APP traffic prediction method, device, computer equipment and storage medium

Abstract

The present invention relates to the field of data analysis, and in particular to an APP traffic prediction method, device, computer equipment and storage medium. Obtain a preset historical time period corresponding to a target time period; the target time period includes at least one prediction time period; the preset historical time period includes at least two unit time periods; obtain historical traffic data within the preset historical time period from a specified APP; determine the unit traffic data of the specified APP within the unit time period based on the historical traffic data; use the unit time period as an independent variable and the unit traffic data as a dependent variable, input them into a first preset mathematical model, determine the unit empirical coefficient between the independent variable and the dependent variable through the first preset mathematical model; determine the target traffic within the target time period based on the unit empirical coefficient and the unit traffic data. This method obtains accurate predicted target traffic data through the unit empirical coefficient, which can improve the accuracy of traffic prediction, improve the operation status of the APP, and improve user stickiness and conversion rate.

Description

Technical Field

[0001] The present invention relates to the field of data analysis, and in particular to an APP traffic prediction method, device, computer equipment and storage medium. Background Art

[0002] Usually, after an APP (Application) is launched, operators need to monitor and analyze APP indicator data related to traffic. Among them, active indicators, year-on-year and month-on-month indicators, etc. are key indicators to measure the operation status of the APP after it is launched. By analyzing the key indicator data after the APP is launched, the operation status of the APP can be well monitored, and reasonable operation goals and operation plans can be set according to the current operation status.

[0003] In the existing technology, the operation goals are usually set based on statistical data and the experience of operators, which is less flexible. When the key indicator data of the APP changes greatly, it is difficult to accurately predict the traffic data of the APP, resulting in the inability to accurately predict the operation goals, which in turn leads to poor APP operation and reduced user stickiness and conversion rate. Summary of the invention

[0004] Based on this, it is necessary to provide an APP traffic prediction method, device, computer equipment and storage medium to address the above technical problems, so as to solve the problem that the existing technology cannot accurately predict APP traffic.

[0005] An APP traffic prediction method, comprising:

[0006] Acquire a preset historical time period corresponding to a target time period; the target time period includes at least one predicted time period; the preset historical time period includes at least two unit time periods;

[0007] Obtain historical traffic data within the preset historical time period from the specified APP;

[0008] Determine, according to the historical traffic data, the unit traffic data corresponding to the designated APP in each of the unit time periods;

[0009] The unit time period is used as an independent variable and the corresponding unit flow data is used as a dependent variable, and the results are input into a first preset mathematical model, and a unit empirical coefficient between the independent variable and the dependent variable is determined by the first preset mathematical model;

[0010] The target flow rate within the target time period is determined according to the unit empirical coefficient and the unit flow rate data.

[0011] An APP traffic prediction device, comprising:

[0012] A historical time period acquisition module, used to acquire a preset historical time period corresponding to a target time period; the target time period includes at least one predicted time period; the preset historical time period includes at least two unit time periods;

[0013] A historical traffic data acquisition module is used to acquire the historical traffic data within the preset historical time period from a specified APP;

[0014] A unit flow data module, used to determine the unit flow data corresponding to the specified APP in each unit time period according to the historical flow data;

[0015] A unit empirical coefficient module, used to input the unit time period as an independent variable and the corresponding unit flow data as a dependent variable into a first preset mathematical model, and determine the unit empirical coefficient between the independent variable and the dependent variable through the first preset mathematical model;

[0016] The target flow module is used to determine the target flow within the target time period according to the unit experience coefficient and the unit flow data.

[0017] A computer device includes a memory, a processor, and computer-readable instructions stored in the memory and executable on the processor. When the processor executes the computer-readable instructions, the above-mentioned APP traffic prediction method is implemented.

[0018] One or more readable storage media storing computer-readable instructions, which, when executed by one or more processors, enable the one or more processors to execute the APP traffic prediction method as described above.

[0019] The above-mentioned APP traffic prediction method, device, computer equipment and storage medium obtain a preset historical time period corresponding to the target time period; the target time period includes at least one prediction time cycle; the preset historical time period includes at least two unit time periods; the historical traffic data within the preset historical time period is obtained from the specified APP; according to the historical traffic data, the unit traffic data corresponding to the specified APP in each unit time period is determined; the unit time period is used as an independent variable, and the corresponding unit traffic data is used as a dependent variable, and input into the first preset mathematical model, and the unit experience coefficient between the independent variable and the dependent variable is determined by the first preset mathematical model; the target traffic within the target time period is determined according to the unit experience coefficient and the unit traffic data. This method determines the unit experience coefficient between the independent variable and the dependent variable through the first preset mathematical model; determines the target traffic within the target time period according to the unit experience coefficient and the unit traffic data. Accurately predicting the target traffic data according to the unit experience coefficient can improve the accuracy of traffic prediction, improve the operation status of the APP, and improve user stickiness and conversion rate. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments of the present invention will be briefly introduced below. Obviously, the accompanying drawings in the following description are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative labor.

