CN109583670B - Control experiment method, system and computer device based on time slice rotation - Google Patents
Control experiment method, system and computer device based on time slice rotation Download PDFInfo
- Publication number
- CN109583670B CN109583670B CN201710899136.4A CN201710899136A CN109583670B CN 109583670 B CN109583670 B CN 109583670B CN 201710899136 A CN201710899136 A CN 201710899136A CN 109583670 B CN109583670 B CN 109583670B
- Authority
- CN
- China
- Prior art keywords
- experiment
- time slice
- group
- time
- experimental
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0631—Resource planning, allocation, distributing or scheduling for enterprises or organisations
- G06Q10/06311—Scheduling, planning or task assignment for a person or group
Landscapes
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Engineering & Computer Science (AREA)
- Strategic Management (AREA)
- Economics (AREA)
- Entrepreneurship & Innovation (AREA)
- Development Economics (AREA)
- Educational Administration (AREA)
- Operations Research (AREA)
- Marketing (AREA)
- Game Theory and Decision Science (AREA)
- Quality & Reliability (AREA)
- Tourism & Hospitality (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Complex Calculations (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The embodiment of the disclosure provides a comparison experiment method, a system, a computer device and a computer readable storage medium based on time slice rotation, which are applied to the technical field of experiment testing, wherein the comparison experiment method based on time slice rotation comprises the following steps: determining experimental test information; selecting a time slice and determining a time slice rotation rule according to the experimental test information; performing an experiment according to the time slices and the time slice rotation rule to obtain a first group of experiment data and a second group of experiment data; and after the experiment is finished, counting experiment index results according to the first group of experiment data and the second group of experiment data to generate an experiment conclusion. The accuracy of experimental data statistics is improved by grouping, contrasting and counting the experimental data, and the problem of inter-group interference caused by a traditional experimental test method (random shunt) is avoided.
Description
Technical Field
The embodiment of the disclosure relates to the technical field of experimental tests, in particular to a comparison experiment method, a comparison experiment system, a computer device and a computer readable storage medium based on time slice rotation.
Background
AB Test (control experiment) the effectiveness of an experiment is often assessed by comparing the performance of different groups of users by randomly dividing the users into two groups, one of the groups being a control group and one of the groups being an experimental group. In a trip scene, passengers issue orders, after the drivers are assigned orders or actively grab the orders, the passengers share the drivers together, if the passengers are divided into an experimental group and a comparison group according to a passenger random shunting mode, the results obtained by the experiment can be influenced by the interference between the passengers in the experimental group and the passengers in the comparison group, and the method is unreliable.
Disclosure of Invention
The disclosed embodiments are directed to solving at least one of the technical problems of the related art or the related art.
To this end, an aspect of the embodiments of the present disclosure is to provide a control experiment method based on time slice rotation.
Another aspect of the embodiments of the present disclosure is to provide a control experiment system based on time slice rotation.
It is yet another aspect of an embodiment of the present disclosure to provide a computer apparatus.
It is yet another aspect of an embodiment of the present disclosure to provide a computer-readable storage medium.
In view of this, according to an aspect of the embodiments of the present disclosure, a control experiment method based on a rotation of time slices is provided, including: determining experimental test information; selecting a time slice and determining a time slice rotation rule according to the experimental test information; performing an experiment according to the time slices and the time slice rotation rule to obtain a first group of experiment data and a second group of experiment data; and after the experiment is finished, counting experiment index results according to the first group of experiment data and the second group of experiment data to generate an experiment conclusion.
According to the comparison experiment method based on the time slice rotation, after experiment test information needing to be carried out is determined, reasonable time slices are selected, the time slice rotation rule is determined, the acquired two groups of experiment data are guaranteed to have small absolute change and insignificant significance, the time slices are time intervals, one experiment is divided into a plurality of time slices, and the time slice rotation rule is that the experiment is carried out according to which experiment period and which rotation time period of the time slices. And then, carrying out an experiment to obtain the experimental data of the first group and the experimental data of the second group, further counting the experimental index results to obtain an experimental conclusion, improving the accuracy of experimental data counting by grouping comparison statistics of the experimental data, and avoiding the problem of inter-group interference caused by the traditional experimental test method (random shunt).
The above control experiment method based on time slice rotation according to the embodiment of the present disclosure may further have the following technical features:
in the above technical solution, preferably, the step of selecting a time slice and determining a time slice rotation rule according to the experimental test information specifically includes: selecting a time slice range according to the experimental test information; acquiring a first group of historical experimental data and a second group of historical experimental data, and calculating absolute change values and significance values of the first group of historical experimental data and the second group of historical experimental data; and selecting a time slice in the time slice range according to the absolute change value and the significance value, and determining a time slice rotation rule.
In this technical solution, an approximate range of the time slice is selected according to the test content included in the experimental test information, for example, if the time slice interference of the experimental test content is likely to be large, the time slice is as large as possible, and if the main influence of the test content is short-term effect, the time slice which is relatively small may be selected. After the approximate range of the time slice is determined, the absolute change and the significance of the two groups of experimental data are respectively calculated according to historical experimental data, a time slice as large as possible is selected according to the absolute change and the significance, a time slice rotation rule is determined, the reasonable time slice and the time slice rotation rule are selected, the fact that the absolute change of the two groups of acquired experimental data is small and the significance is not significant can be guaranteed, and experimental misjudgment caused by inter-group interference of the experimental data is avoided.
