Grouping Method for Users, Device, and Grouping Method for Data Objects

Abstract

This specification provides a method for grouping users, a device, and a method for grouping data objects. Based on this method, when the number of sample users is relatively small and it is necessary to conduct a grouping experiment using the limited sample users, the identity identifiers, relevance features, and randomness features of the sample users can be obtained first; among them, the relevance features include the features that affect the experimental results of the current grouping experiment; the randomness features include the features that do not affect the experimental results of the current grouping experiment; concatenate the identity identifiers, relevance features, and randomness features of the sample users to obtain the feature string of the sample users; then, according to the feature string of the sample users, divide the sample users into the corresponding user groups. Thus, better randomness can be maintained for different groupings, and the different user groups obtained from each grouping have better homogeneity. Furthermore, the above user groups can be fully utilized for grouping experiments to obtain grouping experiment results with relatively high reference value.

Description

Technical Field

[0001] This specification belongs to the field of Internet technologies, and particularly relates to a method and apparatus for grouping users and a method for grouping data objects. Background Art

[0002] In the Internet field, before a functional application is launched, it is often necessary to conduct user tests on the functional application to be launched. At this time, it is necessary to first group the users participating in the test, and then conduct an AB test on different groups to complete the user test.

[0003] Based on the existing grouping methods, most of them directly perform random splitting on the users participating in the test. In the case of a large number of users participating in the test, the above grouping method can obtain a good grouping effect. However, in the case of a small number of users participating in the test, when using the above grouping method for grouping, problems such as poor homogeneity of users in different groups and unsatisfactory grouping effects are likely to occur during each grouping. For example, due to the poor homogeneity of the control group and the test group obtained by grouping, it may occur that the experimental results obtained from the grouping experiment based on the above control group and test group are significantly unreasonable; and it is also impossible to ensure the randomness of different groupings. For example, it may occur that the grouping results of the first experiment are almost the same as those of the second experiment, thereby affecting the experimental results of the AB test.

[0004] In response to the above problems, no effective solution has been proposed yet. Summary of the Invention

[0005] This specification provides a method and apparatus for grouping users and a method for grouping data objects, which can make different groupings maintain good randomness, and the sample users included in different user groups obtained by each grouping have good homogeneity. Furthermore, the above user groups can be fully utilized to conduct corresponding grouping experiments to obtain experimental results with relatively high reference value.

[0006] An embodiment of this specification provides a method for grouping users, including: obtaining the identity identifier, relevance feature, and randomness feature of a sample user; where the relevance feature includes a feature that affects the experimental result of the current grouping experiment; the randomness feature includes a feature that does not affect the experimental result of the current grouping experiment; splicing the identity identifier, relevance feature, and randomness feature of the sample user to obtain a feature string of the sample user; and dividing the sample user into the corresponding user group according to the feature string of the sample user.

[0007] The embodiments of this specification also provide a method for grouping data objects, including: obtaining the identification information, relevance features, and randomness features of the data objects; wherein, the relevance features include the features that affect the experimental results of the current grouping experiment; the randomness features include the features that do not affect the experimental results of the current grouping experiment; concatenating the identification information, relevance features, and randomness features of the data objects to obtain a feature string of the data objects; and dividing the data objects into the corresponding object groups according to the feature string of the data objects.

[0008] The embodiments of this specification also provide a grouping device for users, including: an obtaining module, configured to obtain the identity identification, relevance features, and randomness features of sample users; wherein, the relevance features include the features that affect the experimental results of the current grouping experiment; the randomness features include the features that do not affect the experimental results of the current grouping experiment; a concatenating module, configured to concatenate the identity identification, relevance features, and randomness features of the sample users to obtain a feature string of the sample users; and a grouping module, configured to divide the sample users into the corresponding user groups according to the feature string of the sample users.

[0009] The embodiments of this specification also provide a server, including a processor and a memory for storing processor-executable instructions, and when the processor executes the instructions, the relevant steps of the user grouping method and / or the data object grouping method are implemented.

[0010] The embodiments of this specification also provide a computer-readable storage medium, on which computer instructions are stored, and when the instructions are executed by a processor, the relevant steps of the user grouping method and / or the data object grouping method are implemented.

[0011] Based on the user grouping method, device, and data object grouping method provided in this specification, when the number of sample users is relatively small, but it is necessary to use the above limited sample users for grouping experiments, the identity identification, relevance features, and randomness features of the sample users can be obtained first; wherein, the above relevance features include the features that affect the experimental results of the current grouping experiment; the above randomness features include the features that do not affect the experimental results of the current grouping experiment; concatenate the identity identification, relevance features, and randomness features of the sample users to obtain a feature string of the sample users; then divide the sample users into the corresponding user groups according to the feature string of the sample users to complete the user grouping for the current grouping experiment and obtain multiple user groups. This can make the grouping results of different times maintain good randomness, and the sample users included in different user groups obtained by each grouping have good homogeneity, effectively reducing the influence of the grouping results on subsequent grouping experiments, and further enabling the above user groups to be fully utilized for specific grouping experiments to obtain grouping experiment results with better accuracy and higher reference value. Brief Description of the Drawings

[0012] To more clearly illustrate the embodiments of this specification, the accompanying drawings required for use in the embodiments will be briefly introduced below. The accompanying drawings in the following description are only some embodiments described in this specification. For those of ordinary skill in the art, without creative efforts, other accompanying drawings can be obtained based on these drawings.