[0021] Figure 1 This is a schematic diagram of an application environment of an APP traffic prediction method in one embodiment of the present invention;

[0022] Figure 2 This is a flow chart of an APP traffic prediction method according to an embodiment of the present invention;

[0023] Figure 3 It is a structural schematic diagram of an APP traffic prediction device in one embodiment of the present invention;

[0024] Figure 4 is a schematic diagram of a computer device in one embodiment of the present invention. DETAILED DESCRIPTION

[0025] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0026] The APP traffic prediction method provided in this embodiment can be applied to Figure 1 In an application environment, a client communicates with a server. The client includes but is not limited to various personal computers, laptops, smart phones, tablet computers, and portable wearable devices. The server can be implemented with an independent server or a server cluster consisting of multiple servers.

[0027] In one embodiment, if Figure 2 As shown, a method for predicting APP traffic is provided, and the method is applied in Figure 1 The server in the example is used as an example to illustrate the following steps:

[0028] S10. Acquire a preset historical time period corresponding to a target time period; the target time period includes at least one prediction time period; and the preset historical time period includes at least two unit time periods.

[0029] It is understandable that the target time period refers to the time period for which the customer needs to predict the traffic in the application software of the customer terminal, which includes but is not limited to one year, one quarter or one month. For example, if the customer needs to predict the traffic of a specified APP for one quarter (July to September), the target time period can be set to July 1 to September 30. Among them, the prediction time period refers to a time period in the target time period. For example, if the target time period is one quarter, the prediction time period can be one month, and one target time period contains three prediction time periods. The preset historical time period can refer to a time period adjacent to the target time period.

[0030] The preset historical time period can be set according to the actual situation, including at least two unit time periods. For example, if the target time period is from July 1 to July 31, the preset historical time period can be set to from April 1 to June 30. Among them, the unit time period refers to a time period in the preset historical time period. For example, if the preset historical time period is three months, the unit time period can be one month, and one preset historical time period includes three unit time periods. The unit time period and the prediction time period are the same in time length.

[0031] S20. Obtain historical traffic data within the preset historical time period from a specified APP.

[0032] It is understandable that the designated APP refers to the APP that the customer needs to predict the traffic on the customer terminal. Among them, traffic refers to the number of times or number of users access the server (designated APP). Historical traffic data refers to the traffic data of the designated APP in the past period of time.

[0033] Specifically, if the traffic database of the specified APP stores monitored historical traffic data, the historical traffic data within the preset historical time period is directly obtained from the traffic database of the specified APP. If the traffic database of the specified APP does not store monitored historical traffic data, the historical traffic data of the specified APP within the preset historical time period is obtained through a pre-set monitoring tool.

[0034] S30. Determine, based on the historical traffic data, the unit traffic data corresponding to the designated APP in each of the unit time periods.

[0035] It is understandable that historical traffic data includes all traffic data of a specified APP within a preset historical time period. Among them, all traffic data includes traffic data of new users, old users, and all users in different dimensions. Different dimensions include but are not limited to traffic data of online activities within a specified APP, traffic data of products, and traffic data of channels.

[0036] The unit time period is a time period in the preset historical time period. The unit traffic data is the traffic data of the specified APP in the unit time period.

[0037] According to the historical traffic data, the unit traffic data of the specified APP in the unit time period can be determined. Preferably, after obtaining the historical traffic data, the historical traffic data can be cleaned by a preset cleaning method to clean out junk data such as duplicate data, test data and abnormal data, and the data obtained after cleaning is determined as the cleaned data. Then, based on the cleaned data, the unit traffic data of the specified APP in each unit time period is calculated. Preferably, after obtaining the cleaned data, several pre-set indicator classification rules can be obtained, and the cleaned data can be classified according to several indicator classification rules to obtain several sets of indicator cleaned data, and each indicator cleaned data can be counted according to the unit time period to obtain the unit traffic data corresponding to each indicator.

[0038] S40, taking the unit time period as an independent variable and the corresponding unit flow data as a dependent variable, inputting them into a first preset mathematical model, and determining the unit empirical coefficient between the independent variable and the dependent variable through the first preset mathematical model.

[0039] It can be understood that an independent variable refers to a variable that can affect other variables, and a dependent variable refers to a variable that changes with the change of the independent variable. For example: Y=f(X), this formula is expressed as: Y changes with the change of X, then Y is the dependent variable, and X is the independent variable. The first preset mathematical model includes a regression algorithm, which is used to determine the unit experience coefficient. Among them, the regression algorithm is a statistical analysis method that performs regression analysis on the unit time period and the unit flow data to obtain the variable relationship between the unit time period and the unit flow data. The regression algorithm includes a linear regression algorithm and a nonlinear regression algorithm. The unit experience coefficient refers to a numerical value that characterizes the variable relationship between the unit time period and the unit flow data.

[0040] Preferably, the unit time period is used as an independent variable and the unit flow data is used as a dependent variable. The unit flow data and the unit time period are subjected to regression analysis using a linear regression algorithm to determine the variable relationship between the unit flow data and the unit time period, that is, to obtain the unit empirical coefficient between the independent variable and the dependent variable. One unit time period corresponds to one unit empirical coefficient.