In any of the above technical solutions, preferably, the time slice rotation rule includes a preset experiment period and a preset rotation time period; performing an experiment according to the time slice and the time slice rotation rule to obtain a first group of experimental data and a second group of experimental data, specifically comprising: dividing the test time into time intervals T according to the time slices and the time slice rotation ruleij(ii) a If (i, j) ∈ S1Will be at time interval TijThe acquired experimental data are used as a first group of experimental data; if (i, j) ∈ S2Will be at time interval TijThe acquired experimental data are used as a second group of experimental data; wherein the time interval TijThe time length of (1, 2) · n, j ═ 1, 2,... said., n, m, n is a preset cycle period, m is a preset experimental period, and S is a time slice1＝{(i，j):(i mod 2＝0∩j mod 2＝0)∪(i mod 2＝1∩j mod 2＝1)}，S2＝{(i，j):(i mod 2＝0∩j mod 2＝1)∪(i mod 2＝1∩j mod 2＝0)}。
In the technical scheme, the time slice rotation rule is that the experiment is carried out according to the time slice according to the preset experiment period and the preset rotation time period, namely the experiment is carried out according to two rotations, one rotation is the experiment period,for example, 1/2017 to 14/2017/1/2017, and another rotation is a rotation period, for example, 00: 00 to 00: 59. 01: 00 to 01: 59. 02: 00 to 02: 59, and the like. After selecting reasonable time slices and time slice rotation rules, dividing test time into time intervals T according to the time slices and the time slice rotation rulesijThe time interval TijThe time length of (i, j) belongs to S1I.e. both i and j are even or both are odd, will be in time interval TijThe obtained experimental data is used as the experimental data of the first group, and when (i, j) belongs to S2I.e. i and j are odd one and even one, will be in time interval TijThe acquired experimental data are used as experimental data of the second group, and the first group and the second group alternately acquire the experimental data, so that the two groups of data are not interfered.
In any of the above technical solutions, preferably, the experimental test information includes test content, test indexes, and test cities.
In the technical scheme, the experimental test information comprises test contents, test indexes and test cities, the test contents comprise the order separation distance and the dynamic dispatching multiple, the test indexes comprise the order response rate, and the test cities are selected by combining specific services.
According to another aspect of the embodiments of the present disclosure, a control experiment system based on time slice rotation is provided, which includes: the determining unit is used for determining experimental test information; the selection unit is used for selecting the time slices and determining the time slice rotation rule according to the experimental test information; the experiment unit is used for carrying out experiments according to the time slices and the time slice rotation rule to obtain a first group of experiment data and a second group of experiment data; and the statistical unit is used for counting the experimental index result according to the experimental data of the first group and the experimental data of the second group after the experiment is finished, and generating an experimental conclusion.
According to the comparison experiment system based on time slice rotation, after the determination unit determines the experiment test information to be carried out, the selection unit selects reasonable time slices and determines the time slice rotation rule to ensure that the absolute change of the two groups of acquired experiment data is small and the significance is not significant, wherein the time slices are time intervals, one experiment is divided into a plurality of time slices, and the time slice rotation rule is that the experiment is carried out according to which experiment period and which rotation time period of the time slices. The experimental unit performs experiments to obtain experimental data of the first group and experimental data of the second group, the statistical unit counts experimental index results to obtain experimental conclusions, the accuracy of experimental data statistics is improved in a mode of grouping comparison statistics of the experimental data, and the problem of inter-group interference caused by a traditional experimental testing method (random shunt) is solved.
The above control experiment system based on time slice rotation according to the embodiment of the present disclosure may further have the following technical features:
in the above technical solution, preferably, the selecting unit is configured to select a time slice range according to the experimental test information; the control experiment system based on the time slice rotation further comprises: the calculation unit is used for acquiring a first group of historical experimental data and a second group of historical experimental data, and calculating an absolute change value and a significance value of the first group of historical experimental data and the second group of historical experimental data; and the selecting unit is also used for selecting the time slice in the time slice range according to the absolute change value and the significance value and determining the time slice rotating rule at the same time.
In this technical solution, the selecting unit selects an approximate range of the time slice according to the test content included in the experimental test information, for example, if the time slice interference of the experimental test content is likely to be large, the time slice is as large as possible, and if the main influence of the test content is short-term effect, the time slice which is relatively small may be selected. After the approximate range of the time slice is determined, the absolute change and the significance of the two groups of experimental data are respectively calculated through the calculating unit according to the historical experimental data, the selecting unit selects a time slice as large as possible according to the absolute change and the significance, then the time slice rotation rule is determined, the reasonable time slice and the time slice rotation rule are selected, the fact that the absolute change of the two groups of acquired experimental data is small and the significance is not significant can be guaranteed, and experimental misjudgment caused by inter-group interference of the experimental data is avoided.