[0013] Figure 1 is a schematic flowchart of a method for grouping users provided by an embodiment of this specification;

[0014] Figure 2 is a schematic diagram of an embodiment of applying the method for grouping users provided by an embodiment of this specification in a scenario example;

[0015] Figure 3 is a schematic diagram of an embodiment of applying the method for grouping users provided by an embodiment of this specification in a scenario example;

[0016] Figure 4 is a schematic flowchart of a method for grouping users provided by another embodiment of this specification;

[0017] Figure 5 is a schematic diagram of the structural composition of a server provided by an embodiment of this specification;

[0018] Figure 6 is a schematic diagram of the structural composition of a user grouping device provided by an embodiment of this specification. Detailed Description of the Embodiments

[0019] To enable those skilled in the art to better understand the technical solutions in this specification, the technical solutions in the embodiments of this specification will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of this specification. Obviously, the described embodiments are only part of the embodiments of this specification, rather than all of them. Based on the embodiments in this specification, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the scope of protection of this specification.

[0020] Refer to Figure 1 As shown, an embodiment of this specification provides a method for grouping users. Specifically implemented, this method may include the following content:

[0021] S101: Obtain the identity identifier, relevance feature, and randomness feature of the sample user; wherein, the relevance feature includes the feature that affects the experimental result of the current grouping experiment; the randomness feature includes the feature that does not affect the experimental result of the current grouping experiment;

[0022] S102: Concatenate the identity identifier, relevance features, and randomness features of the sample user to obtain the feature string of the sample user;

[0023] S103: Divide the sample user into the corresponding user group according to the feature string of the sample user.

[0024] Through the above embodiments, even when the number of sample users is relatively small, the sample users can be grouped well, so that the grouping results obtained each time have good homogeneity, effectively reducing the influence of the grouping results on subsequent grouping experiments; at the same time, it can also ensure that different groupings, for example, the current grouping and the previous grouping or the next grouping, all maintain good randomness, so that each grouping experiment based on the grouping results is independent and does not affect each other.

[0025] In some embodiments, the above user grouping method can be specifically applied to the server side.

[0026] Refer to Figure 2 As shown, the above server can specifically include a background server applied to the network platform (for example, a certain ride-hailing service platform) that can implement functions such as data transmission and data processing. Specifically, the server can be, for example, an electronic device with data operation, storage, and network interaction functions. Or, the server can also be a software program running in the electronic device that provides support for data processing, storage, and network interaction. In this embodiment, the number of servers is not specifically limited. The server can specifically be one server, or several servers, or a server cluster formed by several servers.

[0027] In some embodiments, the above sample user can be specifically understood as a user who is willing to participate in the grouping experiment.

[0028] Among them, the above grouping experiment can be specifically called an AB experiment (or A / B test), which usually cuts out a part of users and randomly divides them into two or more groups with as consistent a population as possible. Among them, one group maintains the existing solution and is called the control group, and the other group or groups use the improved solution and are called the test group; then the test results of the control group and the test group are statistically analyzed to obtain the final experimental result.

[0029] In some embodiments, the above identity identifier of the sample user can be specifically understood as the identification information used to indicate the sample user. Specifically, for example, it can be the username of the sample user, the registered email of the sample user, the name of the sample user, the registered ID of the sample user, the user number of the sample user, and so on.

[0030] It should be noted that all user-related data involved in this specification are obtained and used with the consent and knowledge of the users, and the acquisition, storage, use, processing, etc. of the above user data comply with the relevant provisions of national laws and regulations.

[0031] In some embodiments, the above-mentioned correlation feature can be specifically understood as a feature corresponding to a sample user but capable of affecting the experimental results of the current grouping experiment.

[0032] The above-mentioned randomness feature can be specifically understood as a feature that, relative to the correlation feature, does not affect the experimental results of the current grouping experiment, but has a certain random variation relative to other grouping experiments (for example, the previous grouping experiment or the next grouping experiment).

[0033] Among them, the above-mentioned current grouping experiment can be specifically understood as the AB experiment to be carried out after the current grouping is completed.

[0034] In specific implementation, the above-mentioned correlation feature and randomness feature can be flexibly determined according to the specific application scenario and the specific grouping experiment.

[0035] In some embodiments, the above-mentioned sample users can specifically be a relatively large number of users or a relatively small number of users.

[0036] Based on statistical theory, for a relatively large number of users, when using the conventional method for random grouping, it is often relatively easy to obtain a grouping result with good homogeneity. For a relatively small number of users, when using the conventional method for random grouping, it is easy to have a grouping result with poor homogeneity, which will in turn affect the subsequent grouping experiment. Among them, the above-mentioned good homogeneity can be specifically understood as that the different user groups obtained after grouping have good population consistency.

[0037] In addition, for a relatively small number of users, if the conventional method is used for multiple random groupings, the grouping results obtained in different groupings are likely to be correlated, and the randomness between different groupings cannot be effectively guaranteed. For example, the grouping result obtained in the first grouping may be almost the same as the grouping result obtained in the second grouping.

[0038] The above situations will all affect the subsequent grouping experiment, resulting in the experimental results obtained finally being prone to errors and having relatively low reference value.

[0039] In some embodiments, the above-mentioned sample users can specifically include sample drivers of the online car-hailing service platform, and the above-mentioned grouping experiment includes a business promotion experiment for sample drivers.

[0040] Specifically, for example, refer to Figure 2As shown, the server can be the cloud server of a certain online car-hailing service platform. Currently, this online car-hailing service platform plans to push suitable business activities to the driver users of the platform (for example, providing platform rewards for driver users who receive more orders), in order to encourage driver users to receive and complete more orders on the platform, so as to increase the order-receiving volume and order-completion volume of the platform.

