Data processing method, computing device and storage medium

Abstract

The embodiment of the invention provides a data processing method, computing equipment and a storage medium, and the data processing method comprises the following steps: determining resource use data of a target project; performing resource prediction on the resource use data by using a resource prediction model to obtain a plurality of predicted resource thresholds corresponding to the target project and predicted resource use data corresponding to the target project in a target time range; and determining a target project resource of the target project, and performing resource detection according to the target project resource, the predicted resource use data and the plurality of predicted resource thresholds to determine a resource detection result corresponding to the target project. The problem that the resource use condition cannot be accurately detected is avoided; and the problem that due to the fact that the resource use condition cannot be accurately detected, the execution of the service items has large negative effects due to too fast resource consumption, uneven resource use and the like is solved.

Description

Technical Field

[0001] The embodiments of this specification relate to the field of computer technology, and more particularly to a data processing method. One or more embodiments of this specification also relate to a data processing device, a computing device, a computer-readable storage medium, and a computer program product. Background Art

[0002] During the execution of a service project, in order to achieve better results in the execution of the service project, certain resources may be allocated to the service project for use.

[0003] In the process of using resources, current service projects may not be able to accurately detect resource usage, which may cause the execution of service projects to have a greater negative impact due to reasons such as too fast resource consumption and uneven resource usage; therefore, how to accurately detect resource usage has become an issue that needs to be urgently addressed. Summary of the invention

[0004] In view of this, an embodiment of this specification provides a data processing method. One or more embodiments of this specification also relate to a data processing device, a computing device, a computer-readable storage medium and a computer program product to solve the technical defects existing in the prior art.

[0005] According to a first aspect of an embodiment of this specification, a data processing method is provided, including: Determine resource usage data for target projects; Performing resource forecasting on the resource usage data using a resource forecasting model to obtain a plurality of forecast resource thresholds corresponding to the target project and forecast resource usage data corresponding to the target project within a target time range; The target project resources of the target project are determined, and resource detection is performed according to the target project resources, the predicted resource usage data, and the multiple predicted resource thresholds to determine a resource detection result corresponding to the target project.

[0006] According to a second aspect of an embodiment of this specification, there is provided a data processing device, including: a data determination module configured to determine resource usage data of a target project; A resource prediction module is configured to perform resource prediction on the resource usage data using a resource prediction model to obtain a plurality of predicted resource thresholds corresponding to the target project and predicted resource usage data corresponding to the target project within a target time range; The resource detection module is configured to determine the target project resources of the target project, and perform resource detection according to the target project resources, the predicted resource usage data and the multiple predicted resource thresholds to determine the resource detection result corresponding to the target project.

[0007] According to a third aspect of an embodiment of this specification, a computing device is provided, including: Memory and processor; The memory is used to store computer programs / instructions, and the processor is used to execute the computer programs / instructions. When the computer programs / instructions are executed by the processor, the steps of the above method are implemented.

[0008] According to a fourth aspect of the embodiments of this specification, a computer-readable storage medium is provided, which stores a computer program / instruction, and the steps of the above method are implemented when the computer program / instruction is executed by a processor.

[0009] According to a fifth aspect of the embodiments of this specification, a computer program product is provided, comprising a computer program / instruction, which implements the steps of the above method when executed by a processor.

[0010] One or more embodiments of the present specification provide a data processing method. During resource detection for a target project, resource usage data of the target project for the target project resources can be obtained, and a resource prediction model can be used to perform resource prediction on the resource usage data, thereby efficiently and accurately obtaining multiple predicted resource thresholds corresponding to the target project, and predicted resource usage data corresponding to the target project within a target time range; then, resource detection is performed based on the target project resources, predicted resource usage data, and multiple predicted resource thresholds of the target project, thereby accurately determining the resource detection result corresponding to the target project, and the usage of the target project resources by the target project can be determined through the resource detection result; the problem of being unable to accurately detect resource usage is avoided; and the problem of being unable to accurately detect resource usage, resulting in the execution of the service project having a large negative impact due to reasons such as excessive resource consumption and uneven resource usage, is avoided. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] Figure 1 It is an application schematic diagram of a data processing method provided by an embodiment of this specification; Figure 2 is a flow chart of a data processing method provided by an embodiment of this specification; Figure 3 It is a processing schematic diagram of a data processing method provided by an embodiment of this specification; Figure 4is a structural schematic diagram of a data processing device provided by an embodiment of this specification; Figure 5 It is a structural block diagram of a computing device provided by an embodiment of this specification. DETAILED DESCRIPTION

[0012] Many specific details are described in the following description to facilitate a full understanding of this specification. However, this specification can be implemented in many other ways than those described herein, and those skilled in the art can make similar generalizations without violating the connotation of this specification, so this specification is not limited to the specific implementation disclosed below.

[0013] The terms used in one or more embodiments of this specification are only for the purpose of describing specific embodiments, and are not intended to limit one or more embodiments of this specification. The singular forms of "a", "said" and "the" used in one or more embodiments of this specification and the appended claims are also intended to include plural forms, unless the context clearly indicates other meanings. It should also be understood that the term "and / or" used in one or more embodiments of this specification refers to and includes any or all possible combinations of one or more associated listed items.

