Gray release method and device for mobile application and storage medium

Abstract

The invention discloses a gray release method and device for a mobile application and a storage medium. The method relates to the field of artificial intelligence, and comprises the steps that under the condition that authorization of a first target user is obtained, behavior data, aiming at a target mobile application, of the first target user is obtained, and the behavior data at least comprises the click rate and the sharing rate; inputting the behavior data into a gray scale prediction model, and outputting a posterior probability corresponding to the behavior data by using the gray scale prediction model, the posterior probability being used for representing an acceptance degree of a second target user for the target mobile application; obtaining a target gray scale promotion proportion corresponding to the posterior probability according to a gray scale regulation and control model; and adjusting the gray release of the target mobile application according to the target gray promotion proportion. Through the method and the device, the problem of relatively poor real-time performance of gray release in related technologies is solved.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of artificial intelligence, in particular to a gray release method and device for mobile applications and a storage medium. BACKGROUND

[0002] Gray release is a process in software development and deployment where a new version of an application is gradually and controlledly promoted to a user group to evaluate its performance, stability and user acceptance. Through gray release, software bugs that may exist in the new version of the application can be fixed in a timely manner.

[0003] In the prior art, gray release usually adopts a manual decision-making method to adjust the gray ratio, that is, after sufficient data is collected, the gray ratio is manually adjusted based on data reports. Although this method can ensure the accuracy of the decision, the response time is relatively long and cannot make timely adjustments when the user's acceptance of new functions or system performance indicators changes rapidly.

[0004] At present, there is no effective solution to the problem of poor real-time performance of gray release in the related art. SUMMARY

[0005] The main purpose of the present application is to provide a gray release method, device and storage medium for mobile applications to solve the problem of poor real-time performance of gray release in the related art.

[0006] In order to achieve the above purpose, according to one aspect of the present application, a gray release method for mobile applications is provided. The method comprises: obtaining behavior data of a first target user for a target mobile application in the case of obtaining authorization of the first target user, wherein the behavior data at least includes click rate and sharing rate; inputting the behavior data into a gray prediction model, and outputting posterior probability corresponding to the behavior data by using the gray prediction model, wherein the posterior probability is used to represent the acceptance degree of a second target user for the target mobile application; obtaining a target gray promotion ratio corresponding to the posterior probability according to a gray control model; and adjusting the gray release of the target mobile application according to the target gray promotion ratio.

[0007] Further, obtaining the behavior data of the first target user for the target mobile application comprises: obtaining operation events, response logs and system performance data of the first target user for the target mobile application, and determining initial behavior data according to the operation events, the response logs and the system performance data; filtering invalid data in the initial behavior data to obtain a filtering result; performing index aggregation on the filtering result to obtain an aggregation result; extracting the click rate and the sharing rate from the aggregation result, and determining the behavior data according to the click rate and the sharing rate.

[0008] Further, the training step of the gray-scale prediction model comprises: obtaining historical behavior sample data and a preset prior probability for the target mobile application; training a Bayesian model according to the historical behavior sample data and the preset prior probability to obtain a posterior distribution for the target mobile application; and determining the gray-scale prediction model according to the posterior distribution.

[0009] Further, the obtaining of the target gray-scale promotion ratio corresponding to the posterior probability according to the gray-scale regulation model comprises: receiving an initial gray-scale promotion ratio and a confidence coefficient, and obtaining a preset adjustment coefficient; determining a gray-scale deviation according to the posterior probability, the confidence coefficient and the preset adjustment coefficient; and determining the target gray-scale promotion ratio according to the initial gray-scale promotion ratio and the gray-scale deviation.

[0010] Further, the obtaining of the preset adjustment coefficient comprises: obtaining a version iteration tendency type of the second target user in a case where the second target user authorization is obtained; and determining the preset adjustment coefficient according to the version iteration tendency type.

[0011] Further, after the target gray-scale promotion ratio is determined according to the initial gray-scale promotion ratio and the gray-scale deviation, the method further comprises: obtaining an application crash rate, a memory occupation rate and a technical success rate of the first target user for the target mobile application; inputting the application crash rate, the memory occupation rate and the technical success rate into a risk prediction model to obtain a risk prediction result; and correcting the target gray-scale promotion ratio according to the risk prediction result to obtain a corrected gray-scale promotion ratio, wherein the corrected gray-scale promotion ratio is used to adjust the gray-scale release of the target mobile application.

[0012] In order to achieve the above-mentioned purpose, according to another aspect of the present application, a gray-scale release device for a mobile application is provided. The device comprises: a data acquisition unit configured to obtain behavior data of a first target user for a target mobile application in a case where the first target user authorization is obtained, wherein the behavior data at least comprises a click rate and a sharing rate; a gray-scale prediction unit configured to input the behavior data into a gray-scale prediction model, and output a posterior probability corresponding to the behavior data by using the gray-scale prediction model, wherein the posterior probability is used to represent an acceptance degree of a second target user for the target mobile application; a ratio determination unit configured to obtain a target gray-scale promotion ratio corresponding to the posterior probability according to a gray-scale regulation model; and a gray-scale adjustment unit configured to adjust the gray-scale release of the target mobile application according to the target gray-scale promotion ratio.

[0013] Further, the data acquisition unit comprises: an initial acquisition module, configured to acquire operation events, response logs and system performance data of the first target user for the target mobile application, and determine initial behavior data according to the operation events, the response logs and the system performance data; a data filtering module, configured to filter invalid data in the initial behavior data to obtain a filtering result; a data aggregation module, configured to aggregate indexes for the filtering result to obtain an aggregation result; and a data extraction module, configured to extract a click rate and a sharing rate in the aggregation result, and determine the behavior data according to the click rate and the sharing rate.

