals.

And secondly, selecting a specific algorithm to determine the historical order quantity interval according to the specific quantity of the historical order quantity.

and selecting a specific algorithm according to the specific quantity of the historical order quantity for fitting analysis to obtain a historical order quantity interval of the industry to be identified under the normal probability corresponding to the specific algorithm of each algorithm.

It should be noted that, in the anomaly detection algorithm, some algorithms can still meet the detection accuracy under the condition that the data volume is small, for example: the pareto algorithm can perform anomaly detection for small sample data. Therefore, the corresponding abnormality detection algorithm can be selected according to the size of the data amount.

By the method, a specific abnormal detection algorithm can be selected according to the specific quantity of the order quantity, and the order quantity interval under the normal probability is determined, so that the accuracy of the abnormal detection of the order quantity is ensured.

Optionally, the selecting a specific algorithm according to the specific quantity of the historical order quantity to perform fitting analysis to obtain a historical order quantity interval of the industry to be identified under a normal probability corresponding to the specific algorithm specifically includes:

determining whether the specific number of the historical order quantities is smaller than a first preset threshold value;

and if the specific quantity of the historical order quantity is smaller than the first preset threshold value, determining the historical order quantity interval of the industry to be identified under normal probability by adopting a pareto distribution method.

It should be noted that the preset threshold in the above steps may be defined according to an actual application scenario.

And thirdly, determining a historical order quantity interval according to the distribution trend of the historical order quantity and the specific quantity of the historical order quantity.

and selecting a specific algorithm for fitting analysis according to the distribution trend of the historical order quantity and the specific quantity of the historical order quantity to obtain a historical order quantity interval of the industry to be identified under the normal probability corresponding to the specific algorithm.

It should be noted that, when a specific algorithm is selected, the selection can be performed according to the distribution trend of the order quantity and the specific quantity of the order quantity, and when the normal distribution is obeyed, the normal distribution method is adopted to determine the historical order quantity interval; if the normal distribution is not obeyed, other abnormal detection algorithms are continuously selected according to the specific quantity of the order quantity to determine the historical order quantity interval.

Optionally, the selecting a specific algorithm according to the distribution trend of the historical order quantity and the specific quantity of the historical order quantity to perform fitting analysis, so as to obtain a historical order quantity interval of the industry to be identified under a normal probability corresponding to the specific algorithm, and specifically may include:

performing normality check on the historical order quantity, and judging whether the historical order quantity obeys normal distribution or not to obtain a fourth judgment result;

when the fourth judgment result shows that the historical order quantity obeys normal distribution, determining a historical order quantity interval of the industry to be identified under normal probability according to a four-sigma principle of the normal distribution;

when the fourth judgment result shows that the historical order quantity does not comply with normal distribution, determining whether the specific quantity of the historical order quantity is smaller than a first preset threshold value;

By the method, the specific algorithm model is selected for fitting analysis according to the distribution trend of the order quantity and the specific quantity of the order quantity, and a more appropriate algorithm can be selected for carrying out abnormal detection on the order quantity so as to ensure the detection accuracy.

Optionally, after selecting a specific algorithm according to the historical order quantity for fitting analysis, the method may further include:

determining an upper quantile and a lower quantile corresponding to the historical order quantity;

and determining a historical order quantity interval corresponding to the industry to be identified according to the upper quantile point and the lower quantile point.

It should be noted that the method of determining the quantiles and determining the order amount according to the quantiles can be used in combination with various algorithms for anomaly detection.

The determining the historical order quantity interval corresponding to the industry to be identified according to the upper quantile and the lower quantile may specifically include:

acquiring a normal probability corresponding to the specific algorithm;

determining corresponding upper and lower quantiles in the normal probability;

determining a value of an order quantity corresponding to the upper branch position point as a first end point value of the order quantity interval;

determining a value of the order quantity corresponding to the lower branch position point as a second endpoint value of the order quantity interval;

determining the order quantity interval according to the first endpoint value and the second endpoint value.

Optionally, after it is determined that there is an abnormality in the order quantity to be analyzed in the industry to be identified within the set time period, the method may further include:

and determining abnormal orders in the order quantity to be analyzed.