In any of the above technical solutions, preferably, the time slice rotation rule includes a preset experiment period and a preset rotation time period; an experimental unit, specifically configured to: dividing the test time into time intervals T according to the time slices and the time slice rotation ruleij(ii) a If (i, j) ∈ S1Will be at time interval TijThe acquired experimental data are used as a first group of experimental data; if (i, j) ∈ S2Will be at time interval TijThe acquired experimental data are used as a second group of experimental data; wherein the time interval TijThe time length of (1, 2) · n, j ═ 1, 2,... said., n, m, n is a preset cycle period, m is a preset experimental period, and S is a time slice1＝{(i，j):(i mod 2＝0∩j mod 2＝0)∪(i mod 2＝1∩j mod 2＝1)}，S2＝{(i，j):(i mod 2＝0∩j mod 2＝1)∪(i mod 2＝1∩j mod 2＝0)}。
In this technical solution, the time slice rotation rule is that the experiment is performed according to the time slice according to the preset experiment period and the preset rotation period, that is, the experiment is performed according to two rotations, one rotation is the experiment period, for example, 1/2017 to 1/14/2017, and the other rotation is the rotation period, for example, 00: 00 to 00: 59. 01: 00 to 01: 59. 02: 00 to 02: 59, and the like. After selecting reasonable time slices and time slice rotation rules, dividing test time into time intervals T according to the time slices and the time slice rotation rulesijThe time interval TijThe time length of (i, j) belongs to S1I.e. both i and j are even or both are odd, will be in time interval TijThe obtained experimental data is used as the experimental data of the first group, and when (i, j) belongs to S2I.e. i and j are odd one and even one, will be in time interval TijThe acquired experimental data are used as experimental data of the second group, and the first group and the second group alternately acquire the experimental data, so that the two groups of data are not interfered.
In any of the above technical solutions, preferably, the experimental test information includes test content, test indexes, and test cities.
In the technical scheme, the experimental test information comprises test contents, test indexes and test cities, the test contents comprise the order separation distance and the dynamic dispatching multiple, the test indexes comprise the order response rate, and the test cities are selected by combining specific services.
According to yet another aspect of the embodiments of the present disclosure, a computer device is provided, which includes a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the above-mentioned time slice round robin based control experiment method.
In the computer device provided by the embodiment of the disclosure, after experimental test information required to be performed is determined when the processor executes a computer program, a reasonable time slice is selected and a time slice rotation rule is determined to ensure that absolute changes of two groups of acquired experimental data are small and the significance is not significant, wherein the time slice is a time interval, one experiment is divided into a plurality of time slices, and the time slice rotation rule is that the experiment is performed according to which experimental period and which rotation time period according to the time slices. And then, carrying out an experiment to obtain the experimental data of the first group and the experimental data of the second group, further counting the experimental index results to obtain an experimental conclusion, improving the accuracy of experimental data counting by grouping comparison statistics of the experimental data, and avoiding the problem of inter-group interference caused by the traditional experimental test method (random shunt).
According to yet another aspect of an embodiment of the present disclosure, a computer-readable storage medium is proposed, on which a computer program is stored, which, when being executed by a processor, carries out the steps of the slice-rotation based control experiment method according to any one of the above.
In the computer-readable storage medium provided by the embodiment of the disclosure, after the computer program is executed by the processor to determine the experimental test information to be performed, a reasonable time slice is selected and a time slice rotation rule is determined to ensure that absolute changes of two groups of acquired experimental data are small and the significance is not significant, where the time slice is a time interval, one experiment is divided into a plurality of time slices, and the time slice rotation rule is that the experiment is performed according to which experimental period and which rotation time period according to the time slice. And then, carrying out an experiment to obtain the experimental data of the first group and the experimental data of the second group, further counting the experimental index results to obtain an experimental conclusion, improving the accuracy of experimental data counting by grouping comparison statistics of the experimental data, and avoiding the problem of inter-group interference caused by the traditional experimental test method (random shunt).
Additional aspects and advantages of the disclosed embodiments will be set forth in part in the description which follows or may be learned by practice of the disclosed embodiments.
Drawings
The above and/or additional aspects and advantages of the embodiments of the present disclosure will become apparent and readily appreciated from the following description of the embodiments, taken in conjunction with the accompanying drawings of which:
FIG. 1 shows a schematic flow diagram of a time slice rotation-based control experiment method according to an embodiment of the disclosure;
FIG. 2 shows a schematic flow diagram of a control experiment method based on time slice rotation according to another embodiment of the disclosure;
FIG. 3a shows a schematic block diagram of a time slice rotation based control experiment system of one embodiment of the present disclosure;
FIG. 3b shows a schematic block diagram of a time slice rotation based control experiment system of another embodiment of the disclosed embodiments;
FIG. 4 shows a schematic block diagram of a computer apparatus of one embodiment of the disclosed embodiments.
Detailed Description
In order that the above objects, features and advantages of the embodiments of the present disclosure can be more clearly understood, embodiments of the present disclosure will be described in further detail below with reference to the accompanying drawings and detailed description. It should be noted that the embodiments and features of the embodiments of the present application may be combined with each other without conflict.
In the following description, numerous specific details are set forth in order to provide a thorough understanding of the embodiments of the disclosure, however, the embodiments of the disclosure may be practiced in other ways than those described herein, and therefore the scope of the embodiments of the disclosure is not limited to the specific embodiments disclosed below.