[0041] In order to find the most suitable business activity from multiple candidate business activities and push it to the driver users of the entire platform, it is necessary to first conduct a grouped experiment on multiple candidate business activities, so as to determine the most suitable and best-effective target business activity from multiple candidate business activities according to the experimental results.

[0042] Before conducting the grouped experiment, the server needs to first group the sample drivers who are willing to participate in the grouped experiment. In this scenario example, the number of sample drivers who are willing to participate in the grouped experiment of this platform is relatively small. In order to make full use of the limited number of sample drivers and obtain experimental results with better accuracy and higher reference value, the server can adopt the user grouping method provided in this specification and group the sample drivers according to the sample driver database it holds, so as to divide the sample drivers into three user groups, which are respectively recorded as: User Group 1, User Group 2, and User Group 3.

[0043] When specifically grouping, the server can be connected to the sample driver database of the platform. Among them, the sample driver database stores the identity identifiers of the sample drivers, as well as the attribute data and historical records of the sample drivers. The server can query the sample driver database to obtain the identity identifiers of the sample drivers, as well as the attribute data and historical records of the sample drivers; and further determine the correlation characteristics of the sample drivers according to the attribute data and historical records of the sample drivers; then, the server can obtain the randomness characteristics for the current grouped experiment; and then splice the identity identifiers, correlation characteristics, and randomness characteristics of the sample drivers to obtain the characteristic string of the sample drivers. Furthermore, the sample drivers can be grouped.

[0044] Then randomly select User Group 1 as the control group, select User Group 2 as Test Group 1, and select User Group 3 as Test Group 2.

[0045] Furthermore, the server can, at the same time, push the control business activity (for example, push a blank business activity, that is, do not push a business activity) to the sample drivers in the control group; push Candidate Business Activity 1 to the sample drivers in Test Group 1; and push Candidate Business Activity 2 to the sample drivers in Test Group 2. At the same time, the server can respectively collect and count the average values of the order-completion volumes of the sample drivers in the control group, Test Group 1, and Test Group 2 during the test time period as feedback data. For example, the first feedback data corresponding to the control group, the second feedback data 1 corresponding to Test Group 1, and the second feedback data 2 corresponding to Test Group 2.

[0046] Next, based on the first feedback data, the second feedback data 1, and the second feedback data 2, the server can, through data statistical analysis, determine whether the act of pushing business activities to the sample drivers itself can encourage an increase in the order completion volume of the sample drivers. In the case where it is determined that it can encourage an increase in the order completion volume of the sample drivers, further, based on the second feedback data 1 and the second feedback data 2, through data statistical analysis, determine which of the candidate business activities 1 and candidate business activities 2 can more effectively encourage an increase in the order completion volume of the sample drivers, and determine the candidate activity as the target business activity. Then push the target business activity to all driver users of the platform.

[0047] In some embodiments, the relevance features may specifically include at least one of the following: the order completion volume within a preset time period, the cumulative order completion volume, the time interval between the last order completion time and the current time within a preset time period, etc. Among them, the preset time period can be the most recent week, the most recent day, etc.

[0048] Furthermore, the above-mentioned relevance features may also include: the order receiving volume within a preset time period, the statistical result of location information within a preset time period, the order completion income within a preset time period, and so on.

[0049] Of course, it should be noted that the above-listed relevance features are only illustrative. In specific implementation, according to the specific situation and processing requirements, other appropriate features may also be introduced as the above-mentioned relevance features. This specification does not make any limitations in this regard.

[0050] In some embodiments, the randomness features include at least one of the following: the experiment date of the current sub-group experiment, the random number generated for the current sub-group experiment, the current weather data (such as temperature, humidity, wind force, etc.), and so on.

[0051] Furthermore, the above random features may also include: the current platform access traffic, the current stock market index, the experiment time of the current sub-group experiment, and so on.

[0052] Of course, it should be noted that the above-listed randomness features are only illustrative. In specific implementation, according to the specific situation and processing requirements, other appropriate features may also be introduced as the above-mentioned randomness features. This specification does not make any limitations in this regard.

[0053] In some embodiments, during specific implementation, the identity identifier, relevance features, and randomness features of a sample user can be concatenated together to obtain a feature string of the sample user. The feature string obtained in this way not only takes into account the identity identifier and relevance features of the corresponding sample user, but also the randomness features of the sample user. Subsequently, when randomly grouping the sample users based on the above feature string, on the one hand, since the feature string used for grouping already contains the relevance features that affect the current grouping experiment, it is possible to make the different user groups obtained by grouping have better homogeneity; on the other hand, since the feature string used for grouping also contains randomness features that have random differences from other grouping experiments, it is possible to make the grouping results obtained in this grouping be randomly different from the grouping results obtained in other groupings, ensuring the randomness of different groupings.

[0054] In some embodiments, in order to use both the identity identifier and relevance features of the sample user, and at the same time to avoid the leakage of the identity identifier and relevance features of the sample user and protect the data privacy of the sample user, during specific implementation, the obtained identity identifier and relevance features of the sample user can be encrypted respectively according to a preset encryption rule to obtain the ciphertext data of the identity identifier of the sample user and the ciphertext data of the relevance features; then the ciphertext data of the identity identifier of the sample user, the ciphertext data of the relevance features, and the randomness features are concatenated to obtain a feature string for the sample user. This can avoid the leakage of the identity identifier and relevance features of the sample user during the process of concatenating the feature string, and at the same time can also prevent a third party from reverse-inferring relevant information such as the identity identifier and relevance features of the sample user based on the feature string of the sample user.