[0014] It should be understood that although the terms first, second, etc. may be used to describe various information in one or more embodiments of this specification, this information should not be limited to these terms. These terms are only used to distinguish the same type of information from each other. For example, without departing from the scope of one or more embodiments of this specification, the first may also be referred to as the second, and similarly, the second may also be referred to as the first. Depending on the context, the word "if" as used herein may be interpreted as "at the time of" or "when" or "in response to determining".

[0015] In addition, it should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in one or more embodiments of this specification are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with the relevant laws, regulations and standards of the relevant countries and regions, and provide corresponding operation entrances for users to choose to authorize or refuse.

[0016] First, the terms involved in one or more embodiments of this specification are explained.

[0017] Activity timeliness detection: refers to the validity period and budget detection of marketing activities, including the budget consumption of prizes, coupon templates, instant discount activities, credit cards / digital financial rights; activity, prizes, rights, tasks, groups, and delivery status and validity period detection.

[0018] Time series forecasting: refers to a method of analyzing the internal laws of historical data series and predicting future data trends.

[0019] Budget warning: proactively discovers that activity budgets are insufficient and triggers an alarm.

[0020] ‌Prophet‌: is a forecasting model based on time series.

[0021] Time series data: also known as time series data, which is data collected at different times and used to describe situations where the phenomenon changes over time. This time series data can be a collection of data recorded in chronological order.

[0022] In the process of executing a service project, in order to achieve better results, certain resources can be allocated to the service project for use. In the process of using resources, the current service project may not be able to accurately detect the resource usage, which may cause the execution of the service project to have a greater negative impact due to reasons such as too fast resource consumption and uneven resource usage.

[0023] For example, in the operation scenarios of specific services such as credit cards and loans of financial institutions, there will be a large number of marketing activities, and each marketing activity is associated with a large number of prizes. Through the analysis of historical online problems, it is found that among the online problems related to the design of activity configuration, the two major categories of problems, insufficient budget and expired activities, account for more than 560% to 57%; further analysis of the specific causes of the problems is mainly due to: 1. The noise of insufficient budget alarm is large: specifically, it means that the budget consumption is a fixed threshold alarm, and it is impossible to accurately issue an alarm in the case of insufficient budget according to the actual situation; 2. The alarm has no closed-loop capability: it means that the alarm notification channels are scattered, and the notification is not centralized, which is easy to be ignored; 3. The alarm channels are scattered: there is no unified alarm in the activity dimension, and the alarm information is in various forms, which leads to unclear alarms.

[0024] Based on this, a data processing method is provided in this specification. One or more embodiments of this specification also involve a data processing device, a computing device, a computer-readable storage medium and a computer program product, which are described in detail one by one in the following embodiments.

[0025] See also Figure 1 , Figure 1 A schematic diagram of an application of a data processing method provided according to an embodiment of the present specification is shown. Figure 1It can be seen that the server 104 can obtain the time series data of budget usage (i.e., resource usage data), and then input the time series data into the Prophet model (i.e., resource prediction model) for prediction, so as to obtain the budget consumption prediction data for the future time (i.e., predicted resource usage data) and the budget consumption dynamic threshold (i.e., predicted resource threshold), and then perform risk analysis based on the budget consumption prediction data for the future time, the budget consumption dynamic threshold and the current remaining budget of the marketing activity, so as to obtain the alarm information for the activity budget corresponding to the marketing activity (i.e., resource detection result); after obtaining the alarm information, the server 104 can send the alarm information to the client 102, so as to warn the user of the risk.

[0026] See also Figure 2 , Figure 2 A flow chart of a data processing method provided according to an embodiment of the present specification is shown, which specifically includes the following steps.

[0027] Step 202: Determine resource usage data of the target project.

[0028] Among them, the target project can be understood as a project that provides target services, and the target project is a service project; the service project can be a reward service, a subsidized service, a sponsored activity, etc.; for example, the target project (service project) can be a marketing project, a marketing activity, a reward activity, etc.

[0029] The resource usage data can be understood as the usage data of the target project resources during the execution of the target project. The resource usage data can be the resource usage data within the budget historical time range, for example, the resource usage data in the past day, week, or month; the resource usage data can be time series data, that is, the resource usage data can be resource usage chronological data.

[0030] The target project resources may be resources allocated to the target project, and may be cloud service resources, data processing resources, funds, prizes, coupons, etc.; the cloud service resources may be cloud storage resources, cloud computing resources; the data processing resources may be understood as bandwidth resources for data transmission, storage resources for data storage, etc.; the target project resources may also be resources used for venues (such as warehouses, booths, etc.) and resources used for publicity channels (such as display boards, advertising spaces, advertising screens).

[0031] In one or more embodiments provided in this specification, the resource usage data may be data that has been preprocessed, thereby ensuring the accuracy and integrity of the resource usage data and improving the accuracy and efficiency of subsequent resource predictions; specifically, determining the resource usage data of the target project includes: Determine resource usage data to be processed corresponding to multiple types of service items, and perform data preprocessing on the resource usage data to be processed to obtain resource usage data of various types of service items; A target project is determined from the various types of service projects, and resource usage data of the target project is determined from the resource usage data, wherein the target project is any type of service project among the multiple types of service projects.