Optionally, after determining the abnormal order in the industry to be identified, the method may further include:

obtaining an order record corresponding to the order to be analyzed; the order record at least comprises a commercial tenant corresponding to the order;

and determining the merchant corresponding to the merchant information contained in the order to be analyzed as an abnormal merchant according to the order record.

In practical application, when the platform detects that the order quantity of a certain industry is abnormal, the platform can further analyze the order of the industry, firstly determine the merchant with higher contribution degree to the abnormal order quantity, and then further check the merchant. If it is determined that there is a violation in the order size for a merchant, it may be further determined which specific orders for that merchant have an anomaly. If further checking determines that one or more merchants have abnormal orders, the merchants can be allowed to correct the orders or send warning information to the merchants in the abnormal orders.

In some embodiments, the platform may issue some reward or subsidy for the merchant, and the merchant with the order size satisfying the condition may receive the subsidy or reward. At this point, the platform may disqualify its domain rewards or subsidies for merchants with unusual order volumes.

Certainly, when the abnormal order quantity exists in a certain industry, the abnormal order quantity does not mean that the industry has an abnormal order, so that the industry with the abnormal order quantity can be further analyzed, and the order can be evaluated and supervised from the aspects of merchants, commodities corresponding to the order, transaction information, buyer information and the like.

Therefore, in the above step, after determining, according to the order record, the merchant corresponding to the merchant information included in the order to be analyzed as an abnormal merchant, the method may further include:

and the platform generates alarm prompt information, wherein the alarm prompt information is used for prompting the platform that the merchant has the risk of abnormal order quantity.

The platform can also send warning information to the abnormal merchant, and the warning information is used for prompting the abnormal merchant to process the abnormal order.

Through the method in the above embodiment, the embodiment of the present specification can achieve the following technical effects:

1) Obtaining the historical order quantity of industries to be identified in a set historical time period; selecting a specific algorithm according to the historical order quantity to perform fitting analysis, and obtaining a historical order quantity interval of the industry to be identified under normal probability; acquiring the order quantity to be analyzed in a set time period of the industry to be identified; if the order quantity to be analyzed is located outside the historical order quantity interval, it can be determined that the order quantity to be analyzed of the industry to be identified in the set time period is abnormal. By the method, the abnormal condition of the order quantity in the industry can be detected according to the order quantity of the industry, so that the trading orders in the platform can be effectively monitored.

2) In the cross-border trading platform, the order quantity is sourced from all industry merchants all over the world, and cross-border trading has corresponding supervisors, so that the obtained order quantity data are more detailed and more authoritative.

3) The abnormal order quantity of the industries is detected according to the order quantity, and the platform can be assisted to effectively supervise each industry on the platform, so that a more powerful basis is provided for the decision of an authority mechanism.

Based on the same idea, the embodiment of the present specification further provides a device corresponding to the method. Fig. 4 is a schematic structural diagram of an order quantity abnormality detection apparatus provided in an embodiment of the present specification. As shown in fig. 4, the apparatus may include:

a historical order quantity obtaining module 402, configured to obtain a historical order quantity of an industry to be identified in a set historical time period;

a historical order quantity interval determining module 404, configured to select a specific algorithm according to the historical order quantity to perform fitting analysis, so as to obtain a historical order quantity interval of the industry to be identified under a normal probability;

an order quantity to be analyzed obtaining module 406, configured to obtain an order quantity to be analyzed in a set time period for the industry to be identified;

a first determining module 408, configured to determine whether the order quantity to be analyzed is located in the historical order quantity interval, so as to obtain a first determination result;

the order quantity abnormity determining module 410 is configured to determine that the order quantity to be analyzed of the industry to be identified is abnormal in a set time period when the first determination result indicates that the order quantity to be analyzed is located outside the historical order quantity interval.

The examples of this specification also provide some specific embodiments of the process based on the apparatus of fig. 4, which is described below.

Optionally, the apparatus may further include:

the second judgment module is used for judging whether the historical order quantity is greater than or equal to a detection threshold value or not to obtain a second judgment result;

the order quantity abnormity detection module is used for carrying out abnormity detection on the historical order quantity when the second judgment result shows that the historical order quantity is larger than or equal to the detection threshold value;

and the stop detection module is used for not performing abnormal detection on the historical order quantity when the second judgment result shows that the historical order quantity is smaller than the detection threshold value.