In an embodiment of the first aspect of the embodiment of the present disclosure, a comparison experiment method based on a time slice round robin is provided, and fig. 1 shows a schematic flow diagram of the comparison experiment method based on the time slice round robin in one embodiment of the present disclosure. Wherein, the method comprises the following steps:
104, selecting a time slice and determining a time slice rotation rule according to the experimental test information;
106, performing an experiment according to the time slices and the time slice rotation rule to obtain a first group of experiment data and a second group of experiment data;
and 108, after the experiment is finished, counting experiment index results according to the first group of experiment data and the second group of experiment data to generate an experiment conclusion.
According to the comparison experiment method based on the time slice rotation, after experiment test information needing to be carried out is determined, reasonable time slices are selected, the time slice rotation rule is determined, the acquired two groups of experiment data are guaranteed to have small absolute change and insignificant significance, the time slices are time intervals, one experiment is divided into a plurality of time slices, and the time slice rotation rule is that the experiment is carried out according to which experiment period and which rotation time period of the time slices. And then, carrying out an experiment to obtain the experimental data of the first group and the experimental data of the second group, and further counting the experimental index results to obtain an experimental conclusion, so that the accuracy of experimental data counting is improved by carrying out grouping comparison statistics on the experimental data, and the problem of inter-group interference caused by the traditional experimental testing method (random shunt) due to the inter-group interference caused by the traditional experimental testing method (random shunt) is avoided.
Fig. 2 shows a schematic flow chart of a control experiment method based on time slice rotation according to another embodiment of the present disclosure. Wherein, the method comprises the following steps:
and 212, after the experiment is finished, counting experiment index results according to the first group of experiment data and the second group of experiment data to generate an experiment conclusion.
The time slice rotation rule comprises a preset experiment period and a preset rotation time period; time interval TijThe time length of (1, 2) · n, j ═ 1, 2,... said., n, m, n is a preset cycle period, m is a preset experimental period, and S is a time slice1＝{(i，j):(i mod 2＝0∩j mod 2＝0)∪(i mod 2＝1∩j mod 2＝1)}，S2＝{(i，j):(i mod 2＝0∩j mod 2＝1)∪(i mod 2＝1∩j mod 2＝0)}。
In this embodiment, the approximate range of time slices is selected based on the test content included in the experimental test information, e.g., if the time slice interference of the experimental test content is likely to be large, the largest time slice is possible, and if the test content mainly affects short-term effects, the relatively smaller time slice may be selected. After the rough range of the time slice is determined, the absolute change and the significance of two groups of experimental data are respectively calculated according to historical experimental data, a time slice as large as possible is selected according to the absolute change and the significance, and the rotation of the time slice is determinedThe rule can ensure that the absolute change of the acquired two groups of experimental data is small and the significance is not significant by selecting a reasonable time slice and a time slice rotation rule, and avoids experimental misjudgment caused by interference between experimental data groups. The time slice rotation rule is that the experiment is performed according to the preset experiment period and the preset rotation period according to the time slice, that is, the experiment is performed according to two rotations, one rotation is the experiment period, for example, 1/2017 to 1/14/2017, and the other rotation is the rotation period, for example, 00: 00 to 00: 59. 01: 00 to 01: 59. 02: 00 to 02: 59, and the like. After selecting reasonable time slices and time slice rotation rules, dividing test time into time intervals T according to the time slices and the time slice rotation rulesijThe time interval TijThe time length of (i, j) belongs to S1I.e. both i and j are even or both are odd, will be in time interval TijThe obtained experimental data is used as the experimental data of the first group, and when (i, j) belongs to S2I.e. i and j are odd one and even one, will be in time interval TijThe acquired experimental data are used as experimental data of the second group, and the first group and the second group alternately acquire the experimental data, so that the two groups of data are not interfered.
In one embodiment of the present disclosure, preferably, the experimental test information includes test content, test indexes, and test cities.
In this embodiment, the experimental test information includes test content, test indexes, and test cities, where the test content includes, for example, a separation distance and a dynamic tone multiple, the test indexes include, for example, an order response rate, and the test cities are selected in combination with specific services.
In an embodiment of the second aspect of the embodiments of the present disclosure, a control experiment system based on a slice round robin is proposed, and fig. 3a shows a schematic block diagram of a control experiment system 300 based on a slice round robin according to an embodiment of the present disclosure. Among other things, the system 300 includes:
a determination unit 302 for determining experimental test information;
the selecting unit 304 is used for selecting the time slices and determining a time slice rotation rule according to the experimental test information;
the experiment unit 306 is used for performing experiments according to the time slices and the time slice rotation rule to obtain a first group of experiment data and a second group of experiment data;
and the counting unit 308 is configured to count an experiment index result according to the first group of experiment data and the second group of experiment data after the experiment is finished, and generate an experiment conclusion.