[0032] Among them, the resource usage data to be processed can be understood as resource usage data that requires data preprocessing; the data preprocessing can be understood as a data cleaning operation, which includes "project relationship association", "feature extraction", "historical data backtracking", etc.

[0033] In one or more embodiments provided in this specification, the preprocessing of the resource usage data to be processed to obtain resource usage data of various types of service items includes: Filtering the resource usage data to be processed to obtain resource usage data of various types of service items; and / or Extracting features from the resource usage data to be processed to obtain resource usage data for various types of service items; and / or The historical resource usage data corresponding to the various types of service items are determined from the data storage unit, and the historical resource usage data and the to-be-processed resource usage data are used as resource usage data for the various types of service items.

[0034] The data storage unit may be understood as a unit for storing historical resource usage data, and the data storage unit may be a storage device such as a database, a hard disk, or a memory.

[0035] Performing feature extraction on the resource usage data to be processed can be understood as extracting key technical feature data from the resource usage data to be processed, and using the key technical feature data as resource usage data for various types of service items.

[0036] Specifically, the data processing method provided in this specification can obtain the to-be-processed resource usage data corresponding to various types of service items from the resource usage data storage unit (such as a database, hard disk, server) corresponding to various types of service items through a data processing unit.

[0037] The data processing unit may be understood as a unit for processing resource usage data to be processed, and the data processing unit may be a server, a service end, a data processing system, a data processing platform, and the like.

[0038] Then, the resource usage data to be processed is subjected to data preprocessing operations such as data filtering, feature extraction and / or combination with historical resource usage data, so as to obtain resource usage data of various types of service items, thereby improving the performance and accuracy of the resource usage data.

[0039] Then, for various types of service items, a target item is determined from the various types of service items, and resource usage data of the target item is determined from the resource usage data, and a resource detection operation can be performed on the target item subsequently; the target item can be determined from various types of service items in a polling manner; or, the target item can be any type of service item among multiple types of service items.

[0040] Step 204: Perform resource prediction on the resource usage data using a resource prediction model to obtain a plurality of predicted resource thresholds corresponding to the target project and predicted resource usage data corresponding to the target project within a target time range.

[0041] Among them, the resource prediction model can be understood as a model that predicts resource usage based on resource usage data. The resource prediction model can be a neural network model or a deep learning model. When the resource usage data is time series data, the resource prediction model can be a model that can make predictions based on time series data. For example, the resource prediction model can be a Prophet model.

[0042] The predicted resource threshold may be a dynamic threshold, and different predicted resource thresholds may be obtained by performing resource prediction based on different resource usage data. For example, multiple remaining prize quantity thresholds (i.e., multiple predicted resource thresholds) for a reward marketing activity (i.e., a target project); the multiple remaining prize quantity thresholds may be 10%, 15%, and 30% of the total prize quantity allocated to the reward marketing activity.

[0043] Predicted resource usage data can be understood as resource usage data for target project resources within a target time range, obtained by predicting using a resource prediction model; for example, predicting the number of prizes (i.e., predicted resource usage data) that a reward marketing campaign (i.e., target project) will distribute to users within the next day (e.g., within the target time range).

[0044] The target time range can be set according to the actual application scenario, for example, the target time range can be the next day, the next two days, the next week, etc., and no specific limitation is made here. The target time range is used to predict the usage of the target project resources of the target project in the target time range (i.e., predict resource usage data).

[0045] In one or more embodiments provided in this specification, in the process of using the resource prediction model to perform resource prediction on the resource usage data, the resource usage data needs to be input into the resource prediction model for resource prediction; the inputting the resource usage data into the resource prediction model for resource prediction includes: According to a preset data format, the data format of the resource usage data is adjusted to obtain the adjusted resource usage data; thereby ensuring that the resource prediction model can process data in a uniform format, thereby improving the efficiency of resource prediction.

[0046] The adjusted resource usage data is input into the resource prediction model, so as to use the resource prediction model to perform resource prediction on the resource usage data, thereby obtaining multiple predicted resource thresholds corresponding to the target project and the predicted resource usage data corresponding to the target project within the target time range.

[0047] Among them, this method adjusts the data format of the resource usage data according to a preset data format to obtain the adjusted resource usage data, and can also determine preset data prompt information (promote), splice the resource usage data with the preset data prompt information (i.e., the preset data format), thereby obtaining input data (i.e., the adjusted resource usage data), and the input data conforms to the input format of the resource prediction model. Subsequently, the input data can be input into the resource prediction model, and the resource prediction model can be used to perform resource prediction on the resource usage data.