[0055] In some embodiments, referring to Figure 3 As shown, during specific concatenation, according to a preset concatenation rule, for example, in the order of first the identity identifier, then the relevance features, and finally the randomness features, the identity identifier, relevance features, and randomness features of the same sample user are concatenated in sequence to obtain a feature string for the sample user.

[0056] In some embodiments, when dividing the sample users into the corresponding user groups according to the feature string of the sample user, during specific implementation, the following content may be included:

[0057] S1: Encrypt the feature string of the sample user according to a preset encryption algorithm to obtain the ciphertext string of the sample user;

[0058] S2: Determine the number of user groups in the current grouping experiment;

[0059] S3: Divide the sample users into corresponding user groups according to the ciphertext strings of the sample users and the number of user groups in the current sub-group experiment.

[0060] In some embodiments, the preset encryption algorithm may specifically include: md5 encryption algorithm, etc. Among them, the above md5 encryption algorithm may specifically be a kind of one-way hashing algorithm. In specific implementation, in addition to using the md5 encryption algorithm, other suitable encryption algorithms may also be used, for example, hash algorithm, etc. as the preset encryption algorithm.

[0061] In specific implementation, by encrypting the feature strings of the sample users according to the preset encryption algorithm, the directly concatenated feature strings can be converted into strings in a unified format for subsequent processing; at the same time, through the encryption process, it is also possible to avoid the leakage of associated features, identity identifiers, etc. spliced in the feature strings during subsequent processing, and can better protect information security.

[0062] In some embodiments, in specific implementation, the number of user groups in the current sub-group experiment may be determined according to the specific situation of the current sub-group experiment. For example, if there are two candidate business activities participating in the current sub-group experiment, the number of user groups in the current sub-group experiment may be determined to be 3, etc.

[0063] In some embodiments, the above-mentioned dividing the sample users into corresponding user groups according to the ciphertext strings of the sample users and the number of user groups in the current sub-group experiment may specifically include the following contents in specific implementation:

[0064] S1: Perform decimal conversion on the ciphertext strings of the sample users to obtain the digital strings of the sample users;

[0065] S2: Perform a remainder operation according to the digital strings of the sample users and the number of user groups in the current sub-group experiment to determine the user group corresponding to the sample users; and divide the sample users into this user group.

[0066] In specific implementation, by performing decimal conversion on the ciphertext strings of the sample users, the ciphertext strings of the sample users can be further converted into unified decimal digital strings that are convenient for subsequent operations.

[0067] In some embodiments, before performing decimal conversion on the ciphertext strings of the sample users, it may further include: detecting whether the ciphertext strings of the sample users belong to decimal digital strings; in the case of determining that the ciphertext strings of the sample users belong to decimal digital strings, a remainder operation may also be directly performed on the ciphertext strings of the sample users to divide the sample users into the corresponding user groups.

[0068] In some embodiments, a remainder operation is performed according to the digital string of the sample user and the number of user groups in the current sub-group experiment to determine the user group corresponding to the sample user. Specifically, it may include: determining the remainder indication parameters corresponding to each user group according to the number of user groups in the current sub-group experiment; dividing the digital string of the sample user by the number of user groups in the current sub-group experiment to obtain a remainder value; comparing the remainder value with the remainder indication parameter, and determining the user group corresponding to the sample user according to the comparison result.

[0069] Specifically, for example, if the number of user groups in the current sub-group experiment is 3, the remainder indication parameter corresponding to user group 1 can be determined as 0, the remainder indication parameter corresponding to user group 2 can be determined as 1, and the remainder indication parameter corresponding to user group 3 can be determined as 2. Then, the digital string of the current sample user can be divided by 3 to obtain the remainder value. Comparing the remainder value with the remainder indication parameter, the corresponding comparison result can be obtained. According to the comparison result, if the remainder value is equal to 0, it can be determined that the user group corresponding to the sample user is user group 1; if the remainder value is equal to 1, it can be determined that the user group corresponding to the sample user is user group 2; if the remainder value is 2, it can be determined that the user group corresponding to the sample user is user group 3. Furthermore, the sample user can be divided into the corresponding user group.

[0070] In the above manner, the user group corresponding to each sample user can be determined; and each sample user can be divided into the corresponding user group to complete the grouping of the sample users and obtain the user grouping for the current sub-group experiment. The user grouping obtained in this way has good homogeneity on the one hand; on the other hand, it has a certain random difference compared with the user groupings of other times, which can support the repeated use of the above sample users for multiple grouping experiments.

[0071] In some embodiments, after dividing the sample users into the corresponding user groups according to the feature strings of the sample users, the method may further include the following content when specifically implemented:

[0072] S1: Determine the test group and the control group from the user groups;

[0073] S2: Push the test service to the sample drivers in the test group and collect the first feedback data of the sample drivers in the test group for the test service; push the control service to the sample drivers in the control group and collect the second feedback data of the sample drivers in the control group for the control service;

[0074] S3: Determine the experimental result of the current sub-group experiment according to the first feedback data and the second feedback data.

[0075] In specific implementation, a user group can be randomly determined from multiple user groups as the control group, and the remaining user groups are determined as the test groups. Among them, the test group can include one or more user groups. Different test services can be pushed to different user groups in the test group, or the same test service can be pushed.