Optionally, the historical order quantity interval determining module 404 may specifically include:

and the first fitting analysis unit is used for selecting a specific algorithm for fitting analysis according to the distribution trend of the historical order quantity to obtain the historical order quantity interval of the industry to be identified under the normal probability corresponding to the specific algorithm.

and the second fitting analysis unit is used for selecting a specific algorithm to perform fitting analysis according to the specific quantity of the historical order quantity to obtain a historical order quantity interval of the industry to be identified under the normal probability corresponding to the specific algorithm.

Optionally, the first fitting analysis unit may be specifically configured to:

Optionally, the second fitting analysis unit may be specifically configured to:

Optionally, the first fitting analysis unit may be further configured to:

Optionally, the first fitting analysis unit may be specifically configured to:

Optionally, the apparatus may further include:

the quantile determining module is used for determining an upper quantile and a lower quantile corresponding to the historical order quantity;

and the historical order quantity interval determining module is used for determining the historical order quantity interval corresponding to the industry to be identified according to the upper quantile point and the lower quantile point.

and the third fitting analysis unit is used for selecting a specific algorithm for fitting analysis according to the distribution trend of the historical order quantity and the specific quantity of the historical order quantity to obtain the historical order quantity interval of the industry to be identified under the normal probability corresponding to the specific algorithm.

Optionally, the third fitting analysis unit may specifically be configured to:

performing normality check on the historical order quantity, and judging whether the historical order quantity is subjected to normal distribution or not to obtain a fourth judgment result;

when the fourth judgment result shows that the historical order quantity obeys normal distribution, determining a historical order quantity interval of the industry to be identified under normal probability according to the four-sigma principle of the normal distribution;

when the fourth judgment result shows that the historical order quantity does not obey normal distribution, determining whether the specific quantity of the historical order quantity is smaller than a first preset threshold value or not;

Optionally, the apparatus may further include:

and the abnormal order determining module is used for determining abnormal orders in the order quantity to be analyzed.

Optionally, the apparatus may further include:

the order record acquisition module is used for acquiring the order record corresponding to the order to be analyzed; the order record at least comprises a commercial tenant corresponding to the order;

and the abnormal merchant determining module is used for determining the merchant corresponding to the merchant information contained in the order to be analyzed as the abnormal merchant according to the order record.

Based on the same idea, the embodiment of the present specification further provides a device corresponding to the above method.

Fig. 5 is a schematic structural diagram of an order quantity abnormality detection apparatus provided in an embodiment of the present specification. As shown in fig. 5, the apparatus 500 may include:

at least one processor 510; and the number of the first and second groups,

a memory 530 communicatively coupled to the at least one processor; wherein, the first and the second end of the pipe are connected with each other,

the memory 530 stores instructions 520 executable by the at least one processor 510 to cause the at least one processor 510 to:

acquiring the order quantity to be analyzed of the industry to be identified in a set time period;

Based on the same idea, the embodiment of the present specification further provides a computer-readable medium corresponding to the above method. The computer readable medium has computer readable instructions stored thereon which are executable by a processor to implement a method comprising:

In the 90's of the 20 th century, improvements to a technology could clearly distinguish between improvements in hardware (e.g., improvements to circuit structures such as diodes, transistors, switches, etc.) and improvements in software (improvements to process flow). However, as technology advances, many of today's process flow improvements have been seen as direct improvements in hardware circuit architecture. Designers almost always obtain the corresponding hardware circuit structure by programming an improved method flow into the hardware circuit. Thus, it cannot be said that an improvement in the process flow cannot be realized by hardware physical blocks. For example, a Programmable Logic Device (PLD), such as a Field Programmable Gate Array (FPGA), is an integrated circuit whose Logic functions are determined by programming the Device by a user. A digital symbol system is "integrated" onto a PLD by the designer's own programming without requiring the chip manufacturer to design and fabricate a dedicated integrated circuit chip. Furthermore, nowadays, instead of manually manufacturing an Integrated Circuit chip, such Programming is often implemented by "logic compiler" software, which is similar to a software compiler used in program development and writing, but the original code before compiling is also written by a specific Programming Language, which is called Hardware Description Language (HDL), and HDL is not only one but many, such as abll (advanced desktop Expression Language), AHDL (alternate Hardware Description Language), traffic, CUPL (computer unified Programming Language), HDCal, jhddl (Java Description Language), lava, lola, HDL, PALASM, rhyd (Hardware Description Language), and the like, which are currently used in the field-Hardware Language. It will also be apparent to those skilled in the art that hardware circuitry that implements the logical method flows can be readily obtained by merely slightly programming the method flows into an integrated circuit using the hardware description languages described above.

The controller may be implemented in any suitable manner, for example, the controller may take the form of, for example, a microprocessor or processor and a computer-readable medium storing computer-readable program code (e.g., software or firmware) executable by the (micro) processor, logic gates, switches, an Application Specific Integrated Circuit (ASIC), a programmable logic controller, and an embedded microcontroller, examples of which include, but are not limited to, the following microcontrollers: ARC 625D, atmel at91SAM, microchip PIC18F26K20, and Silicone Labs C8051F320, the memory controller may also be implemented as part of the control logic of the memory. Those skilled in the art will also appreciate that, in addition to implementing the controller as pure computer readable program code, the same functionality can be implemented by logically programming method steps such that the controller is in the form of logic gates, switches, application specific integrated circuits, programmable logic controllers, embedded microcontrollers and the like. Such a controller may thus be regarded as a hardware component and the means for performing the various functions included therein may also be regarded as structures within the hardware component. Or even means for performing the functions may be conceived to be both a software module implementing the method and a structure within a hardware component.

The systems, devices, modules or units illustrated in the above embodiments may be implemented by a computer chip or an entity, or by a product with certain functions. One typical implementation device is a computer. In particular, the computer may be, for example, a personal computer, a laptop computer, a cellular telephone, a camera phone, a smartphone, a personal digital assistant, a media player, a navigation device, an email device, a game console, a tablet computer, a wearable device, or a combination of any of these devices.

For convenience of description, the above devices are described as being divided into various units by function, respectively. Of course, the functionality of the various elements may be implemented in the same one or more pieces of software and/or hardware in the practice of the present application.

As will be appreciated by one skilled in the art, embodiments of the present invention may be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.

The present invention is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.

These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

In a typical configuration, a computing device includes one or more processors (CPUs), input/output interfaces, network interfaces, and memory.

The memory may include forms of volatile memory in a computer readable medium, 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 a computer-readable medium.

Computer-readable media, including both non-transitory and non-transitory, removable and non-removable media, may implement information storage by any method or technology. The information may be computer readable instructions, data structures, modules of a program, 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 technology, compact disc read only memory (CD-ROM), digital Versatile Disks (DVD) or other optical storage, magnetic cassettes, magnetic tape disk storage or other magnetic storage devices, or any other non-transmission medium which can be used to store information which can be accessed by a computing device. As defined herein, a computer readable medium does not include a transitory computer readable medium such as a modulated data signal and a carrier wave.

It should also be noted that the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrases "comprising a," "8230," "8230," or "comprising" does not exclude the presence of other like elements in a process, method, article, or apparatus comprising the element.

As will be appreciated by one skilled in the art, embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and so forth) having computer-usable program code embodied therein.

The application may be described in the general context of computer-executable instructions, such as program modules, being executed by a computer. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform particular tasks or implement particular abstract data types. The application may also be practiced in distributed computing environments where tasks are performed by remote processing devices that are linked through a communications network. In a distributed computing environment, program modules may be located in both local and remote computer storage media including memory storage devices.

The above description is only an example of the present application and is not intended to limit the present application. Various modifications and changes may occur to those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the scope of the claims of the present application.

Claims

1. An order quantity abnormity detection method comprises the following steps:

selecting a specific algorithm according to the historical order quantity to perform fitting analysis, and obtaining a historical order quantity interval of the industry to be identified under normal probability; the specific algorithm is selected according to the distribution trend of the historical order quantity and/or the specific quantity of the historical order quantity;

2. The method of claim 1, prior to selecting a particular algorithm for fitting analysis based on the historical order quantity, further comprising:

3. The method according to claim 1, wherein the selecting a specific algorithm according to the historical order quantity to perform fitting analysis to obtain a historical order quantity interval of the industry to be identified under normal probability specifically comprises:

and selecting a specific algorithm according to the distribution trend of the historical order quantity to perform fitting analysis, so as to obtain the historical order quantity interval of the industry to be identified under the normal probability corresponding to the specific algorithm.

4. The method according to claim 1, wherein the selecting a specific algorithm according to the historical order quantity to perform fitting analysis to obtain a historical order quantity interval of the industry to be identified under normal probability specifically comprises:

and selecting a specific algorithm according to the specific quantity of the historical order quantity for fitting analysis to obtain a historical order quantity interval of the industry to be identified under the normal probability corresponding to the specific algorithm.

5. The method according to claim 3, wherein a specific algorithm is selected according to the distribution trend of the historical order quantity to perform fitting analysis, so as to obtain a historical order quantity interval of the industry to be identified under a normal probability corresponding to the specific algorithm, and specifically includes:

6. The method according to claim 4, wherein the selecting a specific algorithm according to the specific quantity of the historical order quantity to perform fitting analysis to obtain a historical order quantity interval of the industry to be identified under a normal probability corresponding to the specific algorithm specifically comprises:

7. The method according to claim 5, wherein the performing a normality check on the historical order quantity to determine whether the historical order quantity complies with a normal distribution further comprises, after obtaining a third determination result:

when the third judgment result shows that the historical order quantity does not comply with normal distribution, verifying whether the historical order quantity complies with pareto distribution;

and if the historical order quantity obeys pareto distribution, determining the historical order quantity interval of the industry to be identified under normal probability by adopting a pareto distribution method.

8. The method of claim 5, wherein determining the historical order quantity interval of the industry to be identified with a normal probability according to the four-sigma principle of normal distribution comprises:

9. The method of claim 1, after selecting a particular algorithm for fitting analysis based on the historical order quantity, further comprising:

10. The method according to claim 1, wherein the selecting a specific algorithm according to the historical order quantity to perform fitting analysis to obtain a historical order quantity interval of the industry to be identified under a normal probability specifically comprises:

11. The method according to claim 10, wherein the selecting a specific algorithm according to the distribution trend of the historical order quantity and the specific quantity of the historical order quantity for fitting analysis to obtain the historical order quantity interval of the industry to be identified under the normal probability corresponding to the specific algorithm specifically comprises:

12. The method of claim 1, after determining that there is an anomaly in the amount of orders to be analyzed for a set period of time for the industry to be identified, further comprising:

and determining abnormal orders in the order quantity to be analyzed.

13. The method of claim 12, after determining the anomalous order in the industry to be identified, further comprising:

14. An order quantity abnormality detection apparatus comprising:

the historical order quantity interval determining module is used for selecting a specific algorithm according to the historical order quantity to perform fitting analysis so as to obtain a historical order quantity interval of the industry to be identified under normal probability; the specific algorithm is selected according to the distribution trend of the historical order quantity and/or the specific quantity of the historical order quantity;

15. The apparatus of claim 14, the apparatus further comprising:

16. The apparatus of claim 14, wherein the historical order quantity interval determining module specifically comprises:

17. The apparatus of claim 14, wherein the historical order quantity interval determining module specifically comprises:

and the second fitting analysis unit is used for selecting a specific algorithm for fitting analysis according to the specific quantity of the historical order quantity to obtain a historical order quantity interval of the industry to be identified under the normal probability corresponding to the specific algorithm.

18. The apparatus according to claim 16, wherein the first fitting analysis unit is specifically configured to:

19. The apparatus according to claim 17, wherein the second fitting analysis unit is specifically configured to:

20. The apparatus of claim 18, the first fitting analysis unit, further configured to:

21. The apparatus according to claim 18, wherein the first fitting analysis unit is specifically configured to:

22. The apparatus of claim 14, the apparatus further comprising:

and the historical order quantity interval determining module is used for determining the historical order quantity interval corresponding to the industry to be identified according to the upper quantile and the lower quantile.

23. The apparatus according to claim 14, wherein the historical order quantity interval determining module specifically includes:

24. The apparatus of claim 23, wherein the third fitting analysis unit is specifically configured to:

25. The apparatus of claim 14, the apparatus further comprising:

26. The apparatus of claim 25, further comprising:

27. An order quantity abnormality detection apparatus comprising:

at least one processor; and the number of the first and second groups,

the memory stores instructions executable by the at least one processor to enable the at least one processor to:

28. A computer readable medium having computer readable instructions stored thereon which are executable by a processor to implement the order quantity anomaly detection method of any one of claims 1 to 13.