[0041] S50: Determine the target flow rate within the target time period according to the unit empirical coefficient and the unit flow rate data.

[0042] It can be understood that the target flow refers to the flow data of each prediction period of the specified APP in the target time period obtained by prediction. Specifically, after obtaining the unit experience coefficient, the linear relationship between the unit flow data and the unit time period can be obtained. According to the linear relationship, the flow data of each prediction period of the specified APP in the target time period can be predicted.

[0043] Optionally, after determining the target traffic, the target traffic can be adjusted and updated in conjunction with the traffic data generated in real time to make the prediction of the target traffic more realistic and improve the accuracy of APP traffic prediction.

[0044] In steps S10-S50, by obtaining a preset historical time period corresponding to the target time period; the target time period includes at least one predicted time period; the preset historical time period includes at least two unit time periods; the historical traffic data within the preset historical time period is obtained from the specified APP; according to the historical traffic data, the unit traffic data corresponding to the specified APP in each unit time period is determined; the unit time period is used as an independent variable, and the corresponding unit traffic data is used as a dependent variable, and input into the first preset mathematical model, and the unit experience coefficient between the independent variable and the dependent variable is determined by the first preset mathematical model; the target traffic within the target time period is determined according to the unit experience coefficient and the unit traffic data. This method determines the unit experience coefficient between the independent variable and the dependent variable through the first preset mathematical model; determines the target traffic within the target time period according to the unit experience coefficient and the unit traffic data. This method can accurately predict the target traffic data, and can be adjusted in a linked manner according to the predicted target traffic data, which greatly improves the operation status of the APP and improves user stickiness and conversion rate.

[0045] Optionally, after step S50, that is, after determining the target flow rate in the target time period according to the unit empirical coefficient and the unit flow rate data, the following steps are included:

[0046] S501, obtaining the real-time traffic within the target time period;

[0047] S502, comparing the real-time flow rate with the target flow rate to obtain a comparison result;

[0048] S503, determining the abnormality level of the real-time traffic in the comparison result according to a preset abnormality level rule;

[0049] S504: Generate first traffic warning information of the designated APP according to the abnormality level.

[0050] It is understandable that after predicting the traffic data of the target time period, the real-time traffic generated can be compared with the predicted target traffic to verify the accuracy of the traffic prediction. At the same time, when the real-time traffic is detected to be different from the predicted target traffic, an early warning information can be generated so that the operator can adjust the operation data of the APP in time according to the early warning information.

[0051] Specifically, real-time traffic refers to the traffic data actually generated by the specified APP during the target time period. The real-time traffic data of the specified APP is obtained, and the real-time traffic data is compared with the predicted target traffic to obtain the degree of deviation between the real-time traffic data and the predicted target traffic, that is, the comparison result. The preset abnormal level rule refers to the pre-set rules for determining the abnormal level of real-time traffic data. The first traffic warning information is a pre-set processing plan for each abnormal level, and the specific processing plan can be set according to the severity of the abnormal problem. Different first traffic warning information is generated according to different abnormal levels.

[0052] Preferably, the abnormality level rules can be set to level one, level two, level three and level four. The higher the abnormality level, the more serious the abnormality. Among them, level one means that the specified APP is in normal operation, and the deviation value of the traffic data belongs to the normal network fluctuation phenomenon. The first traffic warning information fed back to the customer terminal can be "continue to monitor". Level four means that the data of the specified APP has a large deviation, and the cause of the data change needs to be investigated. For example, it may be due to a large data change caused by a network failure, system failure or traffic promotion activities, resulting in an abnormal problem. The first traffic warning information fed back to the customer terminal may be "There are serious abnormalities in the APP traffic, please investigate the problem in time."

[0053] Optionally, in step S502, the real-time traffic is compared with the target traffic to obtain a comparison result, and the following further includes:

[0054] S5021. Obtain a flow warning threshold that matches the target flow;

[0055] S5022. If the real-time traffic is greater than the traffic warning threshold, obtaining an access log corresponding to the real-time traffic from the designated APP; the access log includes an access request and a response time of the access request;

[0056] S5023, determining whether the response time is greater than a preset response threshold, and obtaining a determination result;

[0057] S5024. Generate second traffic warning information of the designated APP according to the judgment result.

[0058] It can be understood that the traffic warning threshold is a pre-set warning threshold. The warning threshold is a warning threshold determined by a statistical method based on the predicted target traffic. The warning threshold determined by the target traffic makes the warning threshold more reliable and accurate. Among them, the statistical method can be Poisson distribution, which is a discrete probability distribution used to describe traffic data in a unit time. If it is detected that the real-time traffic data is greater than the alarm threshold, the access log corresponding to the real-time traffic is obtained from the specified APP, and the access log includes the access request and the response time of the access request. By judging whether the response time is greater than the preset response threshold, it is confirmed whether the specified APP can still provide normal access services when the real-time traffic data exceeds the alarm threshold. Among them, the preset response threshold is a pre-set response threshold used to judge whether the response time of the access request is within the normal range. If the response time of the access request is greater than the preset response threshold, a second traffic warning information is sent to indicate that the specified APP has an abnormal operation problem at this time and needs manual maintenance. Among them, the second traffic warning information includes the response time of the access request and the real-time traffic. This method can timely improve the operation status of the specified APP and improve user stickiness and conversion rate.