In the comparison experiment system 300 based on the rotation of the time slices provided by the embodiment of the present disclosure, after the determination unit 302 determines the experiment test information to be performed, the selection unit 304 selects a reasonable time slice and determines the rotation rule of the time slice to ensure that the absolute change of the acquired two groups of experiment data is small and the significance is not significant, where the time slice is a time interval, one experiment is divided into a plurality of time slices, and the rotation rule of the time slice is that the experiment is performed according to which experiment period and which rotation time period of the time slice. The experiment unit 306 performs an experiment to obtain the first group of experiment data and the second group of experiment data, the statistical unit 308 then calculates the experiment index result to obtain an experiment conclusion, the accuracy of the experiment data statistics is improved by grouping the experiment data and performing comparison statistics, and the problem of inter-group interference caused by a traditional experiment testing method (random shunt) is avoided.
Fig. 3b shows a schematic block diagram of a time slice rotation based control experiment system 300 of another embodiment of the disclosed embodiments. Among other things, the system 300 includes:
a determination unit 302 for determining experimental test information;
the selecting unit 304 is used for selecting the time slices and determining a time slice rotation rule according to the experimental test information;
the experiment unit 306 is used for performing experiments according to the time slices and the time slice rotation rule to obtain a first group of experiment data and a second group of experiment data;
and the counting unit 308 is configured to count an experiment index result according to the first group of experiment data and the second group of experiment data after the experiment is finished, and generate an experiment conclusion.
A selecting unit 304, configured to select a time slice range according to the experimental test information;
the control experiment system 300 based on time slice rotation further comprises: a calculating unit 310, configured to obtain a first group of historical experimental data and a second group of historical experimental data, and calculate an absolute change value and a significance value of the first group of historical experimental data and the second group of historical experimental data;
the selecting unit 304 is further configured to select a time slice from the time slice range according to the absolute change value and the significance value, and determine a time slice rotation rule.
In this embodiment, the selecting unit 304 selects an approximate range of time slices according to the test content included in the experimental test information, for example, if the time slice interference of the experimental test content is likely to be large, the time slice is as large as possible, and if the test content mainly affects the short-term effect, the time slice is relatively small. After the approximate range of the time slice is determined, the absolute change and the significance of the two groups of experimental data are respectively calculated through the calculating unit 310 according to historical experimental data, the selecting unit 304 selects a time slice as large as possible according to the absolute change and the significance, then the time slice rotation rule is determined, the reasonable time slice and the time slice rotation rule are selected, the acquired two groups of experimental data can be guaranteed to have smaller absolute change and insignificant significance, and experimental misjudgment caused by interference among the experimental data groups is avoided.
In one embodiment of the present disclosure, preferably, the time slice rotation rule includes a preset experiment period and a preset rotation period; the experiment unit 306 is specifically configured to: dividing the test time into time intervals T according to the time slices and the time slice rotation ruleij(ii) a If (i, j) ∈ S1Will be at time interval TijThe acquired experimental data are used as a first group of experimental data; if (i, j) ∈ S2Will be at time interval TijThe acquired experimental data are used as a second group of experimental data; wherein the time interval TijThe time length of (1, 2) · n, j ═ 1, 2,... said., n, m, n is a preset cycle time period, m is a preset experimental period，S1＝{(i，j):(i mod 2＝0∩j mod 2＝0)∪(i mod 2＝1∩j mod 2＝1)}，S2＝{(i，j):(i mod 2＝0∩j mod 2＝1)∪(i mod 2＝1∩j mod 2＝0)}。
In this embodiment, the time slice rotation rule is to perform the experiment according to the preset experiment period and the preset rotation period according to the time slice, that is, the experiment is performed according to two rotations, one rotation is the experiment period, for example, 1/2017 to 1/14/2017, and the other rotation is the rotation period, for example, 00: 00 to 00: 59. 01: 00 to 01: 59. 02: 00 to 02: 59, and the like. After selecting reasonable time slices and time slice rotation rules, dividing test time into time intervals T according to the time slices and the time slice rotation rulesijThe time interval TijThe time length of (i, j) belongs to S1I.e. both i and j are even or both are odd, will be in time interval TijThe obtained experimental data is used as the experimental data of the first group, and when (i, j) belongs to S2I.e. i and j are odd one and even one, will be in time interval TijThe acquired experimental data are used as experimental data of the second group, and the first group and the second group alternately acquire the experimental data, so that the two groups of data are not interfered.
In one embodiment of the present disclosure, preferably, the experimental test information includes test content, test indexes, and test cities.
In this embodiment, the experimental test information includes test content, test indexes, and test cities, where the test content includes, for example, a separation distance and a dynamic tone multiple, the test indexes include, for example, an order response rate, and the test cities are selected in combination with specific services.
In a specific embodiment of the disclosure, when strategies such as order splitting, dynamic dispatching and the like are optimized in a travel scene, passengers and drivers are simultaneously influenced, random distribution strategies such as passengers, drivers and orders are adopted, and data conclusions of data influenced by interference of data of an experimental group and data of a comparison group are unreliable. The control experiment method based on the rotation of the time slices is applied to a travel scene, and the problem of interference can be solved well mainly under the short-time influence. In practice, when the transportation capacity is influenced by factors such as abnormal weather, holidays or limit numbers in the experiment period, the data result is unreliable, and the number of days of the experiment can be removed to prolong the experiment period. Performing a control experiment on the passenger to collect data, the method comprising:
(1) and determining the content, the experiment index and the experiment city of the experiment test, wherein the test content comprises the order separation distance and the dynamic dispatching multiple, the experiment index comprises the order response rate, and the experiment city is selected by combining the corresponding business.