[0048] In one or more embodiments provided in this specification, the future guarantees that the predicted resource usage data within the future time range can be accurately determined, and multiple predicted resource thresholds corresponding to different preset time ranges can be accurately determined. The method can use time series data and a prediction model based on time series to perform resource prediction; specifically, the resource usage data is resource usage time series data; Using a resource prediction model to perform resource prediction on the resource usage data, obtaining a plurality of predicted resource thresholds corresponding to the target project, and predicted resource usage data corresponding to the target project within a target time range, including: The resource prediction model is used to perform resource prediction on the resource usage time series data to obtain multiple predicted resource thresholds corresponding to the target project and predicted resource usage data corresponding to the target project within the target time range, wherein each of the multiple predicted resource thresholds corresponds to a different preset time range.

[0049] Among them, resource usage time series data can be understood as time series resource usage data, such as budget consumption time series data.

[0050] The multiple preset time ranges may be multiple different prediction time ranges. For example, the multiple prediction resource thresholds may be prediction resource threshold A, prediction resource threshold B, and prediction resource threshold C. The preset time range corresponding to the distribution may be the next day, the next two days, or the next three days. It should be noted that the preset time ranges corresponding to the prediction resource thresholds may or may not have time range overlaps.

[0051] Step 206: Determine the target project resources of the target project, and perform resource detection according to the target project resources, the predicted resource usage data, and the multiple predicted resource thresholds to determine the resource detection result corresponding to the target project.

[0052] The resource detection result may be understood as the result obtained after detecting the usage of the target project resources; the predicted detection result may be an alarm message, a prompt message, and the like.

[0053] In one or more embodiments provided in this specification, performing resource detection according to the target project resources, the predicted resource usage data, and the multiple predicted resource thresholds to determine the resource detection result corresponding to the target project includes: Calculate according to the target project resources and the predicted resource usage data to obtain the remaining resource value of the project corresponding to the target time range; Determining a target predicted resource threshold by comparing the remaining resource value of the project with each predicted resource threshold, wherein the remaining resource value of the project is less than the target predicted resource threshold; The resource detection result corresponding to the target project is determined according to the target predicted resource threshold and the remaining resource value of the project.

[0054] The remaining resource value of the project can be understood as the remaining resource value obtained by subtracting the predicted resource usage data from the target project resources.

[0055] Taking the application of the data processing method provided in this specification in the budget problem and activity timeliness problem detection scenarios as an example, the data processing method is explained, wherein the target project is a marketing activity, and the target project resources can be the prizes (budget) of the marketing activity; the resource prediction model is the Prophet model, and the resource usage data is the original time series data.

[0056] Based on this, this method cleans data through offline tasks to obtain the original time series data of the activity budget, and predicts the original data through the Prophet model to obtain the budget consumption forecast data (i.e., predicted resource usage data) and the budget consumption dynamic threshold (i.e., predicted resource threshold) for the future time. Then, budget risk analysis, risk classification processing and early warning broadcast can be performed based on the budget consumption forecast data and budget consumption dynamic threshold for the future time. The specific method is as follows: First, subtract the budget that will be consumed in the next 24 hours predicted by the algorithm (Prophet model) from the current remaining budget of the marketing campaign to obtain the remaining budget after the budget is consumed in the next 24 hours (i.e. the remaining resource value of the project); Secondly, the remaining budget amount is compared with the dynamic threshold of budget consumption to determine whether the marketing activity has a budget risk; when the remaining budget amount is greater than or equal to multiple dynamic thresholds of budget consumption, it is determined that the marketing activity does not have a budget risk; if the remaining budget amount is less than any dynamic threshold of budget consumption, it is determined that the marketing activity has a budget risk, and it is necessary to determine the risk level and generate corresponding warning information (i.e., resource detection results). It should be noted that the method of comparing the remaining budget amount with the dynamic threshold of budget consumption to determine whether the marketing activity has a budget risk can also be Br <Dt(24) Cb-24*Bcv <Dt(24)→P0、Cb-24*Bcv<Dt(48)→P1、Cb-24*Bcv<Dt(72)→P2。

[0057] In one or more embodiments provided in this specification, in order to timely handle the detected resource risk, the method can determine the resource warning information corresponding to the target project according to the target predicted resource threshold and the remaining resource value of the project, so as to facilitate the processing of the current resource risk according to the resource warning information, specifically; the resource detection result is the resource warning information; The determining, according to the target predicted resource threshold and the remaining resource value of the project, a resource detection result corresponding to the target project includes: Determine a resource usage alarm level corresponding to the target predicted resource threshold; Based on the resource usage alarm level and the remaining resource value of the project, resource alarm information corresponding to the target project is generated.

[0058] Continuing with the above example, when the remaining budget amount is less than the dynamic threshold of budget consumption, it is determined that there is a budget risk in the marketing activity; therefore, it is necessary to determine the risk level; the risk level is determined by determining the dynamic threshold of budget consumption that the remaining budget amount is less than, and determining the risk level corresponding to the dynamic threshold of budget consumption.

[0059] Then, according to the risk level and the predicted remaining budget amount, the corresponding warning information (ie, resource detection result) is generated.

[0060] It should be noted that the resource detection result in this method can also be a resource risk-free detection result. When the current resource risk-free problem is determined based on the resource risk-free detection result, the step of determining the resource usage data of the target project in the above step 202 is continued to be executed; thereby achieving continuous monitoring of resources.