[0076] When specifically conducting the grouping experiment, only the control service for comparison, such as a blank service, can be pushed to the sample drivers in the control group, or no service can be pushed to the sample drivers in the control group. On the contrary, the test service can be pushed to the sample drivers in the test group.

[0077] Among them, the above-mentioned test service can be specifically understood as the service to be pushed to all driver users of the platform. For example, coupon promotion service, completion reward activity, automatic order receiving service, etc. Of course, the above-listed test services are only illustrative. In specific implementation, according to the specific situation and processing requirements, the above test service can also include other types and other content services, such as order snatching function service, actively pushing passenger orders service, pushing return trip orders service, etc. This specification does not limit this.

[0078] The above first feedback data can be specifically understood as the index parameters collected for the sample drivers in the control group after pushing the control service to the control group. For example, the operation data of the sample drivers in the control group for the control service (including: acceptance operation or rejection operation); another example is the number of completed orders and the number of received orders of the sample drivers in the control group after being reached by the control service, etc.

[0079] Similarly, the above second feedback data can be specifically understood as the index parameters collected for the sample drivers in the test group after pushing the test service to the test group. For example, the operation data of the sample drivers in the control group for the control service (including: acceptance operation or rejection operation); another example is the number of completed orders and the number of received orders of the sample drivers in the control group after being reached by the control service, etc.

[0080] In some embodiments, in specific implementation, according to the first feedback data and the second feedback data, through data analysis, it can be determined whether the test service can obtain relatively better service effects compared to the control service; further, it can also be determined which test service among different test services can obtain relatively better service effects, and then the experimental results for the current grouping experiment can be obtained.

[0081] In some embodiments, after determining the experimental results of the current grouping experiment, when the method is specifically implemented, it may further include the following content:

[0082] S1: According to the experimental results of the current grouping experiment, adjust the test service to obtain a target service that meets the requirements;

[0083] S2: Push the target service to driver users of the online car-hailing service platform.

[0084] In this embodiment, specifically, when it is determined according to the experimental results of the current sub-group experiment that the test service does not obtain relatively better service effects compared to the control service, the test service can be recreated; and according to the recreated test service, the next sub-group experiment is carried out on the sample users.

[0085] When it is determined according to the experimental results of the current sub-group experiment that the test service obtains relatively better service effects compared to the control service, and the test service includes one test service, this test service can be determined as the target service.

[0086] When it is determined according to the experimental results of the current sub-group experiment that the test service obtains relatively better service effects compared to the control service, and the test service includes multiple test services, the test service with the relatively best service effect can be selected from the multiple test services as the target service according to the experimental results; or, the test service with the relatively best service effect can be selected from the multiple test services as the service to be pushed, and the service to be pushed is improved specifically according to other test services, and the improved service is used as the target service.

[0087] Through the above embodiments, the target service with relatively good effects can be determined according to the experimental results of the current sub-group experiment, so as to be pushed to driver users of the online car-hailing service platform to obtain better service effects.

[0088] As can be seen from the above, based on the user grouping method provided in this specification, when the number of sample users is relatively small, but it is necessary to use the sample users for sub-group experiments, the identity identifiers, relevance features, and randomness features of the sample users can be obtained first; among them, the above-mentioned relevance features include the features that affect the experimental results of the current sub-group experiment; the above-mentioned randomness features include the features that do not affect the experimental results of the current sub-group experiment, but have certain random differences compared to other sub-group experiments; the identity identifiers, relevance features, and randomness features of the sample users are spliced to obtain the feature string of the sample users; then, according to the feature string of the sample users, the sample users are divided into the corresponding user groups. Thereby, it can be ensured that different sub-groupings can maintain good randomness, so that the above-mentioned limited number of sample users can be reused multiple times for sub-group experiments; and, it can also ensure that the sample users included in different user groups obtained by each sub-grouping have good homogeneity, effectively reducing the influence of the sub-grouping results on subsequent sub-group experiments, and then the above-mentioned user groups can be fully utilized for corresponding sub-group experiments to obtain experimental results with better accuracy and higher reference value.

[0089] Refer to Figure 4 As shown, the embodiment of the present specification also provides a method for grouping data objects. Specifically, when the method is implemented, it may include the following content:

[0090] S401: Obtain the identification information, relevance features, and randomness features of the data object; wherein, the relevance features include the features that affect the experimental results of the current grouping experiment; the randomness features include the features that do not affect the experimental results of the current grouping experiment;

[0091] S402: Concatenate the identification information, relevance features, and randomness features of the data object to obtain the feature string of the data object;

[0092] S403: Divide the data object into the corresponding object group according to the feature string of the data object.

[0093] In some embodiments, the data object may specifically include at least one of the following: sample users, sample signals, sample products, etc. Among them, the sample users may specifically include: sample drivers and / or sample passengers, etc.

[0094] Of course, it should be noted that the above-listed data objects are only illustrative. Specifically, when implemented, according to the specific application scenarios and processing requirements, the grouping method of the data object provided by the embodiment of the present specification can also be applied to group other types of data objects, such as order objects, etc. The present specification does not limit this.

[0095] In some embodiments, after dividing the data object into the corresponding object group according to the feature string of the data object, the method further includes: performing a grouping experiment according to the obtained object group.

[0096] Through the above embodiments, it is possible to more effectively divide a relatively small number of data objects into the corresponding object groups, complete the grouping of the data objects, so that different groupings all maintain good randomness, and the data objects included in different object groups obtained each time have good homogeneity. Furthermore, the object groups can be fully utilized to conduct grouping experiments and obtain experimental results with relatively high reference value.

[0097] The embodiments of this specification also provide a server, including a processor and a memory for storing processor-executable instructions. When specifically implemented, the processor may execute the following steps according to the instructions: obtain the identity identifier, relevance features, and randomness features of a sample user; wherein, the relevance features include features that affect the experimental results of the current grouping experiment; the randomness features include features that do not affect the experimental results of the current grouping experiment; splice the identity identifier, relevance features, and randomness features of the sample user to obtain a feature string of the sample user; and divide the sample user into the corresponding user group according to the feature string of the sample user.

[0098] To be able to more accurately complete the above instructions, refer to Figure 5 As shown, the embodiments of this specification also provide another specific server. Among them, the server includes a network communication port 501, a processor 502, and a memory 503. The above structures are connected by internal cables so that each structure can perform specific data interactions.

[0099] Among them, the network communication port 501 can specifically be used to obtain the identity identifier, relevance features, and randomness features of a sample user; wherein, the relevance features include features that affect the experimental results of the current grouping experiment; the randomness features include features that do not affect the experimental results of the current grouping experiment;

[0100] The processor 502 can specifically be used to splice the identity identifier, relevance features, and randomness features of the sample user to obtain a feature string of the sample user; and divide the sample user into the corresponding user group according to the feature string of the sample user.

[0101] The memory 503 can specifically be used to store the corresponding instruction program.

[0102] In this embodiment, the network communication port 501 can be bound to different communication protocols, so as to send or receive different data virtual ports. For example, the network communication port can be a port responsible for web data communication, can also be a port responsible for FTP data communication, or can also be a port responsible for mail data communication. In addition, the network communication port can also be a physical communication interface or communication chip. For example, it can be a wireless mobile network communication chip, such as GSM, CDMA, etc.; it can also be a Wifi chip; it can also be a Bluetooth chip.

[0103] In this embodiment, the processor 502 may be implemented in any suitable manner. For example, the processor may take the form of, for example, a microprocessor or a processor, a computer-readable medium storing computer-readable program code (such as software or firmware) executable by the (micro)processor, logic gates, switches, an Application Specific Integrated Circuit (ASIC), a programmable logic controller, and an embedded microcontroller, etc. This specification does not make any limitations.

[0104] In this embodiment, the memory 503 may include multiple levels. In a digital system, anything that can store binary data can be a memory; in an integrated circuit, a circuit with a storage function without a physical form is also called a memory, such as RAM, FIFO, etc.; in a system, a storage device with a physical form is also called a memory, such as a memory stick, a TF card, etc.

[0105] This specification embodiment also provides another server, including a processor and a memory for storing instructions executable by the processor. When specifically implemented, the processor may execute the following steps according to the instructions: obtain the identification information, relevance features, and randomness features of a data object; where the relevance features include features that affect the experimental results of the current sub-group experiment; the randomness features include features that do not affect the experimental results of the current sub-group experiment; splice the identification information, relevance features, and randomness features of the data object to obtain a feature string of the data object; divide the data object into the corresponding object group according to the feature string of the data object.

[0106] This specification embodiment also provides a computer storage medium based on the above user grouping method. The computer storage medium stores computer program instructions, which when executed implement: obtain the identity identification, relevance features, and randomness features of a sample user; where the relevance features include features that affect the experimental results of the current sub-group experiment; the randomness features include features that do not affect the experimental results of the current sub-group experiment; splice the identity identification, relevance features, and randomness features of the sample user to obtain a feature string of the sample user; divide the sample user into the corresponding user group according to the feature string of the sample user.

[0107] The embodiments of this specification also provide a computer storage medium based on the above data object grouping method. The computer storage medium stores computer program instructions, which when executed implement the following: obtaining the identification information, relevance features, and randomness features of a data object; wherein, the relevance features include features that affect the experimental results of the current grouping experiment; the randomness features include features that do not affect the experimental results of the current grouping experiment; concatenating the identification information, relevance features, and randomness features of the data object to obtain a feature string of the data object; and dividing the data object into the corresponding object group according to the feature string of the data object.

[0108] In this embodiment, the above storage medium includes but is not limited to Random Access Memory (RAM), Read-Only Memory (ROM), Cache, Hard Disk Drive (HDD), or Memory Card. The memory can be used to store computer program instructions. The network communication unit can be set according to the standards specified by the communication protocol and is used for the interface of network connection communication.

[0109] In this embodiment, the functions and effects specifically implemented by the program instructions stored in this computer storage medium can be explained in comparison with other embodiments and will not be elaborated here.

[0110] Refer to Figure 6 As shown, at the software level, the embodiments of this specification also provide a grouping device for users, which specifically may include the following structural modules:

[0111] An obtaining module 601, which can specifically be used to obtain the identity identification, relevance features, and randomness features of a sample user; wherein, the relevance features include features that affect the experimental results of the current grouping experiment; the randomness features include features that do not affect the experimental results of the current grouping experiment;

[0112] A concatenating module 602, which can specifically be used to concatenate the identity identification, relevance features, and randomness features of the sample user to obtain a feature string of the sample user;

[0113] A grouping module 603, which can specifically be used to divide the sample user into the corresponding user group according to the feature string of the sample user.

[0114] In some embodiments, the sample user may specifically include sample drivers of an online car-hailing service platform, and the grouping experiment may specifically include a business promotion experiment for the sample drivers.