(2) And selecting a proper time slice, and determining a time slice rotation rule. The time slices were selected in combination with experimental test content and AA analysis. If the interference of time slices with different test contents is likely to be large, large time slices are selected as much as possible, for example, within 3 hours, and if the test contents of the experiment mainly affect short-term effects, relatively small time slices may be selected, for example, within 30 minutes or within 1 hour. After determining the approximate time slice range according to the test content, AA analysis is performed. The AA analysis is to determine whether the difference between the time slices of the experimental group and the control group is small when the experiment is not carried out, so that the misjudgment of the experimental result is prevented from being influenced by the difference caused by the time slices after the experiment is carried out. There are two ways of AA analysis:
mode 1: the data of two weeks before the experiment are counted, the absolute change and significance of the experimental group and the control group are calculated, a time slice as large as possible is selected, and the absolute change is small and is not statistically significant. If the statistical significance is achieved, the absolute change is very small, the experimental misjudgment is not influenced, and the selection can be carried out. But the former is preferred.
Mode 2: and selecting data of 3 months before the experiment, wherein the selection time is long, the holiday data need to be removed, and absolute change data of a plurality of experimental groups and control groups in two weeks are obtained by utilizing smooth movement according to the day, so that the distribution of the absolute change of the indexes is carved. And calculating the mean value and the significance of the index, and selecting a time slice as large as possible, wherein the mean value is small and is not significant. If the statistical significance is significant, the absolute change is very small, and the experimental misjudgment is not influenced, the selection can be carried out. But the former is preferred. Wherein, the mode 1 has smaller calculation amount and faster result, and the mode 2 has larger calculation amount but more result information. Mode 2 may be considered when the calculation speed problem is not considered. After the time slice is determined, the experiment period and the rotation time period are determined, and the time slice rotation rule can be determined.
The grouping being rotated according to the selected time slices, in particular by dividing the time into Tij1, 2.. said., n, j 1, 2.. said., m, time interval TijThe time length of (i) is represented by time slice, wherein Z-1 and Z-0 represent experimental group and control group, respectively, if (i, j) ∈ S1If Z is 0; if (i, j) ∈ S2If Z is 1; wherein S is1＝{(i，j):(i mod 2＝0∩j mod 2＝0)∪(i mod 2＝1∩j mod 2＝1)}，S2＝{(i，j):(i mod 2＝0∩j mod 2＝1)∪(i mod 2＝1∩j mod 2＝0)}。
(3) And (5) carrying out online experiment and waiting for the experiment to be ended. And determining an experimental subject to execute an experiment or a comparison strategy according to the time slice rotation rule, wherein the experimental subject can select an order, a passenger and the like. The experimental period is typically 2 weeks, and for smaller experiments 4 weeks may be selected.
As shown in table 1, the rotation includes two cycles, the first cycle means that the experimental group and the control group alternately appear in different time intervals in one day, and the second cycle means that the experimental group and the control group alternately appear in the first time interval of each day of the experimental period, taking 1 hour as an example, the experimental period: 2017/01/01 to 2017/1/14, time interval: 00: 00 to 00: 59. 01: 00 to 01: 59. 02: 00 to 02: 59, and the like.
TABLE 1
(4) And (5) counting the experimental index results after the experiment is finished, and giving an experimental conclusion. For example, when the experimental index is the response rate, the response rate is the response amount/the number of orders, and the number of orders are defined as Yij,XijThe response rates of the experimental group and the control group are respectively
And calculating significance T (Z ═ 1) -T (Z ═ 0).
In a third aspect of the embodiments of the present disclosure, a computer apparatus 400 is provided, and fig. 4 shows a schematic block diagram of the computer apparatus 400 according to an embodiment of the present disclosure. Wherein, this computer device 400 includes:
a memory 402, a processor 404 and a computer program stored on the memory 402 and executable on the processor 404, the processor 404 when executing the computer program implementing the steps of any of the above-described slice-round based control experiment methods.
In the computer device 400 provided in the embodiment of the present disclosure, after the processor 404 executes the computer program to determine the experimental test information to be performed, a reasonable time slice is selected and a time slice rotation rule is determined to ensure that absolute changes of two groups of acquired experimental data are relatively small and the significance is not significant, where the time slice is a time interval, one experiment is divided into a plurality of time slices, and the time slice rotation rule is that an experiment is performed according to which experimental cycle and which rotation time period according to the time slice. And then, carrying out an experiment to obtain the experimental data of the first group and the experimental data of the second group, further counting the experimental index results to obtain an experimental conclusion, improving the accuracy of experimental data counting by grouping comparison statistics of the experimental data, and avoiding the problem of inter-group interference caused by the traditional experimental test method (random shunt).
An embodiment of the fourth aspect of the embodiments of the present disclosure provides a computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, implements the steps of the slice-rotation-based control experiment method according to any one of the above.