[0061] In one or more embodiments provided in this specification, when the current target project meets multiple resource risk levels, a highest level can be selected from multiple resource risk levels (i.e., resource usage warning levels) to ensure timely processing of high risks. Specifically, the target predicted resource threshold is multiple; The determining, according to the target predicted resource threshold and the remaining resource value of the project, a resource detection result corresponding to the target project includes: Determine a resource usage alarm level corresponding to each target prediction resource threshold among multiple target prediction resource thresholds; Determine, according to the resource usage alarm level, a selected predicted resource threshold from the multiple target predicted resource thresholds, wherein the resource usage alarm level of the selected predicted resource threshold is greater than or equal to the resource usage alarm levels of other predicted resource thresholds in the multiple target predicted resource thresholds; The resource detection result corresponding to the target project is determined according to the selected predicted resource threshold and the remaining resource value of the project.

[0062] Using the above example, when the remaining budget is less than multiple dynamic budget consumption thresholds, it is determined that there is a budget risk in the marketing activity; however, since multiple dynamic budget consumption thresholds correspond to different risk levels (i.e., resource usage warning levels), it is necessary to determine the risk level corresponding to each dynamic budget consumption threshold, and select the highest risk level and the dynamic budget consumption threshold of the highest risk level (i.e., select the predicted resource threshold). Then, based on the highest risk level corresponding to the target predicted resource threshold and the predicted remaining budget, the corresponding warning information (i.e., resource detection result) is generated.

[0063] In one or more embodiments provided in this specification, after performing resource detection according to the target project resources, the predicted resource usage data, and the multiple predicted resource thresholds and determining the resource detection result corresponding to the target project, the method further includes: Determine unit information of a detection result receiving unit corresponding to the target item; The resource detection result is sent to the detection result receiving unit according to the unit information.

[0064] Among them, the detection result receiving unit can be understood as a unit that receives resource detection results. For example, the detection result receiving unit can be a mail server, instant messaging software, a mobile terminal, a client, etc.

[0065] The unit information may be the email address, user ID of instant messaging software, IP address, etc.

[0066] Continuing with the above example, after budget risk analysis and risk grading, early warning broadcasts can be executed. Specifically, the alarm data can be sent to the person in charge of the marketing activity via instant messaging software, and the work order can be closed-loop processed to ensure that risk issues are handled in a timely manner.

[0067] In one or more embodiments provided in this specification, the sending, according to the unit information, the resource detection result to the detection result receiving unit includes: Determining a target resource usage alarm level corresponding to the resource detection result, wherein the target resource usage alarm level is used to generate the resource detection result; Determine a result sending strategy corresponding to the target resource usage alarm level; The resource detection result is sent to the detection result receiving unit according to the unit information by utilizing the result sending strategy.

[0068] Continuing with the above example, this method sets different result sending strategies (notification strategies) for different risk levels; the risk levels can be P0, P1, and P2; for P0, a notification strategy of notifying once every 30 minutes can be adopted; for P1, a notification strategy of notifying once every hour can be adopted; for P2, a notification strategy of notifying once a day can be adopted, thereby achieving targeted sending of test results.

[0069] Based on the notification strategy, an early warning is broadcast to the person in charge of the marketing activity.

[0070] In one or more embodiments provided in this specification, the data processing method further includes: Determine the project execution end time of the target project and the preset time threshold corresponding to the target project; Calculate the time difference between the current time and the end time of the project execution; In the case where the time difference is less than the preset time threshold, generating a time detection result corresponding to the target item based on the time difference; The unit information of the detection result receiving unit corresponding to the target item is determined, and the time detection result is sent to the detection result receiving unit according to the unit information.

[0071] The project execution end time may be the end time of the marketing activity, and the forecast time may be set according to actual conditions, for example, three days, one day, etc.

[0072] The time detection result can be understood as time-sensitive alarm information or time-sensitive prompt information.

[0073] Using the above example, during the execution of a marketing campaign, one or more timeliness alarm thresholds can be set for the marketing campaign, such as one day, three days, one week, etc. If the time difference between the current time and the campaign deadline of the marketing campaign is less than the timeliness alarm prediction (one week), then a timeliness alarm message can be generated and sent to the person in charge of the marketing campaign for warning, thereby detecting the timeliness of the target project and ensuring the timeliness of the target project.

[0074] One or more embodiments of the present specification provide a data processing method. During resource detection for a target project, resource usage data of the target project for the target project resources can be obtained, and a resource prediction model can be used to perform resource prediction on the resource usage data, thereby efficiently and accurately obtaining multiple predicted resource thresholds corresponding to the target project, and predicted resource usage data corresponding to the target project within a target time range; then, resource detection is performed based on the target project resources, predicted resource usage data, and multiple predicted resource thresholds of the target project, thereby accurately determining the resource detection result corresponding to the target project, and the usage of the target project resources by the target project can be determined through the resource detection result; the problem of being unable to accurately detect resource usage is avoided; and the problem of being unable to accurately detect resource usage, resulting in the execution of the service project having a large negative impact due to reasons such as excessive resource consumption and uneven resource usage, is avoided.