[0014] Further, the device further comprises: a sample acquisition unit, configured to acquire historical behavior sample data and a preset prior probability for the target mobile application; a model training unit, configured to train a Bayesian model according to the historical behavior sample data and the preset prior probability to obtain a posterior distribution for the target mobile application; and a model determination unit, configured to determine the gray-scale prediction model according to the posterior distribution.

[0015] Further, the proportion determination unit comprises: a parameter acquisition module, configured to receive an initial gray-scale promotion proportion and a confidence coefficient, and acquire a preset adjustment coefficient; a bias determination module, configured to determine a gray-scale bias according to the posterior probability, the confidence coefficient and the preset adjustment coefficient; and a promotion proportion determination module, configured to determine the target gray-scale promotion proportion according to the initial gray-scale promotion proportion and the gray-scale bias.

[0016] Further, the parameter acquisition module comprises: a type acquisition submodule, configured to acquire a version iteration tendency type of the second target user in a case where the second target user is authorized; and an adjustment coefficient determination submodule, configured to determine the preset adjustment coefficient according to the version iteration tendency type.

[0017] Further, the proportion determination unit further comprises: a performance acquisition module, configured to acquire an application crash rate, a memory occupancy rate and a technical success rate of the first target user for the target mobile application after determining the target gray-scale promotion proportion according to the initial gray-scale promotion proportion and the gray-scale bias; a risk prediction module, configured to input the application crash rate, the memory occupancy rate and the technical success rate into a risk prediction model to obtain a risk prediction result; and a proportion correction module, configured to correct the target gray-scale promotion proportion according to the risk prediction result to obtain a corrected gray-scale promotion proportion, wherein the corrected gray-scale promotion proportion is used to adjust the gray-scale release of the target mobile application.

[0018] According to another aspect of the present application, a computer readable storage medium is provided, which comprises a stored program, wherein the program controls the device where the computer readable storage medium is located to execute any one of the gray-scale release methods for a mobile application when the program is running.

[0019] According to another aspect of the present application, an electronic device is provided, comprising one or more processors, a memory, and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, and the one or more programs comprise instructions for performing any one of the methods for gray release of mobile applications.

[0020] According to another aspect of the present application, a computer program product is provided, comprising computer instructions for implementing the steps of any one of the methods for gray release of mobile applications when executed by a processor.

[0021] In the embodiments of the present application, the behavior data of the first target user for the target mobile application is obtained in the case that the first target user is authorized, wherein the behavior data at least comprises click rate and sharing rate; the behavior data is input into a gray prediction model, and the posterior probability corresponding to the behavior data is output by using the gray prediction model, wherein the posterior probability is used to represent the acceptance degree of the second target user for the target mobile application; the target gray promotion proportion corresponding to the posterior probability is obtained according to a gray regulation model; and the gray release of the target mobile application is adjusted according to the target gray promotion proportion, thereby solving the technical problem of poor real-time performance of the gray release in the prior art.

[0022] By obtaining the behavior data of the first target user and predicting the posterior probability according to the gray prediction model, the target gray promotion proportion is obtained, thereby realizing the dynamic adjustment of the gray release proportion and improving the real-time performance of the gray release. BRIEF DESCRIPTION OF DRAWINGS

[0023] The accompanying drawings, which form a part of the present application, are intended to provide further understanding of the present application, and are used to interpret the illustrative embodiments of the present application and their descriptions, and do not constitute improper limitations to the present application. In the drawings:

[0024] Figure 1 Fig. 1 shows a hardware structure block diagram of a computer terminal for implementing the method for gray release of mobile applications;

[0025] Figure 2 Fig. 2 is a flowchart of the method for gray release of mobile applications according to the embodiments of the present application;

[0026] Figure 3 Fig. 3 is a schematic diagram of the device for gray release of mobile applications according to the embodiments of the present application;

[0027] Figure 4 Fig. 4 is a structure block diagram of an electronic device according to the embodiments of the present application. DETAILED DESCRIPTION

[0028] In order to better understand the technical scheme of the present application, the technical scheme in the embodiments of the present application will be clearly and completely described in the following with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, but not all the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work should belong to the scope of protection of the present application.

[0029] It should be noted that the terms "first", "second" and the like in the description and claims of the present application and the above drawings are used to distinguish similar objects, and do not necessarily indicate a specific order or a chronological sequence. It should be understood that the data thus used can be interchanged under appropriate circumstances, so that the embodiments of the present application described herein can be implemented in an order other than that illustrated or described herein. In addition, the terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusion, for example, a process, method, system, product or device including a series of steps or units does not have to be limited to those steps or units clearly listed, but can include other steps or units not clearly listed or inherent to these processes, methods, products or devices.

[0030] It should be noted that the collected information (including but not limited to user equipment information, user personal information, etc.) and data (including but not limited to data for display, analyzed data, etc.) involved in the present application are information and data authorized by the user or authorized by all parties, and the collection, storage, use, processing, transmission, provision, disclosure and application of related data comply with relevant laws, regulations and standards, necessary security measures are taken, do not violate public order and good customs, and provide corresponding operation portal for user to choose authorization or refusal. For example, the system and the interface between the related users or institutions provide the corresponding operation portal for the user to choose to agree or refuse the automatic decision result; if the user chooses to refuse, the expert decision process is entered.

[0031] Embodiment 1

[0032] According to the embodiments of the present application, a method embodiment for gray release of mobile application is also provided. It should be noted that the steps shown in the flowchart of the drawings can be executed in a computer system such as a set of computer executable instructions, and although the logical order is shown in the flowchart, in some cases, the steps shown or described herein can be executed in an order different from that shown herein.

[0033] The method embodiment provided by the embodiment of the present application can be executed in a mobile terminal, a computer terminal or a similar computing device. Figure 1A hardware structure block diagram of a computer terminal (or mobile device) for implementing a gray release method for mobile applications is shown. As shown in Figure 1 The computer terminal 10 (or mobile device) can include one or more processors 102 (the processor 102 can include, but not limited to, a microprocessor MCU or a programmable logic device FPGA processing device, etc.), a memory 104 for storing data, and a transmission device 106 for communication functions. In addition, it can also include a display, an input / output interface (I / O interface), a universal serial bus (USB) port (which can be included as one of the ports in the BUS bus), a network interface, a power supply and / or a camera. Those skilled in the art can understand that Figure 1 The structure shown is only schematic, which does not limit the structure of the above-mentioned electronic device. For example, the computer terminal 10 can include more or less components than those shown in Figure 1 or have a different configuration than that shown in Figure 1 .

[0034] It should be noted that the one or more processors 102 and / or other data processing circuits described above can be referred to herein as "data processing circuits" in general. The data processing circuit can be embodied in whole or in part as software, hardware, firmware or any other combination. In addition, the data processing circuit can be a single independent processing module, or any one of the other elements combined into the computer terminal 10 (or mobile device) in whole or in part. As referred to in the embodiments of the present application, the data processing circuit is a processor control (for example, selection of a variable resistance terminal path connected to an interface).

[0035] The memory 104 can be used to store software programs and modules of application software, such as program instructions / data storage devices corresponding to the gray release method for mobile applications in the embodiments of the present application. The processor 102 executes various functional applications and data processing by running the software programs and modules stored in the memory 104, that is, implements the above-mentioned gray release method for mobile applications. The memory 104 can include a high-speed random access memory, and can also include a non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some examples, the memory 104 can further include a memory remotely disposed with respect to the processor 102, which can be connected to the computer terminal 10 through a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network and a combination thereof.

[0036] The transmission device 106 is configured to receive or send data via a network. The network can include a wireless network provided by a communication provider of the computer terminal 10. In an example, the transmission device 106 includes a network interface controller (NIC) that can be connected to other network devices through a base station to communicate with the Internet. In an example, the transmission device 106 can be a radio frequency (RF) module configured to communicate with the Internet wirelessly.

[0037] The display can be a liquid crystal display (LCD) that is touch screen, for example, which can enable a user to interact with a user interface of the computer terminal 10 (or the mobile device).

[0038] In the above operating environment, the present application provides a gray release method for a mobile application, as shown in Figure 2 Figure 2 is a flowchart of the gray release method for a mobile application according to Embodiment 1 of the present application.

[0039] In step S201, behavior data of a first target user for a target mobile application is obtained in a case where the first target user authorizes.

[0040] Optionally, the target mobile application is an application that has a new version (or new function) and is targeted by the gray release, and the authorization of the first target user can be determined by showing the user a privacy policy and data collection terms and obtaining the active check of the user. The first target user is a user who has used the target mobile application. It should be noted that the behavior data at least includes a click rate and a sharing rate.

[0041] For example, the original version of an application is 1.0 version, and a 2.0 version of the application is pushed out with version iteration, and the 2.0 version of the application is the target mobile application.

[0042] In step S202, the behavior data is input into a gray prediction model, and a posterior probability corresponding to the behavior data is output by the gray prediction model.

[0043] Optionally, the second target user is a user who has not used the target mobile application (for example, a user who has not received a push of the 2.0 version of the target mobile application), the gray prediction model can be a model obtained by training based on a Bayesian model (which has a significantly faster inference speed than a deep learning model), and the gray prediction model is configured to predict a success rate of the second target user to accept the target mobile application (the acceptance can be in the form of clicking the application to use), and the posterior probability is configured to represent an acceptance degree of the second target user for the target mobile application. ​

[0044] Optionally, after outputting the posterior probability corresponding to the behavior data by using the gray prediction model, a user using the target mobile application in the second target user can be obtained as a third target user, and the gray prediction model is updated by using the third target user (when the gray prediction model is a model trained based on the Bayesian model, updating the gray prediction model means updating the posterior distribution of the gray prediction model by training), so as to realize dynamic updating of the gray prediction model. Since the gray prediction model is obtained according to the Bayesian model training, the updating of the gray prediction model can be incremental training, that is, each time new behavior data is obtained, a Bayesian update can be triggered once (for example, the posterior distribution of the last time is used as the prior probability of the current time training, and the gray prediction model is updated in combination with new behavior data to obtain a new posterior distribution, and a new gray prediction model is determined according to the new posterior distribution. For Beta conjugate distribution, the posterior distribution can be directly obtained by addition calculation, which improves the real-time performance of the updating), so as to realize real-time updating of the behavior data and further ensure the real-time performance of the gray release.

[0045] In step S203, the target gray promotion ratio corresponding to the posterior probability is obtained according to the gray regulation model.

[0046] Optionally, the gray regulation model can be a gray regulation decision formula, and the posterior probability is converted into the target gray promotion ratio by inputting the posterior probability into the gray regulation decision formula.

[0047] In step S204, the gray release of the target mobile application is adjusted according to the target gray promotion ratio.

[0048] For example, the initial gray promotion ratio of the target mobile application is a, and after obtaining the target gray promotion ratio b, the promotion ratio of the gray release of the target mobile application is adjusted from a to b.

[0049] In summary, by obtaining the behavior data of the first target user, the posterior probability is predicted according to the gray prediction model, and then the target gray promotion ratio is obtained, so as to realize dynamic adjustment of the gray release ratio and improve the real-time performance of the gray release.

[0050] In order to improve the real-time performance of the gray release, optionally, obtaining the behavior data of the first target user for the target mobile application includes: obtaining operation events, response logs and system performance data of the first target user for the target mobile application, and determining initial behavior data according to the operation events, the response logs and the system performance data; filtering invalid data in the initial behavior data to obtain a filtering result; performing index aggregation on the filtering result to obtain an aggregation result; extracting the click rate and the sharing rate in the aggregation result, and determining the behavior data according to the click rate and the sharing rate.