In the computer-readable storage medium provided by the embodiment of the disclosure, after the computer program is executed by the processor to determine the experimental test information to be performed, a reasonable time slice is selected and a time slice rotation rule is determined to ensure that absolute changes of two groups of acquired experimental data are small and the significance is not significant, where the time slice is a time interval, one experiment is divided into a plurality of time slices, and the time slice rotation rule is that the experiment is performed according to which experimental period and which rotation time period according to the time slice. And then, carrying out an experiment to obtain the experimental data of the first group and the experimental data of the second group, further counting the experimental index results to obtain an experimental conclusion, improving the accuracy of experimental data counting by grouping comparison statistics of the experimental data, and avoiding the problem of inter-group interference caused by the traditional experimental test method (random shunt).
In the description herein, reference to the term "one embodiment," "some embodiments," "a specific embodiment," or the like, means that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the embodiments of the disclosure. In this specification, the schematic representations of the terms used above do not necessarily refer to the same embodiment or example. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.
The above description is only a preferred embodiment of the present disclosure and is not intended to limit the present disclosure, and various modifications and changes may be made to the present disclosure by those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the disclosed embodiments should be included in the scope of protection of the disclosed embodiments.
Claims (8)
1. A control experiment method based on a time slice rotation, comprising:
determining experimental test information;
selecting a time slice and determining a time slice rotation rule according to the experimental test information;
performing an experiment according to the time slice and the time slice rotation rule to obtain a first group of experiment data and a second group of experiment data;
after the experiment is finished, counting experiment index results according to the first group of experiment data and the second group of experiment data to generate an experiment conclusion;
the step of selecting the time slice and determining the time slice rotation rule according to the experimental test information specifically comprises:
selecting a time slice range according to the experimental test information;
acquiring a first group of historical experimental data and a second group of historical experimental data, and calculating absolute change values and significance values of the first group of historical experimental data and the second group of historical experimental data;
and selecting the time slice in the time slice range according to the absolute change value and the significance value, and simultaneously determining the time slice rotation rule.
2. The method of claim 1, wherein the time slice rotation rule comprises a preset experiment period and a preset rotation period;
the step of performing an experiment according to the time slice and the time slice rotation rule to obtain a first group of experimental data and a second group of experimental data specifically includes:
dividing the test time into time intervals T according to the time slices and the time slice rotation ruleij；
If (i, j) ∈ S1Will be at said time interval TijThe acquired experimental data are used as the experimental data of the first group;
if (i, j) ∈ S2Will be at said time interval TijThe acquired experimental data are used as the experimental data of the second group;
wherein the time interval TijThe time length of (a) is the time slice, i is 1, 2,. ·1＝{(i，j):(i mod 2＝0∩j mod 2＝0)∪(i mod 2＝1∩j mod 2＝1)}，S2＝{(i，j):(i mod 2＝0∩j mod 2＝1)∪(i mod 2＝1∩j mod 2＝0)}。
3. The control experiment method based on time slice rotation of claim 1 or 2, wherein the experiment test information comprises test contents, test indexes, test cities.
4. A time slice rotation-based control experiment system, comprising:
the determining unit is used for determining experimental test information;
the selection unit is used for selecting the time slices and determining the time slice rotation rule according to the experimental test information;
the experiment unit is used for carrying out experiments according to the time slices and the time slice rotation rule to obtain a first group of experiment data and a second group of experiment data;
the statistical unit is used for counting the experimental index result according to the experimental data of the first group and the experimental data of the second group after the experiment is finished, and generating an experimental conclusion;
the selection unit is used for selecting a time slice range according to the experimental test information;
the control experiment system based on time slice rotation further comprises:
the calculation unit is used for acquiring a first group of historical experimental data and a second group of historical experimental data, and calculating an absolute change value and a significance value of the first group of historical experimental data and the second group of historical experimental data;
the selecting unit is further configured to select the time slice in the time slice range according to the absolute change value and the significance value, and determine the time slice rotation rule at the same time.
5. The time slice rotation based control experiment system of claim 4, wherein the time slice rotation rule comprises a preset experiment period and a preset rotation period;
the experimental unit is specifically configured to:
dividing the test time into time intervals T according to the time slices and the time slice rotation ruleij；
If (i, j) ∈S1Will be at said time interval TijThe acquired experimental data are used as the experimental data of the first group;
if (i, j) ∈ S2Will be at said time interval TijThe acquired experimental data are used as the experimental data of the second group;
wherein the time interval TijThe time length of (a) is the time slice, i is 1, 2,. ·1＝{(i，j):(i mod 2＝0∩j mod 2＝0)∪(i mod 2＝1∩j mod 2＝1)}，S2＝{(i，j):(i mod 2＝0∩j mod 2＝1)∪(i mod 2＝1∩j mod 2＝0)}。
6. The control experiment system based on time slice rotation of claim 4 or 5, wherein the experiment test information comprises test content, test index, test city.
7. A computer arrangement comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the processor, when executing the computer program, carries out the steps of the slice-rotation based control experiment method according to any one of claims 1 to 3.
8. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the steps of the slice-rotation based control experiment method according to any one of claims 1 to 3.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710899136.4A CN109583670B (en) | 2017-09-28 | 2017-09-28 | Control experiment method, system and computer device based on time slice rotation |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710899136.4A CN109583670B (en) | 2017-09-28 | 2017-09-28 | Control experiment method, system and computer device based on time slice rotation |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109583670A CN109583670A (en) | 2019-04-05 |
| CN109583670B true CN109583670B (en) | 2020-12-04 |
Family
ID=65913917
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710899136.4A Active CN109583670B (en) | 2017-09-28 | 2017-09-28 | Control experiment method, system and computer device based on time slice rotation |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109583670B (en) |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112465604A (en) * | 2020-12-14 | 2021-03-09 | 深圳依时货拉拉科技有限公司 | Order distribution method and device, computer equipment and computer readable storage medium |
| CN113448876B (en) * | 2021-08-31 | 2021-11-19 | 腾讯科技（深圳）有限公司 | Service testing method, device, computer equipment and storage medium |
| CN115361340B (en) * | 2022-08-24 | 2024-02-27 | 深圳依时货拉拉科技有限公司 | AB experiment shunting method and device, computer equipment and storage medium |
| CN117033946B (en) * | 2023-09-21 | 2023-12-22 | 深圳依时货拉拉科技有限公司 | AB experiment method, AB experiment device, computer equipment and computer readable storage medium |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1436005A (en) * | 2002-01-30 | 2003-08-13 | 华邦电子股份有限公司 | Method and system for measuring accuracy of video/audio output synchronization, and analysis means |
| CN101727383A (en) * | 2008-10-16 | 2010-06-09 | 上海市医疗保险信息中心 | Simulation test method and system of database |
| CN101785022A (en) * | 2007-07-03 | 2010-07-21 | 3M创新有限公司 | System and method for generating time-slot samples to which content may be assigned for measuring effects of the assigned content |
| CN104915726A (en) * | 2015-05-11 | 2015-09-16 | 四川汇源吉迅数码科技有限公司 | Traffic flow database generation method based on time slice division |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7773985B2 (en) * | 2003-09-22 | 2010-08-10 | United Parcel Service Of America, Inc. | Symbiotic system for testing electromagnetic signal coverage in areas near transport routes |
| US7183970B1 (en) * | 2005-01-31 | 2007-02-27 | The United States Of America As Represented By The Secretary Of The Navy | Missile time-space-position-information unit test software |
-
2017
- 2017-09-28 CN CN201710899136.4A patent/CN109583670B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1436005A (en) * | 2002-01-30 | 2003-08-13 | 华邦电子股份有限公司 | Method and system for measuring accuracy of video/audio output synchronization, and analysis means |
| CN101785022A (en) * | 2007-07-03 | 2010-07-21 | 3M创新有限公司 | System and method for generating time-slot samples to which content may be assigned for measuring effects of the assigned content |
| CN101727383A (en) * | 2008-10-16 | 2010-06-09 | 上海市医疗保险信息中心 | Simulation test method and system of database |
| CN104915726A (en) * | 2015-05-11 | 2015-09-16 | 四川汇源吉迅数码科技有限公司 | Traffic flow database generation method based on time slice division |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109583670A (en) | 2019-04-05 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109583670B (en) | Control experiment method, system and computer device based on time slice rotation | |
| Liu et al. | A new improved NEH heuristic for permutation flowshop scheduling problems | |
| CN107807625B (en) | End-to-end-based automatic driving system comfort evaluation method and device | |
| Kanet et al. | A weighted modified due date rule for sequencing to minimize weighted tardiness | |
| WO2018095065A1 (en) | Method and apparatus for allocating data object, and electronic device | |
| CN109325185B (en) | Method, device and equipment for determining getting-on point and storage medium | |
| CN108960431A (en) | The prediction of index, the training method of model and device | |
| CN109033966B (en) | Detour detection model training method and device, and detour detection method and device | |
| CN110209686A (en) | Storage, querying method and the device of data | |
| CN108615129B (en) | Transport capacity monitoring method and device and electronic equipment | |
| CN104063265A (en) | Method for comprehensively evaluating virtual resources | |
| WO2016206557A1 (en) | Risk identification method and apparatus | |
| CN111325374A (en) | Method and device for predicting order cancellation probability and electronic equipment | |
| CN111967964B (en) | Intelligent recommending method and device for bank client sites | |
| CN106096047A (en) | User partition preference computational methods based on Information Entropy and system | |
| CN108898831B (en) | Road section condition evaluation method and system based on road high-definition checkpoint data | |
| CN108806241A (en) | A kind of determination method and device of common driving route | |
| CN112183899A (en) | Method, device, equipment and storage medium for determining safety degree prediction model | |
| CN109377317A (en) | Data processing method and device | |
| Ko et al. | Measuring control delay components using second-by-second GPS speed data | |
| CN108040250B (en) | Method and device for evaluating performance of smart television | |
| CN111160385B (en) | Method, device, equipment and storage medium for aggregating mass location points | |
| CN114723071B (en) | Federal learning method and device based on client classification and information entropy | |
| CN106681837A (en) | Data sheet based data eliminating method and device | |
| CN101944142A (en) | Method for determining optimal fusion of multi-source information of spacecraft flight parameters |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |