AB experiment shunting method and device, experiment method, equipment and medium

Abstract

The invention provides an AB experiment shunting method and device, an experiment method, equipment and a medium. The shunting method comprises the following steps: acquiring experiment flow; obtaining a time slice code and a space slice code of each user in the experimental traffic according to a preset rule; and according to the time slice codes and the space slice codes of the users, the users are shunted to each experiment group preset by the AB experiment to obtain a shunting result, and the users with the same time slice codes and the same space slice codes are shunted to the same experiment group. According to the method, people in the same time and space receive the same strategy, so that algorithm price discrimination is avoided fundamentally; moreover, under the same time, the flows in different spaces enter different experiment groups, so that the problem of experiment interference caused by short-time supply and demand imbalance caused by shunting of a pure time slice is solved; meanwhile, the sizes of the time slice and the space slice can be finely set, fine operation is carried out, and the application scene of the experiment is further improved.

Description

Technical Field

[0001] The present invention relates to the field of computer technology, and in particular to an AB experiment diversion method and device, an experimental method, equipment, and a medium. Background Art

[0002] With the rapid development of information technologies such as the Internet and cloud computing, a variety of applications have emerged in the application market to meet the diverse needs of users. During the operation of applications, AB experiments are often used to evaluate and improve the performance of applications. When conducting AB experiments, a large number of users need to be diverted to multiple groups in the AB experiment (for example, blank group, control group, and experimental group) according to preset strategies, and then the users are asked to execute the corresponding groups in the application, and the usage data of users in each group is collected to evaluate and improve the application.

[0003] When conducting AB experiments, it is often necessary to set up a diversion method to divert the experimental traffic to different groups. Among these diversion methods, simple random diversion is the most common, and the random object is a single variable. When performing random diversion, there are simple time slice random diversion, simple user random diversion, or other forms of random diversion.

[0004] Take the random diversion of time slices as an example. Each time slice is a random object. The traffic in the same time slice either enters the experimental group or the control group. Assume that every half hour is a time slice, that is, timelot=30min; each time slice timestamp=tsn, there are 24 hours a day, a total of 48 time slices, so n=48. When conducting experiments under this diversion method, assuming that the strategy of the experimental group is better than that of the control group, the driver has a high order response rate after entering the experimental group; the driver has a low order response rate after entering the control group. This does not mean that the strategy of the experimental group is better. For example: the order response rate of the experimental group from 9:00 to 9:30 is higher, the capacity occupancy rate is higher, and most of the capacity is taking orders from 9:00 to 9:30; 9:30-10:00 is the control group. Since the capacity is still occupied by the orders from 9:00 to 9:30, it will lead to insufficient capacity from 9:30 to 10:00, and the order response rate is significantly lower than the previous time slice. At this time, short-term imbalance between supply and demand (supply << demand) will cause experimental interference, resulting in inaccurate and unreliable experimental results using the time slice random diversion method.

[0005] Taking the example of simple random user diversion, each user is a random object, with different user IDs, and inconsistent random results. Suppose user A and user B are friends, and they place an order at the same place and time. When user A and user B compare, they find that the prices are different, resulting in a poor user experience. At the same time, because marketing strategies vary from person to person, it may be mistaken for algorithmic price discrimination. Summary of the invention

[0006] In view of the above defects in the prior art, the present invention provides an AB experiment diversion method and device, experimental method, equipment, and medium to solve the technical problems of uneven experimental traffic distribution and the existence of being mistaken for algorithm price discrimination.

[0007] To achieve the above-mentioned purpose and other related purposes, the present invention provides an AB experiment diversion method, an AB experiment diversion method, comprising: obtaining experimental traffic; obtaining the time slice code and space slice code of each user in the experimental traffic according to preset rules; diverting the user to each experimental group preset in the AB experiment according to the time slice code and the space slice code of the user, and obtaining a diversion result, wherein users with the same time slice code and the same space slice code are diverted to the same experimental group.

[0008] In one embodiment of the present invention, according to preset rules, the time slice code of each user in the experimental traffic is obtained, including: according to the preset start time and time slice length, the time period corresponding to each time slice is obtained; according to the time period corresponding to each time slice, the code of each time slice is obtained, wherein the codes of all time slices are different; according to the code of the time slice and the time characteristics of the user, the time slice code of the user is obtained.