[0115] In some embodiments, the relevance feature may specifically include at least one of the following: the number of completed orders within a preset time period, the cumulative number of completed orders, the time interval between the last completed order time and the current time within a preset time period, etc.

[0116] In some embodiments, the randomness feature may specifically include at least one of the following: the experiment date of the current sub-group experiment, the random number generated for the current sub-group experiment, the current weather data, etc.

[0117] In some embodiments, when the grouping module 603 is specifically implemented, the sample users may be divided into the corresponding user groups according to the following method: encrypt the feature string of the sample user according to a preset encryption algorithm to obtain the ciphertext string of the sample user; determine the number of user groups in the current sub-group experiment; and divide the sample users into the corresponding user groups according to the ciphertext string of the sample user and the number of user groups in the current sub-group experiment.

[0118] In some embodiments, the preset encryption algorithm may specifically include: the md5 encryption algorithm, etc.

[0119] In some embodiments, when the grouping module 603 is specifically implemented, the sample users may be divided into the corresponding user groups according to the following method: perform decimal conversion on the ciphertext string of the sample user to obtain the digital string of the sample user; perform a modulo operation according to the digital string of the sample user and the number of user groups in the current sub-group experiment to determine the user group corresponding to the sample user; and divide the sample users into this user group.

[0120] In some embodiments, after dividing the sample users into the corresponding user groups according to the feature string of the sample users, when the device is specifically implemented, it may also be used to determine a test group and a control group from the user groups; push test services to the sample drivers in the test group and collect first feedback data of the sample drivers in the test group for the test services; push control services to the sample drivers in the control group and collect second feedback data of the sample drivers in the control group for the control services; and determine the experimental result of the current sub-group experiment according to the first feedback data and the second feedback data.

[0121] In some embodiments, after determining the experimental result of the current sub-group experiment, when the device is specifically implemented, it may also be used to adjust the test service according to the experimental result of the current sub-group experiment to obtain a target service that meets the requirements; and push the target service to the driver users of the online ride-hailing service platform.

[0122] It should be noted that the units, devices, or modules described in the above embodiments can be specifically implemented by computer chips or entities, or by products with certain functions. For the convenience of description, when describing the above devices, they are divided into various modules according to functions and described separately. Of course, when implementing this specification, the functions of each module can be implemented in the same or multiple software and / or hardware, or the modules that implement the same function can be implemented by a combination of multiple sub-modules or sub-units, etc. The device embodiments described above are merely illustrative. For example, the division of the units is only a logical function division, and there may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed coupling or direct coupling or communication connection to each other can be through some interfaces, and the indirect coupling or communication connection of the devices or units can be in electrical, mechanical, or other forms.

[0123] As can be seen from the above, based on the user grouping device provided in the embodiments of this specification, it is possible to make the grouping of different times maintain good randomness, and the sample users included in the different user groups obtained by each grouping have good homogeneity. Furthermore, the above user groups can be fully utilized for grouping experiments to obtain experimental results with relatively high reference value.

[0124] The embodiments of this specification also provide a grouping device for data objects. Specifically, when implemented, it can include the following structural modules:

[0125] An acquisition module, which can be specifically used to acquire the identification information, relevance features, and randomness features of the data object; among them, the relevance features include the features that affect the experimental results of the current grouping experiment; the randomness features include the features that do not affect the experimental results of the current grouping experiment;

[0126] A splicing module, which can be specifically used to splice the identification information, relevance features, and randomness features of the data object to obtain a feature string of the data object;

[0127] A grouping module, which can be specifically used to divide the data object into the corresponding object group according to the feature string of the data object.

[0128] In some embodiments, the data object can specifically include at least one of the following: sample users, sample signals, sample products, etc.

[0129] Through the above embodiments, the grouping device for data objects provided based on the embodiments of this specification can ensure good randomness in different groupings of data objects, and the data objects included in different object groups obtained by each grouping have good homogeneity. Furthermore, the above object groups can be fully utilized for grouping experiments to obtain experimental results with relatively high reference value.

[0130] Although this specification provides method operation steps as described in the embodiments or flowcharts, more or fewer operation steps may be included based on conventional or non-creative means. The step sequences listed in the embodiments are only one of the many ways of executing steps and do not represent the only execution sequence. When actually executed by a device or client product, it can be executed in the method sequence shown in the embodiments or the drawings or in parallel (for example, in an environment with parallel processors or multi-threaded processing, or even in a distributed data processing environment). The term "including", "comprising" or any other variant thereof is intended to cover non-exclusive inclusion, such that a process, method, product or device including a series of elements not only includes those elements but also includes other elements not explicitly listed, or also includes elements inherent to such process, method, product or device. Without further limitation, there is no exclusion of the existence of additional identical or equivalent elements in the process, method, product or device including the said elements. The terms "first", "second", etc. are used to denote names and do not denote any particular order.

[0131] Those skilled in the art also know that in addition to implementing the controller in the form of pure computer-readable program code, the method steps can be logically programmed to enable the controller to be implemented in the form of logic gates, switches, application-specific integrated circuits, programmable logic controllers, embedded microcontrollers, etc. to achieve the same functions. Therefore, such a controller can be regarded as a hardware component, and the devices included therein for implementing various functions can also be regarded as the structures within the hardware component. Or even, the devices for implementing various functions can be regarded as both software modules for implementing the method and the structures within the hardware component.

[0132] This specification can be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, classes, etc. that perform specific tasks or implement specific abstract data types. This specification can also be practiced in a distributed computing environment where tasks are performed by remote processing devices connected through a communication network. In a distributed computing environment, program modules can be located in local and remote computer storage media including storage devices.

[0133] As can be seen from the description of the above embodiments, those skilled in the art can clearly understand that this specification can be implemented by means of software plus a necessary general hardware platform. Based on such an understanding, the technical solution of this specification can essentially be embodied in the form of a software product, which can be stored in a storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions for causing a computer device (which can be a personal computer, mobile terminal, server, or network device, etc.) to execute the methods described in various embodiments or some parts of the embodiments of this specification.

[0134] The various embodiments in this specification are described in a progressive manner. For the same or similar parts between the various embodiments, reference can be made to each other. The key point of each embodiment is to illustrate the differences from other embodiments. This specification can be used in many general or special computer system environments or configurations. For example: personal computers, server computers, handheld or portable devices, tablet devices, multi-processor systems, microprocessor-based systems, set-top boxes, programmable electronic devices, network PCs, minicomputers, mainframe computers, distributed computing environments including any of the above systems or devices, and so on.

[0135] Although this specification is depicted through embodiments, those of ordinary skill in the art know that this specification has many variations and changes without departing from the spirit of this specification. It is hoped that the appended claims will cover these variations and changes without departing from the spirit of this specification.

Claims

1. A method for grouping users, characterized in that, Including: Obtain the identity identifier, relevance features, and randomness features of a sample user; wherein, the relevance features include features that affect the experimental results of the current sub-group experiment; the randomness features include features that do not affect the experimental results of the current sub-group experiment; the relevance features include at least one of the following: the number of completed orders within a preset time period, the cumulative number of completed orders, the time interval between the last completed order time within a preset time period and the current time; the randomness features include at least one of the following: the experimental date of the current sub-group experiment, the random number generated for the current sub-group experiment, the current weather data; Concatenate the identity identifier, relevance features, and randomness features of the sample user to obtain a feature string of the sample user; Divide the sample user into the corresponding user group according to the feature string of the sample user.

2. The method according to claim 1, characterized in that, The sample user includes sample drivers of an online car-hailing service platform, and the sub-group experiment includes a business promotion experiment for sample drivers.

3. The method according to claim 1, characterized in that, Dividing the sample user into the corresponding user group according to the feature string of the sample user includes: Perform encryption processing on the feature string of the sample user according to a preset encryption algorithm to obtain a ciphertext string of the sample user; Determine the number of user groups in the current sub-group experiment; Divide the sample user into the corresponding user group according to the ciphertext string of the sample user and the number of user groups in the current sub-group experiment.

4. The method according to claim 3, characterized in that, Dividing the sample user into the corresponding user group according to the ciphertext string of the sample user and the number of user groups in the current sub-group experiment includes: Perform decimal conversion on the ciphertext string of the sample user to obtain a digital string of the sample user; Perform a remainder operation according to the digital string of the sample user and the number of user groups in the current sub-group experiment to determine the user group corresponding to the sample user; and divide the sample user into this user group.

5. The method according to claim 2, characterized in that, After dividing the sample user into the corresponding user group according to the feature string of the sample user, the method further includes: Determine a test group and a control group from the user group; Push a test service to the sample drivers in the test group and collect first feedback data of the sample drivers in the test group for the test service; push a control service to the sample drivers in the control group and collect second feedback data of the sample drivers in the control group for the control service; Determine the experimental results of the current sub-group experiment according to the first feedback data and the second feedback data.

6. The method according to claim 5, characterized in that, After determining the experimental results of the current sub-group experiment, the method further includes: Adjust the test service according to the experimental results of the current sub-group experiment to obtain a target service that meets the requirements; Push the target service to the driver users of the online car-hailing service platform.

7. A method for grouping data objects, characterized in that, Including: Obtain the identification information, relevance features, and randomness features of the data object; wherein, the relevance features include features that affect the experimental results of the current sub-group experiment; the randomness features include features that do not affect the experimental results of the current sub-group experiment; the relevance features include at least one of the following: the number of completed orders within a preset time period, the cumulative number of completed orders, the time interval between the last completed order time within a preset time period and the current time; the randomness features include at least one of the following: the experimental date of the current sub-group experiment, the random number generated for the current sub-group experiment, the current weather data; Concatenate the identification information, relevance features, and randomness features of the data object to obtain the feature string of the data object; Divide the data object into the corresponding object group according to the feature string of the data object.

8. A user grouping device, characterized in that, It includes: An acquisition module, configured to acquire the identity identification, relevance features, and randomness features of the sample user; wherein, the relevance features include features that affect the experimental results of the current sub-group experiment; the randomness features include features that do not affect the experimental results of the current sub-group experiment; the relevance features include at least one of the following: the number of completed orders within a preset time period, the cumulative number of completed orders, the time interval between the last completed order time within a preset time period and the current time; the randomness features include at least one of the following: the experimental date of the current sub-group experiment, the random number generated for the current sub-group experiment, the current weather data; A concatenation module, configured to concatenate the identity identification, relevance features, and randomness features of the sample user to obtain the feature string of the sample user; A grouping module, configured to divide the sample user into the corresponding user group according to the feature string of the sample user.

9. A server, characterized in that, It includes a processor and a memory for storing processor-executable instructions, and when the processor executes the instructions, it implements the steps of the method according to any one of claims 1 to 6, or 7.

10. A computer-readable storage medium, characterized in that, Computer instructions are stored thereon, and when the instructions are executed by the processor, it implements the steps of the method according to any one of claims 1 to 6, or 7.

Images