[0075] The following combination Figure 3 , taking the application of the data processing method provided in this specification in the detection scenarios of activity budget problems and timeliness problems as an example, the data processing method is further explained. Figure 3 A processing schematic diagram of a data processing method provided by an embodiment of this specification is shown.

[0076] based on Figure 3 It can be seen that the specific implementation process of this method may include steps 1 to 5: Step 1: Confirm the scope of activities (activity dimension / project dimension) that require budget check through configuration.

[0077] Specifically, this solution can perform activity timeliness detection and activity budget problem detection on a large number of marketing activities; therefore, in the process of performing activity timeliness detection and activity budget problem detection, it is necessary to configure and confirm the activity scope (i.e., marketing activities) for budget detection and timeliness detection.

[0078] Step 2: Clean the data through offline tasks to obtain the original time series data of the activity budget.

[0079] Specifically, this solution can obtain the budget consumption time series data of each marketing activity from the data processing platform through the "data synchronization" operation; After obtaining the budget consumption time series data, you can perform data cleaning operations such as "data filtering", "project relationship association", "feature extraction", and "historical data backtracking" on the budget consumption time series data to obtain the original time series data of the activity budget.

[0080] Among them, data filtering can be understood as: filtering out time series data with format errors, redundancy, etc. in the budget consumption time series data.

[0081] Project relationship association can be understood as associating budget consumption time series data with related projects or marketing activities.

[0082] Feature extraction can be understood as: extracting key feature data (such as budget usage value and time) from the budget consumption time series data and removing non-key feature data.

[0083] Historical data backtracking can be understood as: after obtaining the budget consumption time series data of a marketing campaign in the past week through "data synchronization", you can obtain all the historical data obtained by the marketing campaign when "data synchronization" was executed in the past.

[0084] Step 3: Train the original data through the Prophet model to obtain the budget consumption forecast data and dynamic threshold of budget consumption in the future.

[0085] Specifically, this solution can combine budget detection with the Prophet time series prediction model to provide a budget detection method based on algorithm prediction; in this process, it is necessary to use training data to train the Prophet model; wherein, the training time includes training samples and sample labels, the training samples can be the original time series data of the activity budget, and the sample labels can be the actual budget consumption prediction data and the actual budget consumption dynamic threshold corresponding to the training samples.

[0086] After completing the model training, the Prophet model can be used to predict the original data to obtain the budget consumption forecast data and dynamic threshold of budget consumption within the future time range.

[0087] The budget consumption forecast data can be Figure 3 The Br in is the budget that the algorithm predicts will be consumed in the next 24 hours.

[0088] The dynamic threshold for budget consumption can be Figure 3 Dt in which Dt is the alarm threshold value recommended by the algorithm based on the consumption rate of the past budget; the dynamic threshold value Dt may be multiple, and each dynamic threshold value Dt corresponds to a different time range; Figure 3 Dt(24) refers to the dynamic threshold corresponding to 24 hours, Dt(48) refers to the dynamic threshold corresponding to 48 hours, and Dt(72) refers to the dynamic threshold corresponding to 72 hours.

[0089] in, Figure 3 Cb refers to the current actual remaining budget; Bcv refers to the hourly budget consumption rate.

[0090] Step 4: Offline data flows back online, and inspections are triggered through scheduling tasks to conduct budget risk analysis, risk classification processing, and early warning broadcasts.

[0091] Specifically, after obtaining the consumption forecast data and the budget consumption dynamic threshold through the offline data processing operations of the above steps 1 to 3, the consumption forecast data and the budget consumption dynamic threshold are returned (sent) to the online data processing module for activity budget detection and activity timeliness detection.

[0092] Among them, activity budget detection needs to trigger inspections through scheduling tasks, so as to detect the budget risk issues of each marketing activity (i.e. service activity) in turn. The specific implementation method is as follows: 1. Scheduling tasks: From multiple marketing activities, determine the target marketing activities that need budget testing (e.g. Figure 3 (“Multiple Prizes with One Coupon” event).

[0093] 2. Determine the current actual remaining budget Cb (i.e., target project resources) of the target marketing activity, the budget consumption forecast data Br (i.e., forecast resource usage data), multiple dynamic thresholds Dt (i.e., multiple forecast resource thresholds), and the bottom-line threshold (fixed percentage).

[0094] The bottom-line threshold may be preset, for example, 5%.

[0095] 3. Conduct budget risk analysis and risk grading.

[0096] Among them, the risk level of this solution (i.e., resource usage alarm level) is divided into three levels, namely P0, P1, and P2; among them, P0 has the highest risk level and P2 has the lowest risk level.

[0097] The specific methods for budget risk analysis and risk grading include the following four methods: (1)Br < Dt(24) Cb - 24 * Bcv < Dt(24) → P0 The meaning of the above content is: If Br < Dt(24) or Cb - 24 * Bcv < Dt(24), it is determined that there is a budget risk in the target marketing activity, and the risk level of this budget risk is P0 level.

[0098] (2)Cb - 24 * Bcv < Dt(48) → P1 The meaning of the above content is: If Cb - 24 * Bcv < Dt(48), it is determined that there is a budget risk in the target marketing activity, and the risk level of this budget risk is P1 level.