[0009] In one embodiment of the present invention, the spatial slice coding of the experimental traffic is obtained according to preset rules, including: obtaining the coding of each of the spatial slices according to a preset spatial area and its corresponding coding, wherein the coding of all spatial slices is different; obtaining the spatial slice coding of the user according to the coding of the spatial slice and the location characteristics of the user.

[0010] In one embodiment of the present invention, the code of each of the spatial slices is obtained according to a preset spatial area and its corresponding code, including: obtaining the code of each of the spatial slices according to a preset administrative area and its corresponding code, or according to a preset geographic fence and its corresponding code; obtaining the spatial slice code of the user according to the code of the spatial slice and the location characteristics of the user, including: obtaining the spatial slice code of the user according to the code of the spatial slice and the starting location characteristics or the end location characteristics of the user.

[0011] In one embodiment of the present invention, users are diverted to the experimental groups preset in the AB experiment according to the time slice coding and the space slice coding of the users to obtain the diversion results, including: obtaining the time-space slice coding of the users according to the time slice coding and the space slice coding of the users; and diverting the users to the experimental groups preset in the AB experiment according to the time-space slice coding of the users to obtain the diversion results.

[0012] In one embodiment of the present invention, the user is diverted to each experimental group preset in the AB experiment according to the space-time slice code of the user to obtain the diversion result, including: obtaining a hash value of the space-time slice code according to the space-time slice code of the user; and diverting the user to the corresponding experimental group according to the remainder between the hash value of the user and the number of experimental groups of the AB experiment to obtain the diversion result.

[0013] To achieve the above-mentioned purpose and other related purposes, the present invention also provides an AB experiment diversion device, including: a data acquisition unit, used to obtain experimental traffic; a coding calculation unit, used to obtain the time slice code and space slice code of each user in the experimental traffic according to preset rules; a diversion unit, used to divert the user to each experimental group preset in the AB experiment according to the user's time slice code and the space slice code, and obtain a diversion result, wherein users with the same time slice code and the same space slice code are diverted to the same experimental group.

[0014] To achieve the above-mentioned purpose and other related purposes, the present invention also provides an AB experimental method, including: obtaining experimental requirements; formulating an experimental plan based on the experimental requirements, in which the experimental traffic is distributed to each experimental group according to the above-mentioned diversion method; starting the experimental plan and collecting experimental data generated during the experiment; and obtaining targeted indicator analysis results based on the experimental data.

[0015] To achieve the above-mentioned purpose and other related purposes, the present invention also provides an electronic device, including a processor, a memory and a communication bus; the communication bus is used to connect the processor and the memory; the processor is used to execute a computer program stored in the memory to implement a method provided in any one of the above embodiments.

[0016] To achieve the above-mentioned object and other related objects, the present invention further provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program is used to enable a computer to execute the method provided in any one of the above-mentioned embodiments.

[0017] Beneficial effects of the present invention: The present invention proposes an AB experiment diversion method and device, experimental method, equipment, and medium, which diverts users with the same time slice code and space slice code to the same experimental group, so that people in the same time and space receive the same strategy, fundamentally avoiding the occurrence of algorithm price discrimination; and, at the same time, traffic in different spaces enters different experimental groups, alleviating the problem of experimental interference caused by short-term supply and demand imbalance brought about by the diversion of simple time slices; at the same time, the size of the time slice and space slice can be finely set to perform fine operations, further improving the applicable scenarios of the experiment. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0019] Figure 1 A flow chart of a flow diversion method provided by an embodiment of the present invention;

[0020] Figure 2 A flow chart of obtaining time slice coding provided by an embodiment of the present invention;

[0021] Figure 3 A flow chart of obtaining spatial slice coding provided by an embodiment of the present invention;

[0022] Figure 4 A flow chart based on spatiotemporal slice coding and splitting provided in one embodiment of the present invention;

[0023] Figure 5 A flowchart of shunting based on encoded hash values ​​provided by an embodiment of the present invention;

[0024] Figure 6 A schematic diagram of a diversion result provided by an embodiment of the present invention;

[0025] Figure 7 A schematic diagram of the structure of a flow diversion device provided in one embodiment of the present invention;

[0026] Figure 8 A flow chart of an AB experimental method provided by one embodiment of the present invention;

[0027] Fig. 9 A schematic diagram of the structure of an electronic device provided by an embodiment of the present invention.