[0059] Optionally, if the response time of the access request is less than or equal to the preset response threshold, the second traffic warning information will not be sent, indicating that the sudden increase in traffic data may be due to the promotion activities of the designated APP. The designated APP can still operate normally, reducing the false alarm rate of the warning information.

[0060] Optionally, the prediction time period includes at least two time nodes; after S50, that is, after determining the target flow in the target time period according to the unit experience coefficient and the unit flow data, it also includes:

[0061] S504: Determine the unit average flow rate within the preset historical time period according to the historical flow data.

[0062] It is understandable that the unit average flow refers to the flow data obtained by averaging all unit flow data. Specifically, based on the historical flow data, the unit flow data in each unit time period in the preset historical time period can be obtained, and the unit average flow can be obtained by averaging all unit flow data.

[0063] S505: Obtain pre-node traffic information before a specified time node in the prediction time period.

[0064] It is understandable that the prediction time period includes at least two time nodes. For example, the prediction time period is one week, and the time node can be one day, that is, the prediction time period includes 7 time nodes. The specified time node is any time node in the prediction time period. The pre-node traffic information refers to all traffic information before the specified time node, including cumulative traffic data and node traffic data. Among them, the cumulative traffic data includes the cumulative traffic at each time node before the specified time node. The node traffic data includes the traffic at each time node before the specified time node.

[0065] S506. Input the unit average flow, the pre-node flow information and the target flow into a second preset data model, so that the second preset data model performs node flow splitting processing on the target flow to obtain the node target flow of the specified time node.

[0066] It is understandable that the second preset data model includes several preset rules for determining the target flow at a specified time node. Node flow splitting processing refers to the process of splitting the target flow into several parts according to the time node to obtain the node target flow corresponding to each time node. For example, the target flow is the predicted flow in July. The target flow is split in days, and the target flow is split into 31 parts to obtain the predicted flow for each day in July. The node target flow refers to the predicted flow at each time node obtained by splitting the target flow according to the time node.

[0067] Specifically, after the unit average flow, pre-node flow information and target flow are input into the second preset data model, if at least one time node is included before the specified time node, the newly added flow data of each time node before the specified time node in the prediction time period is determined according to the pre-node flow information through the first preset rule. According to the newly added flow data and the node flow data, the flow increase rate at the time node is determined according to the second preset rule. The flow weight of the target flow on the unit average flow is determined according to the third preset rule. According to the accumulated flow data and the target flow, the remaining target flow in the prediction time period is determined according to the fourth preset rule. The remaining flow proportion of the remaining target flow to the node target flow is determined according to the fifth preset rule. The target flow is subjected to node flow splitting processing according to the remaining target flow, the remaining flow proportion, the flow increase rate and the flow weight, and the predicted flow at the specified time node, that is, the node target flow, can be obtained.

[0068] In steps S504-S506, the unit average flow in the preset historical time period is determined based on the historical flow data; the pre-node flow information before the specified time node in the prediction time period is obtained; the unit average flow, the pre-node flow information and the target flow are input into the second preset data model, so that the second preset data model performs node flow splitting processing on the target flow to obtain the node target flow at the specified time node. After obtaining the target flow, the target flow can be further split into node flows based on the analysis of the historical flow data to improve the accuracy and flexibility of flow prediction.

[0069] Optionally, the pre-node flow information includes cumulative flow data and node flow data; in step S506, the unit average flow, the pre-node flow information and the target flow are input into a second preset data model so that the second preset data model performs node flow splitting processing on the target flow to obtain the node target flow at the specified time node, including:

[0070] S5061. If at least one of the time nodes is included before the designated time node, the flow information before the node is processed by a first preset rule to obtain newly added flow data of each of the time nodes before the designated time node in the prediction time period.

[0071] It is understandable that the specified time node may include one time node or at least one time node before the specified time node. When at least one of the time nodes is included before the specified time node, the pre-node traffic information includes accumulated traffic data and node traffic data. The first preset rule is used to calculate the newly added traffic data, and the first preset rule may be:

[0072] D i =C i –C i-1

[0073] Among them, D i is the new traffic data of the specified APP at time node i, C i is the cumulative traffic data of the specified APP at time node i, C i-1 It is the cumulative traffic data of the specified APP at time node i-1.

[0074] S5062. Process the newly added traffic data and the node traffic data by using a second preset rule to obtain the traffic increase rate at the time node.

[0075] It is understandable that the node traffic data includes the traffic at each time node before the specified time node. The second preset rule is used to calculate the traffic increase rate, and the second preset rule can be:

[0076] E i =D i / B i

[0077] Among them, E i is the traffic increase rate of the specified APP at time node i, D i is the new traffic data of the specified APP at time node i, B i It is the node traffic data of the specified APP at time node i.

[0078] S5063. Determine the flow weight of the target flow on the unit average flow through a third preset rule.

[0079] It can be understood that the flow weight is the weight of the target flow on the unit average flow. The third preset rule is used to determine the flow weight of the target flow on the unit average flow. Among them, the third preset rule can be:

[0080] A=M / N

[0081] Among them, A is the flow weight, M is the target flow, and N is the unit average flow.

[0082] S5064. Process the accumulated flow data and the target flow using a fourth preset rule to obtain the remaining target flow over the predicted time period.

[0083] It is understandable that the remaining target flow is the difference flow between the specified APP and the target flow at the specified time node. The fourth preset rule is used to determine the remaining target flow. Among them, the fourth preset rule can be:

[0084] F i =MC i-1

[0085] Among them, F i is the remaining target flow of the specified APP at time node i, M is the target flow, C i-1 It is the cumulative traffic data of the specified APP at time node i-1.

[0086] S5065. Determine the remaining flow ratio of the remaining target flow in the node flow data through the fifth preset rule.

[0087] It can be understood that the remaining traffic proportion is the proportion of the remaining target traffic in the node traffic data. The fifth preset rule is used to determine the remaining traffic proportion. Among them, the fifth preset rule can be:

[0088] G i =F i / B i

[0089] Among them, G i is the remaining traffic proportion of the specified APP at time node i, F i is the remaining target flow of the specified APP at time node i, B i It is the node traffic data of the specified APP at time node i.

[0090] S5066. Determine the node target flow according to the remaining target flow, the remaining flow proportion, the flow increase rate and the flow weight.

[0091] It is understandable that the node target flow is the predicted flow at each time node. The node target flow can be obtained by processing the remaining target flow, the remaining flow proportion, the flow rate and the flow weight through the second preset data model. Among them, the second preset data model includes:

[0092] H i =(F i *G i ) / (E i *A)

[0093] Where i is a positive integer greater than or equal to 2, H i is the node target flow of the specified APP at time node i. i is the remaining target flow of the specified APP at time node i, G i is the remaining traffic proportion of the specified APP at time node i, E i is the traffic increase rate of the specified APP at time node i, and A is the traffic weight.

[0094] Preferably, when a time node may be included before the specified time node, the second preset data model includes:

[0095] H i =(F i *G i ) / E i

[0096] Where i is 1, H i is the node target flow of the specified APP at time node i. i is the remaining target flow of the specified APP at time node i, G i is the remaining traffic proportion of the specified APP at time node i, E i It is the traffic increase rate of the specified APP at time node i.

[0097] Preferably, when performing node traffic splitting processing on the target traffic and obtaining the node target traffic of the specified time node, the months can also be classified according to the number of days in each month, or the dates can be classified according to holidays such as weekends and holidays, so as to obtain the target traffic of a certain month or the target traffic of a certain day.

[0098] Optionally, in step S30, determining the unit flow data of the specified APP in the unit time period according to the historical flow data includes:

[0099] S301, cleaning the historical traffic data by a preset cleaning method to obtain cleaned data;

[0100] S302: Calculate the unit flow data of the designated APP in the unit time period according to the cleaning data.

[0101] Understandably, data cleaning refers to the process of reviewing and verifying data, with the goal of removing duplicate information, correcting existing errors, and ensuring data consistency. Preset cleaning methods refer to pre-set rules for data cleaning, including but not limited to mathematical statistics and data mining. Preset cleaning methods can transform dirty data into data that meets data quality requirements.

[0102] Specifically, the historical traffic data is cleaned by a preset cleaning method, and dirty data that does not meet the requirements is deleted to obtain cleaned data that meets the requirements. After obtaining the cleaned data, the unit traffic data of the specified APP in the unit time period is calculated based on the cleaned data.

[0103] Optionally, in step S302, the calculating of the unit flow data of the specified APP in the unit time period according to the cleaning data includes:

[0104] S3021. Obtain indicator classification rules;

[0105] S3022, classifying the cleaned data according to the indicator classification rule to obtain a plurality of indicator cleaned data, where one indicator cleaned data corresponds to one indicator;

[0106] S3023. Count each portion of the indicator cleaning data according to the unit time period to obtain unit flow data corresponding to each of the indicators.

[0107] It is understandable that the indicator classification rule is a pre-set rule for classifying the cleaned data. Preferably, the indicator classification rule can be set to three classification indicators. For example, the three classification indicators are new users, old users and total users. Then, according to the indicator classification rule, the cleaned data can be divided into three parts: new user data, old user data and total user data. Among them, the new user data, the old user data and the total user data are the indicator cleaned data. The classification indicator corresponding to the new user data is the new user, the classification indicator corresponding to the old user data is the old user, and the classification indicator corresponding to the total user data is the total user. Statistics are performed on each indicator cleaned data according to the unit time period to obtain the unit flow data corresponding to each indicator.

[0108] It should be understood that the order of execution of the steps in the above embodiment does not necessarily mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiment of the present invention.

[0109] In one embodiment, an APP traffic prediction device is provided, which corresponds one-to-one to the APP traffic prediction method in the above embodiment. Figure 3 As shown, the APP traffic prediction device includes a historical time period acquisition module 10, a historical traffic data acquisition module 20, a unit traffic data module 30, a unit experience coefficient module 40 and a target traffic module 50. The functional modules are described in detail as follows:

[0110] The historical time period acquisition module 10 is used to acquire a preset historical time period corresponding to a target time period; the target time period includes at least one prediction time period; the preset historical time period includes at least two unit time periods;

[0111] The historical traffic data acquisition module 20 is used to acquire the historical traffic data within the preset historical time period from the specified APP;

[0112] The unit flow data module 30 is used to determine the unit flow data corresponding to the specified APP in each unit time period according to the historical flow data;

[0113] The unit empirical coefficient module 40 is used to input the unit time period as an independent variable and the corresponding unit flow data as a dependent variable into a first preset mathematical model, and determine the unit empirical coefficient between the independent variable and the dependent variable through the first preset mathematical model;

[0114] The target flow module 50 is used to determine the target flow within the target time period according to the unit experience coefficient and the unit flow data.

[0115] Optionally, after the target flow module 50, it includes:

[0116] A real-time traffic module is used to obtain the real-time traffic within the target time period;

[0117] A comparison result module, used for comparing the real-time flow with the target flow to obtain a comparison result;

[0118] An abnormality level module, used to determine the abnormality level of the real-time traffic in the comparison result according to a preset abnormality level rule;

[0119] The first warning information module is used to generate the first traffic warning information of the designated APP according to the abnormal level.

[0120] Optionally, the comparison result module includes:

[0121] An early warning threshold unit, used to obtain a flow early warning threshold matching the target flow;

[0122] An access log unit, configured to obtain an access log corresponding to the real-time traffic from the designated APP if the real-time traffic is greater than the traffic warning threshold; the access log includes an access request and a response time of the access request;

[0123] A judgment result unit, used to judge whether the response time is greater than a preset response threshold and obtain a judgment result;

[0124] The second warning information unit is used to generate the second traffic warning information of the designated APP according to the judgment result.

[0125] Optionally, after the target flow module 50, the following is further included:

[0126] A unit average flow module, used to determine the unit average flow within the preset historical time period according to the historical flow data;

[0127] A pre-node flow information module is used to obtain pre-node flow information before a specified time node in the prediction time period;

[0128] The node target flow module is used to input the unit average flow, the pre-node flow information and the target flow into a second preset data model, so that the second preset data model performs node flow splitting processing on the target flow to obtain the node target flow of the specified time node.

[0129] Optionally, the pre-node traffic information includes accumulated traffic data and node traffic data; the node target traffic module includes:

[0130] A newly added traffic data unit is used to process the traffic information before the node according to a first preset rule to obtain the newly added traffic data of each of the time nodes before the specified time node in the prediction time period if at least one of the time nodes is included before the specified time node;

[0131] A traffic increase rate unit, configured to process the newly added traffic data and the node traffic data according to a second preset rule to obtain a traffic increase rate at the time node;

[0132] a remaining target flow unit, configured to process the accumulated flow data and the target flow according to a fourth preset rule to obtain a remaining target flow over the predicted time period;

[0133] A remaining flow ratio unit, used to determine the remaining flow ratio of the remaining target flow on the node flow data by a fifth preset rule;

[0134] A node target flow unit is used to determine the node target flow according to the remaining target flow, the remaining flow proportion, the flow increase rate and the flow weight.

[0135] Optionally, the unit flow data module 30 includes:

[0136] A data cleaning unit, used to clean the historical traffic data by a preset cleaning method to obtain cleaned data;

[0137] The unit flow data calculation unit is used to calculate the unit flow data of the specified APP within the unit time period according to the cleaning data.

[0138] Optionally, the unit flow data unit includes:

[0139] A classification rule unit, used to obtain indicator classification rules;

[0140] An indicator cleaning data unit, used to classify the cleaning data according to the indicator classification rule to obtain a plurality of indicator cleaning data, where one indicator cleaning data corresponds to one indicator;

[0141] The unit flow data statistics unit is used to count each of the indicator cleaning data according to the unit time period to obtain the unit flow data corresponding to each of the indicators.

[0142] For the specific limitations of the APP traffic prediction device, please refer to the limitations of the APP traffic prediction method above, which will not be repeated here. Each module in the above-mentioned APP traffic prediction device can be implemented in whole or in part by software, hardware and a combination thereof. The above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.

[0143] In one embodiment, a computer device is provided. The computer device may be a terminal, and its internal structure diagram may be as follows: Figure 4 As shown. The computer device includes a processor, a memory, a network interface, a display screen and an input device connected via a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a readable storage medium and an internal memory. The non-volatile storage medium stores an operating system and computer-readable instructions. The internal memory provides an environment for the operation of the operating system and computer-readable instructions in the readable storage medium. The network interface of the computer device is used to communicate with an external server through a network connection. When the computer-readable instructions are executed by the processor, an APP traffic prediction method is implemented. The readable storage medium provided in this embodiment includes a non-volatile readable storage medium and a volatile readable storage medium.

[0144] In one embodiment, a computer device is provided, including a memory, a processor, and computer-readable instructions stored in the memory and executable on the processor, wherein the processor executes the computer-readable instructions to implement the following steps:

[0145] Acquire a preset historical time period corresponding to a target time period; the target time period includes at least one predicted time period; the preset historical time period includes at least two unit time periods;

[0146] Obtain historical traffic data within the preset historical time period from the specified APP;

[0147] Determine, according to the historical traffic data, the unit traffic data corresponding to the designated APP in each of the unit time periods;

[0148] The unit time period is used as an independent variable and the corresponding unit flow data is used as a dependent variable, and the results are input into a first preset mathematical model, and a unit empirical coefficient between the independent variable and the dependent variable is determined by the first preset mathematical model;

[0149] The target flow rate within the target time period is determined according to the unit empirical coefficient and the unit flow rate data.

[0150] In one embodiment, one or more computer-readable storage media storing computer-readable instructions are provided. The readable storage media provided in this embodiment include non-volatile readable storage media and volatile readable storage media. The readable storage media store computer-readable instructions, and when the computer-readable instructions are executed by one or more processors, the following steps are implemented:

[0151] Acquire a preset historical time period corresponding to a target time period; the target time period includes at least one predicted time period; the preset historical time period includes at least two unit time periods;

[0152] Obtain historical traffic data within the preset historical time period from the specified APP;

[0153] Determine, according to the historical traffic data, the unit traffic data corresponding to the designated APP in each of the unit time periods;

[0154] The unit time period is used as an independent variable and the corresponding unit flow data is used as a dependent variable, and the results are input into a first preset mathematical model, and a unit empirical coefficient between the independent variable and the dependent variable is determined by the first preset mathematical model;

[0155] The target flow rate within the target time period is determined according to the unit empirical coefficient and the unit flow rate data.

[0156] Those of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be implemented by instructing related hardware through computer-readable instructions, and the computer-readable instructions can be stored in a non-volatile readable storage medium or a volatile readable storage medium. When the computer-readable instructions are executed, they may include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0157] Those skilled in the art can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example. In actual applications, the above-mentioned functions can be distributed and completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above.

[0158] The embodiments described above are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included in the protection scope of the present invention.

Claims

1. A method for predicting APP traffic, characterized in that: include: Acquire a preset historical time period corresponding to a target time period; the target time period includes at least one predicted time period; the preset historical time period includes at least two unit time periods; Obtain historical traffic data within the preset historical time period from the specified APP; Determine, according to the historical traffic data, the unit traffic data corresponding to the designated APP in each of the unit time periods; The unit time period is used as an independent variable and the corresponding unit flow data is used as a dependent variable, and the results are input into a first preset mathematical model, and a unit empirical coefficient between the independent variable and the dependent variable is determined by the first preset mathematical model; Determine the target flow rate within the target time period according to the unit experience coefficient and the unit flow data; The prediction time period includes at least two time nodes; After determining the target flow rate within the target time period according to the unit experience coefficient and the unit flow data, the method further includes: Determine the unit average flow rate within the preset historical time period according to the historical flow data; Obtaining pre-node traffic information before a specified time node in the prediction time period; Inputting the unit average flow, the pre-node flow information and the target flow into a second preset data model, so that the second preset data model performs node flow splitting processing on the target flow to obtain the node target flow at the specified time node; The pre-node flow information includes cumulative flow data and node flow data; the unit average flow, the pre-node flow information and the target flow are input into a second preset data model so that the second preset data model performs node flow splitting processing on the target flow to obtain the node target flow at the specified time node, including: If at least one of the time nodes is included before the designated time node, the traffic information before the node is processed by a first preset rule to obtain newly added traffic data of each of the time nodes before the designated time node in the prediction time period; Processing the newly added traffic data and the node traffic data by a second preset rule to obtain a traffic increase rate at the time node; Determining the flow weight of the target flow on the average flow of the unit by a third preset rule; Processing the accumulated flow data and the target flow through a fourth preset rule to obtain the remaining target flow over the predicted time period; Determine the remaining flow ratio of the remaining target flow on the node flow data by a fifth preset rule; Determine the node target flow according to the remaining target flow, the remaining flow proportion, the flow increase rate and the flow weight; The first preset rule is: D i =C i –C i-1 Among them, D i is the new traffic data of the specified APP at time node i, C i is the cumulative traffic data of the specified APP at time node i, C i-1 The cumulative traffic data of the specified APP at time node i-1; The second preset rule is: E i =D i / B i Among them, E i is the traffic increase rate of the specified APP at time node i, D i is the new traffic data of the specified APP at time node i, B i It is the node traffic data of the specified APP at time node i; The third preset rule is: A=M / N Among them, A is the flow weight, M is the target flow, and N is the unit average flow; The fourth preset rule is: F i =M-C i-1 Among them, F i is the remaining target flow of the specified APP at time node i, M is the target flow, C i-1 The cumulative traffic data of the specified APP at time node i-1; The fifth preset rule is: G i =F i / B i Among them, G i is the remaining traffic proportion of the specified APP at time node i, F i is the remaining target flow of the specified APP at time node i, B i It is the node traffic data of the specified APP at time node i.

2. The APP traffic prediction method according to claim 1, characterized in that: After determining the target flow rate within the target time period according to the unit empirical coefficient and the unit flow rate data, the method further comprises: Obtaining real-time traffic within the target time period; Compare the real-time flow rate with the target flow rate to obtain a comparison result; Determining the abnormality level of the real-time traffic in the comparison result according to a preset abnormality level rule; According to the abnormality level, first traffic warning information of the designated APP is generated.

3. The APP traffic prediction method according to claim 2, characterized in that: The comparing the real-time flow rate with the target flow rate to obtain a comparison result includes: Obtaining a flow warning threshold that matches the target flow; If the real-time traffic is greater than the traffic warning threshold, obtaining an access log corresponding to the real-time traffic from the designated APP; the access log includes an access request and a response time of the access request; Determine whether the response time is greater than a preset response threshold, and obtain a determination result; Generate second traffic warning information of the designated APP according to the judgment result.

4. The APP traffic prediction method according to claim 1, characterized in that: The determining, according to the historical flow data, the unit flow data of the specified APP in the unit time period includes: Cleaning the historical traffic data by a preset cleaning method to obtain cleaned data; The unit flow data of the designated APP in the unit time period is calculated according to the cleaning data.

5. The APP traffic prediction method according to claim 4, characterized in that: The calculating the unit flow data of the designated APP in the unit time period according to the cleaning data includes: Obtain indicator classification rules; Classify the cleaned data according to the indicator classification rule to obtain a plurality of indicator cleaned data, where one indicator cleaned data corresponds to one indicator; Statistics are collected for each of the indicator cleaning data according to the unit time period to obtain unit flow data corresponding to each of the indicators.

6. An APP traffic prediction device, characterized in that: include: A historical time period acquisition module, used to acquire a preset historical time period corresponding to a target time period; the target time period includes at least one predicted time period; the preset historical time period includes at least two unit time periods; A historical traffic data acquisition module is used to acquire the historical traffic data within the preset historical time period from a specified APP; A unit flow data module, used to determine the unit flow data corresponding to the specified APP in each unit time period according to the historical flow data; A unit empirical coefficient module, used to input the unit time period as an independent variable and the corresponding unit flow data as a dependent variable into a first preset mathematical model, and determine the unit empirical coefficient between the independent variable and the dependent variable through the first preset mathematical model; A target flow module, for determining a target flow within the target time period according to the unit experience coefficient and the unit flow data; The prediction time period includes at least two time nodes; After determining the target flow rate within the target time period according to the unit experience coefficient and the unit flow data, the method further includes: Determine the unit average flow rate within the preset historical time period according to the historical flow data; Obtaining pre-node traffic information before a specified time node in the prediction time period; Inputting the unit average flow, the pre-node flow information and the target flow into a second preset data model, so that the second preset data model performs node flow splitting processing on the target flow to obtain the node target flow at the specified time node; The pre-node traffic information includes accumulated traffic data and node traffic data; the target traffic module includes: If at least one of the time nodes is included before the designated time node, the traffic information before the node is processed by a first preset rule to obtain newly added traffic data of each of the time nodes before the designated time node in the prediction time period; Processing the newly added traffic data and the node traffic data by a second preset rule to obtain a traffic increase rate at the time node; Determining the flow weight of the target flow on the average flow of the unit by a third preset rule; Processing the accumulated flow data and the target flow through a fourth preset rule to obtain the remaining target flow over the predicted time period; Determine the remaining flow ratio of the remaining target flow on the node flow data by a fifth preset rule; Determine the node target flow according to the remaining target flow, the remaining flow proportion, the flow rate and the flow weight; The first preset rule is: D i =C i –C i-1 Among them, D i is the new traffic data of the specified APP at time node i, C i is the cumulative traffic data of the specified APP at time node i, C i-1 The cumulative traffic data of the specified APP at time node i-1; The second preset rule is: E i =D i / B i Among them, E i is the traffic increase rate of the specified APP at time node i, D i is the new traffic data of the specified APP at time node i, B i It is the node traffic data of the specified APP at time node i; The third preset rule is: A=M / N Among them, A is the flow weight, M is the target flow, and N is the unit average flow; The fourth preset rule is: F i =M-C i-1 Among them, F i is the remaining target flow of the specified APP at time node i, M is the target flow, C i-1 The cumulative traffic data of the specified APP at time node i-1; The fifth preset rule is: G i =F i / B i Among them, G i is the remaining traffic proportion of the specified APP at time node i, F i is the remaining target flow of the specified APP at time node i, B i It is the node traffic data of the specified APP at time node i.

7. A computer device comprising a memory, a processor, and computer-readable instructions stored in the memory and executable on the processor, characterized in that: When the processor executes the computer-readable instructions, the APP traffic prediction method as described in any one of claims 1 to 5 is implemented.

8. One or more readable storage media storing computer-readable instructions, wherein when the computer-readable instructions are executed by one or more processors, the one or more processors execute the APP traffic prediction method as described in any one of claims 1 to 5.

Images