[0099] (3)Cb - 24 * Bcv < Dt(72) → P2 The meaning of the above content is: If Cb - 24 * Bcv < Dt(72), it is determined that there is a budget risk in the target marketing activity, and the risk level of this budget risk is P2 level.

[0100] (4)Remaining budget ratio < fallback threshold → P0 The meaning of the above content is: If the remaining budget ratio of the target marketing activity < fallback threshold (i.e., the minimum guarantee threshold), it is determined that there is a budget risk in the target marketing activity, and the risk level of this budget risk is P0 level.

[0101] It should be noted that if the current target marketing activity meets multiple risk levels (for example, Cb - 24 * Bcv can be less than Dt(24), Dt(48), Dt(72), so it meets P0, P1, P2), then it is necessary to select the highest risk level (P0) from multiple risk levels as the finally detected risk level (i.e., the resource usage alarm level corresponding to the selected prediction resource threshold).

[0102] Step Five: The alarm data is sent to the person in charge of the marketing activity through instant messaging software, and the work order is closed for processing.

[0103] Specifically, this solution sets different result sending strategies (notification strategies) for different risk levels; the risk levels can be P0, P1, P2; for P0, a notification strategy of notifying once every 30 minutes can be adopted; for P1, a notification strategy of notifying once every hour can be adopted; for P2, a notification strategy of notifying once a day can be adopted.

[0104] Based on this notification strategy, the alarm information is sent to the person in charge of the marketing activity through instant messaging software for early warning broadcast.

[0105] The person in charge of the marketing activity can perform closed-loop processing of the work order based on the alarm information.

[0106] In addition, it should be noted that this solution can also detect the timeliness of marketing activities. The specific methods are as follows: During the execution of a marketing campaign, one or more time-limited warning thresholds can be set for the marketing campaign, such as one day, three days, one week, etc. If the time difference between the current time and the campaign deadline of the marketing campaign is less than the time-limited warning forecast (such as one week), then a time-limited warning message can be generated and sent to the person in charge of the marketing campaign for warning.

[0107] Based on the above steps, it can be seen that the data processing method in the embodiment of this specification provides a detection scheme for activity timeliness and activity budget problems; specifically, this method processes activity configuration data and combines it with a time series prediction algorithm to provide a general detection scheme for marketing activity timeliness and marketing activity budget, and implements it in actual work to help ensure marketing activities. This scheme obtains the basic configuration of the activity through simple configuration / data cleaning, and implements timeliness detection of various configurations related to the activity (activities, tasks, prizes, people, coupons) based on experience values ​​and algorithm calculations. It can identify timeliness problems and complete the closed loop of "perception"-"evaluation"-"processing".

[0108] In practice, this solution provides a lightweight and practical tool to ensure timeliness and budget for marketing activities, which includes the following technical effects: 1) Full coverage: Supports marketing activities such as activity / task / prize / voucher templates and instant discounts, and supports multiple delivery platforms.

[0109] 2) High efficiency: Combine budget detection with the Prophet time series prediction model to provide an alarm threshold based on algorithm prediction and an alarm level based on the algorithm prediction results. The budget warning efficiency is 100%; 3) Timely alarm: Support instant messaging software alarm (batch alarm) and single-point push, achieving a current work order closed-loop rate of >80%; 4) Flexible configuration: supports configuration of inspection strategies in activity / prize / voucher / delivery / project dimensions to meet various configuration requirements; 5) Low access cost: One-stop inspection supports one-time configuration of project dimensions, automatically covering incremental activities and delivery.

[0110] In addition, this method provides targeted solutions to the above-mentioned defects such as "insufficient budget and high alarm noise", "no closed-loop alarm capability", and "scattered alarm channels": 1) Budget warning based on time series prediction algorithm: thus providing more accurate budget shortfall warning capability; 2) Inspection alarm notification-processing capability closed loop: to avoid the expansion of impact caused by known problems not being handled in time; 3) Unified warning configuration and alarm channels: Supports warning configuration in project / activity dimensions, and broadcasts warning information through a single-point push.

[0111] Corresponding to the above method embodiment, this specification also provides a data processing device embodiment, Figure 4 FIG. 1 is a schematic diagram showing the structure of a data processing device provided by an embodiment of the present specification. Figure 4 As shown, the device comprises: A data determination module 402 is configured to determine resource usage data of a target project; The resource prediction module 404 is configured to perform resource prediction on the resource usage data using a resource prediction model, and obtain a plurality of predicted resource thresholds corresponding to the target project, and predicted resource usage data corresponding to the target project within a target time range; The resource detection module 406 is configured to determine the target project resources of the target project, and perform resource detection according to the target project resources, the predicted resource usage data and the multiple predicted resource thresholds to determine the resource detection result corresponding to the target project.

[0112] Optionally, the resource usage data is resource usage time series data; The resource prediction module 404 is further configured to: The resource prediction model is used to perform resource prediction on the resource usage time series data to obtain multiple predicted resource thresholds corresponding to the target project and predicted resource usage data corresponding to the target project within the target time range, wherein each of the multiple predicted resource thresholds corresponds to a different preset time range.