[0051] Optionally, the client-side tracking of the target mobile application can be performed with the authorization of the first target user, and the user operation events such as button clicks, swipes, sharing times, etc. can be automatically captured through the integrated software development kit. The server-side monitoring of the target mobile application can be performed to collect the program interface response conditions and error logs, such as failed requests, abnormal requests, error codes, and the number of times of distribution, etc. The system performance data of the target mobile application can be collected through the performance probe, such as CPU occupancy, memory usage, application (for the target mobile application) crash rate, network request (for the target mobile application) success rate, etc. Invalid data (such as null and other invalid data) can be filtered through data cleaning, and the filtered results can be standardized. The related indicators (such as click rate and sharing rate) are aggregated and calculated every 60 seconds, such as click rate equals to the number of clicks divided by the exposure, crash rate equals to the number of crashes divided by the number of starts, etc. The aggregation results are obtained, and the sharing rate and click rate are extracted as the behavior data.

[0052] In summary, by obtaining the operation events, response logs, and system performance data of the first target user for the target mobile application, and determining the initial behavior data according to the operation events, response logs, and system performance data, filtering the invalid data in the initial behavior data to obtain the filtered results, and performing indicator aggregation on the filtered results to obtain the aggregation results, the click rate and sharing rate in the aggregation results are extracted, and the behavior data is determined according to the click rate and sharing rate, which lays a data foundation for the gray release and improves the real-time performance of the gray release.

[0053] In order to improve the real-time performance of the gray release, the training step of the gray prediction model includes: obtaining historical behavior sample data and a preset prior probability for the target mobile application; training the Bayesian model according to the historical behavior sample data and the preset prior probability to obtain a posterior distribution for the target mobile application; and determining the gray prediction model according to the posterior distribution.

[0054] Optionally, the historical behavior sample data includes behavior samples and sample labels, and the sample labels can be that the user accepts the target new version application (for example, the user gives four or five stars to the target new version application) or the target does not accept the target new version application (for example, the user gives three stars or less than three stars to the target new version application). The preset prior probability can be Beta (1, 1), that is, all users are equally likely to accept. After training the Bayesian model, the posterior distribution is 0~Beta (10, 90).

[0055] Optionally, a preset acceptance rate threshold can be set for the Bayesian model, that is, the output of the gray prediction model is a probability P (0>0|D) that the acceptance possibility is greater than the preset acceptance rate threshold, D is the behavior data, 0 is the real acceptance rate, 0 is the preset acceptance rate threshold (for example, 5%), Beta distribution can be selected as the prior probability, and the prior probability can be Beta (1, 1) in the case of no information for the first time gray. According to the Bayesian theorem, the posterior probability is defined as the posterior distribution on (0, 1). The integral can also be calculated by the incomplete Beta function, and the corresponding program package is provided in the java program language and the python program language, which will not be described here.

[0056] In summary, by obtaining the historical behavior sample data and the preset prior probability for the target mobile application, the Bayesian model is trained according to the historical behavior sample data and the preset prior probability, and the posterior distribution for the target mobile application is obtained. According to the posterior distribution, the gray prediction model is determined, and the real-time performance of the gray release is improved.

[0057] In order to improve the accuracy of the gray release, optionally, the target gray promotion ratio corresponding to the posterior probability is obtained according to the gray regulation model, including: receiving an initial gray promotion ratio and a confidence coefficient, and obtaining a preset adjustment coefficient; determining a gray deviation according to the posterior probability, the confidence coefficient and the preset adjustment coefficient; determining the target gray promotion ratio according to the initial gray promotion ratio and the gray deviation.

[0058] Optionally, the gray regulation model can be a gray regulation decision formula, and the gray decision formula can be as follows:

[0059] Δ = λ × (P - (1 - V))

[0060] Wherein, Δ represents the gray deviation, λ represents the preset adjustment coefficient, the preset adjustment coefficient is used to balance the speed and stability of adjustment, for example, if the preset adjustment coefficient is higher, the adjustment range of the gray scale ratio will be larger, otherwise it will be smaller, so as to prevent excessive adjustment from bringing instability factors. P represents the posterior probability, V represents the confidence coefficient, the confidence coefficient is used to represent the reliability of the posterior probability, the smaller the confidence coefficient, the higher the confidence requirement for the posterior probability. Determining the target gray scale promotion ratio according to the initial gray scale promotion ratio and the gray deviation can be that the gray deviation is detected whether it is a positive number, in the case that the gray deviation is a positive number, the sum of the gray deviation and the initial gray scale promotion ratio is determined as the target gray scale promotion ratio; in the case that the gray deviation is negative or zero, the initial gray scale promotion ratio is directly taken as the target gray scale promotion ratio. Through the posterior probability, a quantitative basis for the acceptance degree of the target mobile application by the second target user is provided, and by setting the preset adjustment coefficient, the speed and amplitude of the gray scale ratio adjustment can be controlled, for example, in the initial gray scale release stage, in order to speed up the verification, the preset adjustment coefficient can be set to be higher to quickly expand the gray scale ratio; and once the function performance tends to be stable, the preset adjustment coefficient should be appropriately reduced to prevent excessive adjustment caused by single data fluctuation; the smaller the confidence coefficient, the higher the confidence requirement of the gray deviation for the posterior probability, that is, without sufficient reliable data support, the gray scale ratio will not be easily adjusted, thereby ensuring the stability and accuracy of the gray release.

[0061] In summary, by receiving the initial gray scale promotion ratio and the confidence coefficient, and obtaining the preset adjustment coefficient, the gray deviation is determined according to the posterior probability, the confidence coefficient and the preset adjustment coefficient, and the target gray scale promotion ratio is determined according to the initial gray scale promotion ratio and the gray deviation, thereby providing the accuracy of the gray release.