[0028] Explanation of the reference numerals: 101, data acquisition unit; 102, encoding calculation unit; 103, diversion unit; 201, processor; 202, memory. DETAILED DESCRIPTION

[0029] The following describes the embodiments of the present invention through specific specific embodiments, and those skilled in the art can easily understand other advantages and effects of the present invention from the contents disclosed in this specification. It should be noted that the following embodiments and the features in the embodiments can be combined with each other without conflict. In addition to the specific methods, equipment, and materials used in the embodiments, according to the technical personnel in this technical field's grasp of the prior art and the records of the present invention, any methods, equipment, and materials of the prior art that are similar or equivalent to the methods, equipment, and materials described in the embodiments of the present invention can also be used to implement the present invention.

[0030] It should be understood that the terms used in the embodiments of the present invention are for describing specific embodiments rather than for limiting the scope of protection of the present invention. Unless otherwise defined, all technical and scientific terms used in the present invention have the same meaning as those generally understood by those skilled in the art.

[0031] In the following description, numerous details are discussed to provide a more thorough explanation of the embodiments of the present invention. However, it is obvious to those skilled in the art that the embodiments of the present invention can be implemented without these specific details. In some of the embodiments, well-known structures and devices are shown in the form of block diagrams rather than in detail to avoid making the embodiments of the present invention difficult to understand.

[0032] The flowchart and block diagram in the accompanying drawings illustrate the system architecture, functions and operations that may be implemented according to the methods and computer program products of various embodiments disclosed in the present invention. In this regard, each box in the flowchart or block diagram can represent a module, a program segment, or a part of a code, and the module, the program segment, or a part of the code contains one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or the flowchart, and the combination of the boxes in the block diagram and / or the flowchart can be implemented with a dedicated hardware-based system that performs the specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0033] See also Figure 1 , Figure 1 An AB experiment diversion method provided by an embodiment of the present invention includes steps S100 to S300.

[0034] Step S100, obtain experimental traffic. Experimental traffic refers to the total number of visits of the user group selected to participate in the AB experiment. It is usually used to describe the traffic generated by users who actually enter the experimental process and are assigned to different experimental groups (such as Group A, Group B, etc.). It is an important indicator for measuring the scale of the experiment and the ability to collect experimental data. Sufficient experimental traffic is crucial to ensure the statistical significance and reliability of the experimental results. When designing an experiment, it is necessary to determine the required experimental traffic based on the purpose of the experiment and the size of the expected effect to ensure that there is enough data to detect the real differences between the experimental variables.

[0035] Step S200: According to a preset rule, the time slice code and space slice code of each user in the experimental traffic are obtained.

[0036] See also Figure 2 In a specific embodiment of the present invention, according to a preset rule, the time slice code of each user in the experimental traffic is obtained, including steps S211 to S213.

[0037] Step S211: Obtain the time period corresponding to each time slice according to the preset start time and the time slice duration. The preset start time may be, for example, 0 o'clock, and the preset time slice duration may be, for example, 30 minutes. Then, according to these two parameters, the time period corresponding to each time slice may be calculated. There are 48 time periods in total, corresponding to 0:00-0:30, 0:30-1:00, ..., and so on.

[0038] When presetting the time slice duration, the time slice duration can be made smaller than the peak period duration, so that the peak period can be divided into different time slices. For example, according to historical data, it can be known that the peak period duration is generally 1 hour, so the time slice duration can be set to 30 minutes.

[0039] Step S212: Obtain the code (ID) of each time slice according to the time period corresponding to each time slice, wherein the codes of all time slices are different. This process can be implemented by programming or manually, and the specific method depends on the division rules and requirements of the time slices. For example, the following methods can be used: (1) linear numbering, assigning a linearly increasing code to each time period; (2) using a combination of date and time to generate a hash value, and then assigning the code based on the hash value; and so on. It is also possible to define more complex rules to assign time slice codes based on specific business needs, such as considering working days and non-working days, holidays, etc. In actual applications, which method to choose depends on the specific needs and business logic of the time slice. These methods can be flexibly adjusted and expanded to adapt to different scenarios.

[0040] Step S213: Obtain the user's time slice code according to the time slice code and the user's time feature. Taking the freight scenario as an example, in this step, the user's time feature can be, for example, the user's order time. Assuming that the user places an order at 7:35, the corresponding time period can be found to be 7:30-8:00, then the time slice code corresponding to the time period 7:30-8:00 is the user's time slice code.

[0041] See also Figure 3 In a specific embodiment of the present invention, according to a preset rule, the spatial slice coding of the experimental traffic is obtained, including: S221, according to the preset spatial area and its corresponding coding, the coding of each spatial slice is obtained, wherein the coding of all spatial slices is different; S222, according to the coding of the spatial slice and the location characteristics of the user, the spatial slice coding of the user is obtained. The process of obtaining the spatial slice coding of the user is basically similar to the process of time slice coding. First, the coding of each spatial slice needs to be preset, and then the spatial slice of the user is found according to the characteristics of the user to determine the spatial slice coding of the user.

[0042] In a specific embodiment of the present invention, step S221 includes: obtaining the code of each spatial slice according to the preset administrative area and its corresponding code, or according to the preset geographic fence and its corresponding code. That is, the spatial area in step S221 can be an administrative area or a geographic fence. The administrative area and its corresponding code can be understood as the administrative division code planned for different regions according to national standards. Each administrative area has a specific range and code, and the code of each spatial slice can be directly obtained based on this. Or according to the needs of each platform itself, the spatial area is divided by geographic fence technology based on the virtual boundary set based on the geographical location. In addition, other methods can be used to divide and encode the spatial slices, as long as the spatial slice codes corresponding to users in different spaces are different.

[0043] In a specific embodiment of the present invention, step S222 includes: obtaining the user's spatial slice code according to the spatial slice code and the user's starting position feature or the end position feature. That is, in step S222, when obtaining the user's spatial slice code, it can be obtained according to the user's starting position feature or the user's end position feature, so that different users will not have different prices when the starting point and the end point are the same.

[0044] Step S300: Divide the user into experimental groups preset in the AB experiment according to the user's time slice code and space slice code to obtain a diversion result, wherein users with the same time slice code and the same space slice code are diverted into the same experimental group.

[0045] See also Figure 4In a specific embodiment of the present invention, users are diverted to experimental groups preset in an AB experiment according to the time slice coding and space slice coding of the users, and diversion results are obtained, including step S310 and step S320.

[0046] Step S310: Obtain the user's space-time slice code according to the user's time slice code and space slice code. For example, the user's space-time slice code may be directly added to the user's time slice code to obtain the user's space-time slice code.

[0047] Step S320: Divide the user into various experimental groups preset in the AB experiment according to the user's space-time slice coding to obtain a diversion result.

[0048] See also Figure 5 In a specific embodiment of the present invention, step S320 includes: S321, according to the space-time slice code of the user, obtain the hash value of the space-time slice code; S322, according to the remainder between the hash value of the user and the number of experimental groups in the AB experiment, divert the user to the corresponding experimental group to obtain the diversion result. Assuming that there are two groups, namely the experimental group and the control group, then first calculate the hash value of the space-time slice code, and then use the hash value to calculate the two remainders. If the remainder is 0, the user is diverted to the experimental group, and if the remainder is 1, the user is diverted to the control group. In this way, users can be diverted based on the space-time slice code of the user.

[0049] Figure 6 This is the result of the diversion in a specific embodiment of the present invention. As shown in the figure, the horizontal axis represents time, the vertical axis represents space, and each combination of a time slice and a space slice represents a space-time slice. The object of our random diversion is the space-time slice. The traffic in the same space-time slice is either all experimental group or all control group, which corresponds to the step S300: users with the same time slice code and the same space slice code are diverted to the same experimental group. The specific size of the space-time slice is determined by the actual business scenario.

[0050] See also Figure 7 , Figure 7 An AB experiment diversion device provided for one embodiment of the present invention includes a data acquisition unit 101, a coding calculation unit 102 and a diversion unit 103; wherein the data acquisition unit 101 is used to obtain experimental traffic; the coding calculation unit 102 is used to obtain the time slice code and the space slice code of each user in the experimental traffic according to a preset rule; the diversion unit 103 is used to divert the user to each experimental group preset in the AB experiment according to the user's time slice code and space slice code, and obtain a diversion result, wherein users with the same time slice code and the same space slice code are diverted to the same experimental group.

[0051] It should be noted that the flow diversion device of this embodiment is a device corresponding to the above-mentioned flow diversion method, and the functional modules in the flow diversion device may correspond to the corresponding steps in the flow diversion method. The flow diversion device of this embodiment can be implemented in conjunction with the flow diversion method, that is, in the absence of conflict, the relevant technical details mentioned in the flow diversion method of the above-mentioned embodiment can also be applied to the flow diversion device of this embodiment.

[0052] See also Figure 8 , Figure 8 An AB experimental method provided by an embodiment of the present invention includes steps S1 to S4.

[0053] Step S1, obtain experimental requirements. Experimental requirements refer to a set of requirements that are clearly defined in order to achieve specific goals before conducting an AB experiment or any other type of experiment. These requirements define the purpose, scope, and expected results of the experiment. Clear and detailed experimental requirements are the basis for successful experiments. They help team members understand the purpose of the experiment, ensure that all relevant parties agree on the expectations and goals of the experiment, and provide guidance for the planning and execution of the experiment.

[0054] Step S2: formulate an experimental plan according to the experimental requirements, in which the experimental traffic is allocated to each experimental group according to the above-mentioned diversion method. Different from the traditional experimental plan, the experimental plan in this step adopts the above-mentioned diversion method to achieve traffic diversion, that is, the time-space slice diversion method is adopted.

[0055] Step S3, start the experiment plan and collect the experimental data generated during the experiment. When all the information is configured, the experiment can be started. After the experiment starts, experimental data will be continuously generated. These experimental data will be collected for subsequent analysis.

[0056] Step S4: Obtain targeted indicator analysis results based on the experimental data.

[0057] It should be noted that, whether it is the diversion method or the experimental method, the step division of the above methods is only for clear description. When implemented, they can be merged into one step or some steps can be split and decomposed into multiple steps. As long as they contain the same logical relationship, they are all within the protection scope of this patent; adding insignificant modifications to the algorithm or process or introducing insignificant designs without changing the core design of the algorithm and process are all within the protection scope of this patent.

[0058] See also Fig. 9 , Fig. 9An electronic device provided for one embodiment of the present invention includes a processor 201, a memory 202 and a communication bus; the communication bus is used to connect the processor 201 and the memory 202; the processor 201 is used to execute a computer program stored in the memory 202 to implement the above-mentioned diversion method or experimental method.

[0059] The above-mentioned electronic device is a device that can automatically perform numerical calculations and / or information processing according to pre-set or stored instructions, and its hardware includes but is not limited to microprocessors, application specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), digital signal processors (DSPs), embedded devices, etc.

[0060] The electronic device mentioned above can be any electronic product that can perform human-computer interaction with a user, such as a personal computer, a tablet computer, a smart phone, a personal digital assistant (PDA), a game console, an interactive network television (IPTV), a smart wearable device, etc.

[0061] The electronic device may also include a network device and / or a user device, wherein the network device includes, but is not limited to, a single network server, a server group consisting of multiple network servers, or a cloud consisting of a large number of hosts or network servers based on cloud computing.

[0062] The network where the above electronic devices are located includes but is not limited to the Internet, wide area network, metropolitan area network, local area network, virtual private network (VPN), etc.

[0063] The above-mentioned processor can be, for example, a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components; the above-mentioned memory may include random access memory (RAM), and may also include non-volatile memory (Non-volatile Memory), such as at least one disk storage.

[0064] An embodiment of the present invention further provides a computer-readable storage medium on which a computer program is stored. The computer program is used to enable a computer to execute the above-mentioned diversion method or experimental method.

[0065] In general, the present invention combines hashkey (time slice + space slice) to achieve multi-variable (time + space, other variables can also be introduced) control of hash results, especially in special scenarios such as surcharges, peak capacity resources are in short supply, and enterprises adjust capacity resources through surcharge strategies to maximize economic benefits. Using the hashkey method, the random results are related to the space-time slice composed of (time slice + space-time slice), so that the random results of the traffic in the same space-time slice are the same, either entering the experimental group or entering the control group. It has the following advantages: (1) In the same space-time, the traffic enters the same group, the strategy received by the user is the same, and there is no algorithm price discrimination; (2) At the same time, in different spaces (regions), the traffic may enter the experimental group or the control group, and the random allocation alleviates the problem of short-sighted supply and demand imbalance caused by simple time slice diversion causing experimental interference; (3) In the same space (region), in different time periods, the traffic may enter the experimental group or the control group, avoiding the deviation caused by the experiment entering only one group in the space and the experimental results are not random enough.

[0066] The present invention can be applied to passenger peak surcharge scenarios to verify the effectiveness and feasibility of surcharge strategies. Different surcharge strategies are required during peak periods and off-peak periods to achieve optimal cost and highest benefits.

[0067] The invention is also applicable to scenarios in which there are resource constraints and refined operations in bilateral markets, and different strategies need to be configured during business peaks and valleys. For example, for a food delivery company, how much is the appropriate errand fee to set at different times and spaces? For example, in a living area, business peaks at meal times, and encounters extreme weather (such as rain / strong wind), there will be a short-term imbalance between supply and demand (supply is far less than demand). At this time, the appropriate business strategy can also be found by using the random method of time and space slices.

[0068] The above embodiments are merely illustrative of the principles and effects of the present invention, and are not intended to limit the present invention. Anyone familiar with the art may modify or alter the above embodiments without departing from the spirit and scope of the present invention. Therefore, all equivalent modifications or alterations made by a person of ordinary skill in the art without departing from the spirit and technical concept disclosed by the present invention shall still be covered by the claims of the present invention.

Claims

1. An AB experiment diversion method, characterized in that: include: Get experimental traffic; According to a preset rule, a time slice code and a space slice code of each user in the experimental traffic are obtained; According to the time slice code and the space slice code of the user, the user is shunted into each experimental group preset in the AB experiment to obtain a shunting result, wherein users with the same time slice code and the same space slice code are shunted into the same experimental group.

2. The AB experiment diversion method according to claim 1, characterized in that: According to the preset rules, the time slice code of each user in the experimental traffic is obtained, including: According to the preset start time and time slice duration, the time period corresponding to each time slice is obtained; Obtaining a code for each time slice according to a time period corresponding to each time slice, wherein the codes for all time slices are different; The time slice code of the user is obtained according to the code of the time slice and the time feature of the user.

3. The AB experiment diversion method according to claim 1, characterized in that: According to the preset rules, the spatial slice coding of the experimental traffic is obtained, including: According to the preset spatial region and its corresponding code, a code of each of the spatial slices is obtained, wherein the codes of all the spatial slices are different; The spatial slice code of the user is obtained according to the code of the spatial slice and the position feature of the user.

4. The AB experiment diversion method according to claim 3, characterized in that: According to the preset spatial region and its corresponding code, the code of each spatial slice is obtained, including: Obtaining a code for each of the spatial slices according to a preset administrative region and its corresponding code, or according to a preset geographic fence and its corresponding code; Obtaining the spatial slice code of the user according to the code of the spatial slice and the location feature of the user includes: The spatial slice code of the user is obtained according to the code of the spatial slice and the starting position feature or the ending position feature of the user.

5. The AB experiment diversion method according to claim 1, characterized in that: Dividing the user into various experimental groups preset in the AB experiment according to the time slice code and the space slice code of the user, and obtaining a diversion result, including: Obtaining a time-space slice code of the user according to the time slice code and the space slice code of the user; According to the space-time slice coding of the user, the user is divided into each experimental group preset in the AB experiment to obtain the diversion result.

6. The AB experiment diversion method according to claim 5, characterized in that: According to the space-time slice code of the user, the user is divided into each experimental group preset in the AB experiment to obtain the diversion result, including: Obtaining a hash value of the space-time slice code according to the space-time slice code of the user; According to the remainder between the hash value of the user and the number of experimental groups of the AB experiment, the user is diverted to the corresponding experimental group to obtain the diversion result.

7. An AB experiment diversion device, characterized in that: include: A data acquisition unit, used to acquire experimental flow; A coding calculation unit, used to obtain a time slice code and a space slice code of each user in the experimental traffic according to a preset rule; The shunting unit is used to shunt the user to each experimental group preset in the AB experiment according to the time slice code and the space slice code of the user, and obtain a shunting result, wherein users with the same time slice code and the same space slice code are shunted to the same experimental group.

8. An AB experimental method, characterized in that: include: Obtain experimental requirements; According to the experimental requirements, an experimental plan is formulated, in which the experimental flow is distributed to each experimental group according to the flow diversion method described in any one of claims 1 to 6; Start the experimental plan and collect experimental data generated during the experiment; According to the experimental data, targeted indicator analysis results are obtained.

9. An electronic device, characterized in that: It comprises a processor, a memory and a communication bus; the communication bus is used to connect the processor and the memory; the processor is used to execute the computer program stored in the memory to implement the shunting method as described in any one of claims 1 to 6 or the experimental method as described in claim 8.

10. A computer-readable storage medium, characterized in that: A computer program is stored thereon, and the computer program is used to enable a computer to execute the diversion method according to any one of claims 1 to 6 or the experimental method according to claim 8.

Images