[0062] In order to improve the accuracy of the gray release, optionally, obtaining the preset adjustment coefficient comprises: obtaining the version iteration tendency type of the second target user in the case that the second target user is authorized; determining the preset adjustment coefficient according to the version iteration tendency type.

[0063] For example, the questionnaire can be issued to the second target user by using the old version application corresponding to the target mobile application, to obtain the version iteration tendency type of the second target user, and the version iteration tendency type can be divided into a first type of tendency (tending to try the new version) and a second type of tendency (tending to continue to use the more stable old version). The preset adjustment coefficient corresponding to the version iteration tendency type can be obtained according to the preset value mapping. The preset adjustment coefficient corresponding to the first type of tendency can be 0.7, and the preset adjustment coefficient corresponding to the second type of tendency can be 0.5. When adjusting the gray release to the users of the first type of tendency in the second target user, the preset adjustment coefficient of 0.7 can be used to calculate the target gray promotion ratio, and when adjusting the gray release to the users of the second type of tendency in the second target user, the preset adjustment coefficient of 0.5 can be used to calculate the target gray promotion ratio, so as to meet the individual needs of different users and make the gray release more accurate.

[0064] In summary, by obtaining the version iteration tendency type of the second target user in the case of obtaining the authorization of the second target user, and determining the preset adjustment coefficient according to the version iteration tendency type, the accuracy of the gray release is improved.

[0065] In order to improve the real-time performance of the gray release, optionally, after determining the target gray promotion ratio according to the initial gray promotion ratio and the gray deviation, the method further includes: obtaining the application crash rate, the memory occupancy rate and the technical success rate of the first target user for the target mobile application; inputting the application crash rate, the memory occupancy rate and the technical success rate into a risk prediction model to obtain a risk prediction result; correcting the target gray promotion ratio according to the risk prediction result to obtain a corrected gray promotion ratio, wherein the corrected gray promotion ratio is used to adjust the gray release of the target mobile application.

[0066] Optionally, the first target user is a user who has used the target mobile application, that is, in the case that a second target user uses the target mobile application due to gray release, the second target user is the new first target user. The application crash rate, the memory occupancy rate and the technical success rate of all users who have used the target mobile application are obtained (that is, the proportion of all users who can successfully run the target mobile application without failure). The risk prediction model can be a model constructed based on risk rules, and the risk prediction model includes a first crash rate threshold, a first memory occupancy rate threshold, a first technical success rate threshold, a second crash rate threshold, a second memory occupancy rate threshold, and a second technical success rate threshold. When the memory occupancy rate is higher than the first memory occupancy rate threshold or the application crash rate is greater than the first crash rate threshold or the technical success rate is less than the first technical success rate threshold, a first rollback instruction is triggered. When the memory occupancy rate is higher than the second memory occupancy rate threshold or the application crash rate is greater than the second crash rate threshold or the technical success rate is less than the second technical success rate threshold, a second rollback instruction is triggered. In the case that the first rollback instruction is triggered, the gray deviation in the target gray promotion proportion is adjusted to 30% of the original gray deviation, and in the case that the second rollback instruction is triggered, the gray deviation in the target gray promotion proportion is adjusted to zero, that is, the gray promotion proportion is no longer increased, so as to realize timely response to the risk situation and repair the program failure, and ensure the user interests.

[0067] In summary, by obtaining the application crash rate, the memory occupancy rate and the technical success rate of the first target user for the target mobile application; inputting the application crash rate, the memory occupancy rate and the technical success rate into the risk prediction model to obtain a risk prediction result; and modifying the target gray promotion proportion according to the risk prediction result to obtain a modified gray promotion proportion, the real-time performance of the gray release is improved.

[0068] It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described herein can be executed in an order different from that shown herein.

[0069] Embodiment 2

[0070] The embodiments of the present application also provide a gray release device for a mobile application. It should be noted that the gray release device for a mobile application of the embodiments of the present application can be used to execute the gray release method for a mobile application provided by the embodiments of the present application. The gray release device for a mobile application provided by the embodiments of the present application is introduced as follows.

[0071] According to the embodiments of the present application, a device for implementing the above-mentioned gray release method for a mobile application is also provided, such asFigure 3 The device comprises:

[0072] The data acquisition unit 301 is configured to acquire behavior data of the first target user for the target mobile application in the case where the first target user authorizes, wherein the behavior data at least includes a click rate and a sharing rate.

[0073] The gray-scale prediction unit 302 is configured to input the behavior data into a gray-scale prediction model and output a posterior probability corresponding to the behavior data by using the gray-scale prediction model, wherein the posterior probability is used to represent an acceptance degree of a second target user for the target mobile application.

[0074] The proportion determination unit 303 is configured to acquire a target gray-scale promotion proportion corresponding to the posterior probability according to a gray-scale regulation model.

[0075] The gray-scale adjustment unit 304 is configured to adjust the gray-scale release of the target mobile application according to the target gray-scale promotion proportion.

[0076] The gray-scale release device for the mobile application provided by the embodiment of the present application acquires the behavior data of the first target user for the target mobile application in the case where the first target user authorizes by using the data acquisition unit 301, wherein the behavior data at least includes a click rate and a sharing rate, inputs the behavior data into a gray-scale prediction model by using the gray-scale prediction unit 302, and outputs a posterior probability corresponding to the behavior data by using the gray-scale prediction model, wherein the posterior probability is used to represent an acceptance degree of a second target user for the target mobile application, acquires the target gray-scale promotion proportion corresponding to the posterior probability according to a gray-scale regulation model by using the proportion determination unit 303, and adjusts the gray-scale release of the target mobile application according to the target gray-scale promotion proportion by using the gray-scale adjustment unit 304, thereby solving the problem of poor real-time performance of the gray-scale release in the related art, and further achieving the effect of improving the real-time performance of the gray-scale release.

[0077] Optionally, in the gray-scale release device for the mobile application provided by the embodiment of the present application, the data acquisition unit 301 comprises: an initial acquisition module configured to acquire operation events, response logs and system performance data of the first target user for the target mobile application, and determine initial behavior data according to the operation events, the response logs and the system performance data; a data filtering module configured to filter invalid data in the initial behavior data to obtain a filtering result; a data aggregation module configured to perform index aggregation on the filtering result to obtain an aggregation result; and a data extraction module configured to extract a click rate and a sharing rate in the aggregation result, and determine the behavior data according to the click rate and the sharing rate.

[0078] Optionally, in the gray release device for mobile applications provided in the embodiments of the present application, the device further comprises: a sample acquisition unit, configured to acquire historical behavior sample data and a preset prior probability for a target mobile application; a model training unit, configured to train a Bayesian model according to the historical behavior sample data and the preset prior probability to obtain a posterior distribution for the target mobile application; and a model determination unit, configured to determine a gray prediction model according to the posterior distribution.

[0079] Optionally, in the gray release device for mobile applications provided in the embodiments of the present application, the proportion determination unit 303 comprises: a parameter acquisition module, configured to receive an initial gray promotion proportion and a confidence coefficient, and acquire a preset adjustment coefficient; a deviation determination module, configured to determine a gray deviation according to the posterior probability, the confidence coefficient and the preset adjustment coefficient; and a promotion proportion determination module, configured to determine a target gray promotion proportion according to the initial gray promotion proportion and the gray deviation.

[0080] Optionally, in the gray release device for mobile applications provided in the embodiments of the present application, the parameter acquisition module comprises: a type acquisition submodule, configured to acquire a version iteration tendency type of a second target user in a case where the second target user is authorized; and an adjustment coefficient determination submodule, configured to determine the preset adjustment coefficient according to the version iteration tendency type.

[0081] Optionally, in the gray release device for mobile applications provided in the embodiments of the present application, the proportion determination unit 303 further comprises: a performance acquisition module, configured to acquire an application crash rate, a memory occupancy rate and a technical success rate of the first target user for the target mobile application after determining the target gray promotion proportion according to the initial gray promotion proportion and the gray deviation; a risk prediction module, configured to input the application crash rate, the memory occupancy rate and the technical success rate into a risk prediction model to obtain a risk prediction result; and a proportion correction module, configured to correct the target gray promotion proportion according to the risk prediction result to obtain a corrected gray promotion proportion, wherein the corrected gray promotion proportion is used to adjust the gray release of the target mobile application.

[0082] It should be noted that the data acquisition unit 301, the gray prediction unit 302, the proportion determination unit 303 and the gray adjustment unit 304 correspond to steps S201 to S204 in Embodiment 1, and the units have the same instances and application scenarios as the corresponding steps, but are not limited to the content disclosed in Embodiment 1. It should be noted that the above modules or units can be hardware components or software components stored in a memory (for example, the memory 104) and processed by one or more processors (for example, the processors 102a, 102b, …, 102n), and the above modules can also be run in the computer terminal 10 provided in Embodiment 1 as a part of the device.

[0083] Embodiment 3

[0084] Embodiments of the present application can provide an electronic device, Figure 4 is a structural block diagram of an electronic device according to an embodiment of the present application. As shown in the figure, the electronic device can include one or more (only one is shown in the figure) processors 1002, a memory 1004, a storage controller, and a peripheral interface, wherein the peripheral interface is connected with a radio frequency module, an audio module, and a display. Figure 4 Figure 4 The memory can be used to store software programs and modules, such as program instructions / modules corresponding to the methods and devices in the embodiments of the present application. The processor executes various functions and data processing by running the software programs and modules stored in the memory, that is, implements the above-mentioned methods. The memory can include a high-speed random access memory, and can further include a non-volatile memory, such as one or more magnetic storage devices, flash memories, or other non-volatile solid-state memories. In some examples, the memory can further include a memory remotely arranged with respect to the processor, which can be connected to the terminal through a network. Examples of the above-mentioned network include but are not limited to the Internet, an intranet, a local area network, a mobile communication network, and a combination thereof.

[0085] The processor can call information and application programs stored in the memory through the transmission device to perform the following steps: obtaining the behavior data of the first target user for the target mobile application in the case of obtaining the authorization of the first target user, wherein the behavior data at least includes click rate and sharing rate; inputting the behavior data into the gray prediction model, and outputting the posterior probability corresponding to the behavior data by using the gray prediction model, wherein the posterior probability is used to represent the acceptance degree of the second target user for the target mobile application; obtaining the target gray promotion proportion corresponding to the posterior probability according to the gray regulation model; and adjusting the gray release of the target mobile application according to the target gray promotion proportion.

[0086] The processor can also call information and application programs stored in the memory through the transmission device to perform the following steps: obtaining the operation event, response log and system performance data of the first target user for the target mobile application, and determining the initial behavior data according to the operation event, response log and system performance data; filtering invalid data in the initial behavior data to obtain a filtering result; performing index aggregation on the filtering result to obtain an aggregation result; extracting the click rate and the sharing rate in the aggregation result, and determining the behavior data according to the click rate and the sharing rate.

[0087] The processor can also call information and application programs stored in the memory through the transmission device to perform the following steps: obtaining the operation event, response log and system performance data of the first target user for the target mobile application, and determining the initial behavior data according to the operation event, response log and system performance data; filtering invalid data in the initial behavior data to obtain a filtering result; performing index aggregation on the filtering result to obtain an aggregation result; extracting the click rate and the sharing rate in the aggregation result, and determining the behavior data according to the click rate and the sharing rate.

[0088] ​The processor can further call information and application programs stored in the memory through the transmission device to perform the following steps: obtaining historical behavior sample data and preset prior probability for the target mobile application; training a Bayesian model according to the historical behavior sample data and the preset prior probability to obtain a posterior distribution for the target mobile application; and determining a gray prediction model according to the posterior distribution.

[0089] The processor can further call information and application programs stored in the memory through the transmission device to perform the following steps: receiving an initial gray promotion ratio and a confidence coefficient, and obtaining a preset adjustment coefficient; determining a gray deviation according to the posterior probability, the confidence coefficient and the preset adjustment coefficient; and determining a target gray promotion ratio according to the initial gray promotion ratio and the gray deviation.

[0090] The processor can further call information and application programs stored in the memory through the transmission device to perform the following steps: obtaining a version iteration tendency type of the second target user in the case of obtaining authorization of the second target user; and determining a preset adjustment coefficient according to the version iteration tendency type.

[0091] The processor can further call information and application programs stored in the memory through the transmission device to perform the following steps: obtaining an application crash rate, a memory occupation rate and a technical success rate of the first target user for the target mobile application; inputting the application crash rate, the memory occupation rate and the technical success rate into a risk prediction model to obtain a risk prediction result; and correcting the target gray promotion ratio according to the risk prediction result to obtain a corrected gray promotion ratio, wherein the corrected gray promotion ratio is used to adjust the gray release of the target mobile application.

[0092] By obtaining behavior data of the first target user for the target mobile application in the case of obtaining authorization of the first target user, wherein the behavior data at least includes a click rate and a sharing rate; inputting the behavior data into a gray prediction model, and outputting a posterior probability corresponding to the behavior data by using the gray prediction model, wherein the posterior probability is used to represent an acceptance degree of a second target user for the target mobile application; obtaining a target gray promotion ratio corresponding to the posterior probability according to a gray regulation model; and adjusting the gray release of the target mobile application according to the target gray promotion ratio, the technical problem of poor real-time performance of the gray release in the prior art is solved.

[0093] Those skilled in the art can understand that Figure 4 The structure shown is only schematic, and the electronic device can also be a terminal device such as a smart phone, a tablet computer, a palm computer, a Mobile Internet Device (MID), a PAD, etc. Figure 4It does not cause limitation to the structure of the electronic device. For example, the electronic device can further include more or less components (such as a network interface, a display device, etc.) or have a different configuration from that shown in FIG. 1. Figure 4 Figure 4

[0094] Those skilled in the art can understand that all or part of the steps in the various methods of the above embodiments can be instructed by programs to the related hardware of the terminal device, and the programs can be stored in a computer readable storage medium, which can include a flash disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, etc.

[0095] Embodiment 4

[0096] The embodiments of the present application also provide a storage medium. Optionally, in the embodiments, the storage medium can be used to store the program code executed by the gray release method for mobile applications provided in Embodiment 1.

[0097] Optionally, in the embodiments, the storage medium can be located in any one of the computer terminals in the computer terminal group in the computer network, or in any one of the mobile terminals in the mobile terminal group.

[0098] Optionally, in the embodiments, the computer readable storage medium is configured to store program code for performing the following steps: obtaining the behavior data of the first target user for the target mobile application in the case of obtaining the authorization of the first target user, wherein the behavior data at least includes a click rate and a sharing rate; inputting the behavior data into the gray prediction model, and outputting the posterior probability corresponding to the behavior data by using the gray prediction model, wherein the posterior probability is used to represent the acceptance degree of the second target user for the target mobile application; obtaining the target gray promotion proportion corresponding to the posterior probability according to the gray regulation model; and adjusting the gray release of the target mobile application according to the target gray promotion proportion.

[0099] Optionally, in the embodiments, the computer readable storage medium is further configured to store program code for performing the following steps: obtaining the operation event, the response log and the system performance data of the first target user for the target mobile application, and determining the initial behavior data according to the operation event, the response log and the system performance data; filtering the invalid data in the initial behavior data to obtain a filtering result; performing index aggregation on the filtering result to obtain an aggregation result; extracting the click rate and the sharing rate in the aggregation result, and determining the behavior data according to the click rate and the sharing rate.

[0100] ​​Optionally, in the embodiment, the computer readable storage medium is further configured to store program code for performing the following steps: obtaining historical behavior sample data and preset prior probability for the target mobile application; training the Bayesian model according to the historical behavior sample data and the preset prior probability to obtain a posterior distribution for the target mobile application; and determining the gray prediction model according to the posterior distribution.

[0101] Optionally, in the embodiment, the computer readable storage medium is further configured to store program code for performing the following steps: receiving an initial gray promotion ratio and a confidence coefficient, and obtaining a preset adjustment coefficient; determining a gray deviation according to the posterior probability, the confidence coefficient and the preset adjustment coefficient; and determining the target gray promotion ratio according to the initial gray promotion ratio and the gray deviation.

[0102] Optionally, in the embodiment, the computer readable storage medium is further configured to store program code for performing the following steps: obtaining a version iteration tendency type of the second target user in a case where the second target user is authorized; and determining the preset adjustment coefficient according to the version iteration tendency type.

[0103] Optionally, in the embodiment, the computer readable storage medium is further configured to store program code for performing the following steps: obtaining an application crash rate, a memory occupation rate and a technical success rate of the first target user for the target mobile application; inputting the application crash rate, the memory occupation rate and the technical success rate into the risk prediction model to obtain a risk prediction result; and correcting the target gray promotion ratio according to the risk prediction result to obtain a corrected gray promotion ratio, wherein the corrected gray promotion ratio is used to adjust the gray release of the target mobile application.

[0104] The application also provides a computer program product adapted to execute the steps of the gray release method for a mobile application when executed on a data processing device.

[0105] The above-mentioned sequence numbers of the embodiments of the application only for description, not represent the pros and cons of the embodiments.

[0106] In the above-described embodiments of the application, the description of each embodiment has its own emphasis, and the parts not described in detail in a certain embodiment can be referred to the relevant description of other embodiments.

[0107] In several embodiments provided in the present application, it should be understood that the disclosed technology can be implemented by other ways. Among them, the above-described device embodiments are only schematic, for example, the division of the units is only a logical function division, and actual implementation can have another division manner, for example, a plurality of units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the coupling or direct coupling or communication connection between the displayed or discussed units can be indirect coupling or communication connection through some interfaces, units or modules, and can be electrical or other forms.

[0108] The units described as separate components can or can not be physically separate, and the components shown as units can or can not be physical units, i.e., they can be located in one place or distributed on a plurality of network units. Part or all of the units can be selected according to actual needs to achieve the purpose of the embodiment.

[0109] In addition, the functional units in each embodiment of the present application can be integrated in one processing unit, or each unit can be physically present separately, or two or more units can be integrated in one unit. The integrated unit can be realized in the form of hardware or in the form of a software functional unit.

[0110] The integrated unit, if realized in the form of a software functional unit and sold or used as an independent product, can be stored in a computer readable storage medium. Based on this understanding, the technical solutions of the present application essentially or the part that contributes to the prior art or the whole or part of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a storage medium, including a plurality of instructions for causing a computer device (which can be a personal computer, a server or a network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present application. The foregoing storage medium includes: a U disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a mobile hard disk, a magnetic disk or an optical disk, and various media that can store program codes.

[0111] The above is only the preferred embodiment of the present application, and it should be pointed out that for ordinary skilled in the art, without departing from the principles of the present application, a number of improvements and refinements can be made, and these improvements and refinements should be considered as the protection scope of the present application.

Claims

1. A gray release method for mobile applications, characterized in that, The method comprises the steps of: In the case of obtaining the authorization of the first target user, obtaining the behavior data of the first target user for the target mobile application, wherein the behavior data at least includes click rate, sharing rate; Input the behavior data into the gray prediction model, and output the posterior probability corresponding to the behavior data by using the gray prediction model, wherein the posterior probability is used to represent the acceptance degree of the second target user for the target mobile application; According to the gray control model, the target gray promotion ratio corresponding to the posterior probability is obtained; According to the target gray promotion ratio, the gray release of the target mobile application is adjusted.

2. The method of claim 1, wherein, The method comprises the steps of: Obtaining the operation event, response log and system performance data of the first target user for the target mobile application, and determining the initial behavior data according to the operation event, the response log and the system performance data; Filtering invalid data in the initial behavior data to obtain a filtering result; Index aggregation is performed on the filtering result to obtain an aggregation result; Extract the click rate and sharing rate in the aggregation result, and determine the behavior data according to the click rate and the sharing rate.

3. The method of claim 1, wherein, The training steps of the gray prediction model comprise: Obtaining historical behavior sample data and preset prior probability for the target mobile application; According to the historical behavior sample data and the preset prior probability, the Bayesian model is trained to obtain the posterior distribution for the target mobile application; According to the posterior distribution, the gray prediction model is determined.

4. The method of claim 1, wherein, According to the gray control model, the target gray promotion ratio corresponding to the posterior probability is obtained, which comprises the steps of: Receiving initial gray promotion ratio and confidence coefficient, and obtaining preset adjustment coefficient; According to the posterior probability, the confidence coefficient and the preset adjustment coefficient, the gray deviation is determined; According to the initial gray promotion ratio and the gray deviation, the target gray promotion ratio is determined.

5. The method of claim 4, wherein, Obtaining the preset adjustment coefficient comprises the steps of: In the case of obtaining the authorization of the second target user, obtaining the version iteration tendency type of the second target user; According to the version iteration tendency type, the preset adjustment coefficient is determined.

6. The method of claim 4, wherein, After determining the target gray promotion ratio according to the initial gray promotion ratio and the gray deviation, the method further comprises the steps of: Obtaining the application crash rate, memory occupation rate and technical success rate of the first target user for the target mobile application; Input the application crash rate, the memory occupation rate and the technical success rate into the risk prediction model to obtain the risk prediction result; According to the risk prediction result, the target gray promotion ratio is modified to obtain the modified gray promotion ratio, wherein the modified gray promotion ratio is used to adjust the gray release of the target mobile application.

7. A grayscale deployment device for mobile applications, characterized in that, The method comprises the steps of: A data acquisition unit is configured to obtain the behavior data of the first target user for the target mobile application in the case of obtaining the authorization of the first target user, wherein the behavior data at least includes click rate, sharing rate; The gray scale prediction unit is configured to input the behavior data into a gray scale prediction model, and output a posterior probability corresponding to the behavior data by using the gray scale prediction model, where the posterior probability is used to represent an acceptance degree of a second target user to the target mobile application. The proportion determination unit is configured to obtain a target gray scale promotion proportion corresponding to the posterior probability according to a gray scale regulation model. The gray scale adjustment unit is configured to adjust a gray scale release of the target mobile application according to the target gray scale promotion proportion.

8. A computer-readable storage medium, characterized in that, The computer readable storage medium comprises a stored executable program, wherein the executable program controls a device where the computer readable storage medium is located to execute the gray scale release method for a mobile application in any one of claims 1 to 6 when the executable program is running.

9. An electronic device, comprising: The computer readable storage medium comprises: a memory storing an executable program; a processor configured to run the program, wherein the program is executed to perform the gray scale release method for a mobile application in any one of claims 1 to 6 when the program is running.

10. A computer program product comprising computer instructions, characterized in that, The computer instructions are executed by the processor to implement the steps of the gray scale release method for a mobile application in any one of claims 1 to 6.

Images