[0113] Optionally, the resource detection module 406 is further configured to: Calculate according to the target project resources and the predicted resource usage data to obtain the remaining resource value of the project corresponding to the target time range; Determining a target predicted resource threshold by comparing the remaining resource value of the project with each predicted resource threshold, wherein the remaining resource value of the project is less than the target predicted resource threshold; The resource detection result corresponding to the target project is determined according to the target predicted resource threshold and the remaining resource value of the project.

[0114] Optionally, the resource detection result is resource warning information; The resource detection module 406 is further configured to: Determine a resource usage alarm level corresponding to the target predicted resource threshold; Based on the resource usage alarm level and the remaining resource value of the project, resource alarm information corresponding to the target project is generated.

[0115] Optionally, the target prediction resource threshold is multiple; The resource detection module 406 is further configured to: Determine a resource usage alarm level corresponding to each target prediction resource threshold among multiple target prediction resource thresholds; Determine, according to the resource usage alarm level, a selected predicted resource threshold from the multiple target predicted resource thresholds, wherein the resource usage alarm level of the selected predicted resource threshold is greater than or equal to the resource usage alarm levels of other predicted resource thresholds in the multiple target predicted resource thresholds; The resource detection result corresponding to the target project is determined according to the selected predicted resource threshold and the remaining resource value of the project.

[0116] Optionally, the data processing device further includes a result sending module, which is configured to: Determine unit information of a detection result receiving unit corresponding to the target item; The resource detection result is sent to the detection result receiving unit according to the unit information.

[0117] Optionally, the result sending module is further configured to: Determining a target resource usage alarm level corresponding to the resource detection result, wherein the target resource usage alarm level is used to generate the resource detection result; Determine a result sending strategy corresponding to the target resource usage alarm level; The resource detection result is sent to the detection result receiving unit according to the unit information by utilizing the result sending strategy.

[0118] Optionally, the target items are target items of multiple different types; The data determination module 402 is further configured to: Determine resource usage data to be processed corresponding to multiple types of service items, and perform data preprocessing on the resource usage data to be processed to obtain resource usage data of various types of service items; A target project is determined from the various types of service projects, and resource usage data of the target project is determined from the resource usage data, wherein the target project is any type of service project among the multiple types of service projects.

[0119] Optionally, the data processing device further includes a time detection module configured to: Determine the project execution end time of the target project and the preset time threshold corresponding to the target project; Calculate the time difference between the current time and the end time of the project execution; In the case where the time difference is less than the preset time threshold, generating a time detection result corresponding to the target item based on the time difference; The unit information of the detection result receiving unit corresponding to the target item is determined, and the time detection result is sent to the detection result receiving unit according to the unit information.

[0120] One or more embodiments of the present specification provide a data processing device, which can obtain the resource usage data of the target project for the target project resources during the process of performing resource detection on the target project, and use the resource prediction model to perform resource prediction on the resource usage data, so as to efficiently and accurately obtain multiple predicted resource thresholds corresponding to the target project, and the predicted resource usage data corresponding to the target project within the target time range; then, perform resource detection based on the target project resources, predicted resource usage data and multiple predicted resource thresholds of the target project, so as to accurately determine the resource detection result corresponding to the target project, and the target project can determine the usage of the target project resources through the resource detection result; avoid the problem of being unable to accurately detect resource usage; and avoid the problem that the execution of the service project will have a greater negative impact due to reasons such as excessive resource consumption and uneven resource usage due to the inability to accurately detect resource usage.

[0121] The above is a schematic scheme of a data processing device of this embodiment. It should be noted that the technical scheme of the data processing device and the technical scheme of the above data processing method belong to the same concept, and the details of the technical scheme of the data processing device that are not described in detail can be referred to the description of the technical scheme of the above data processing method.

[0122] Figure 5 The structure block diagram of a computing device 500 provided according to an embodiment of the present specification is shown. The components of the computing device 500 include but are not limited to a memory 510 and a processor 520. The processor 520 is connected to the memory 510 via a bus 530, and the database 550 is used to store data.

[0123] The computing device 500 also includes an access device 540 that enables the computing device 500 to communicate via one or more networks 560. Examples of these networks include a public switched telephone network (PSTN), a local area network (LAN), a wide area network (WAN), a personal area network (PAN), or a combination of communication networks such as the Internet. The access device 540 may include one or more of any type of network interface (e.g., a network interface card (NIC)) that is wired or wireless, such as an IEEE 802.11 wireless local area network (WLAN) wireless interface, a world-wide interoperability for microwave access (Wi-MAX) interface, an Ethernet interface, a universal serial bus (USB) interface, a cellular network interface, a Bluetooth interface, and a near field communication (NFC).

[0124] In one embodiment of the present specification, the above components of the computing device 500 and Figure 5 Other components not shown in the figure may also be connected to each other, for example, via a bus. It should be understood that Figure 5 The computing device structure block diagram shown is only for the purpose of illustration, and is not intended to limit the scope of this specification. Those skilled in the art can add or replace other components as needed.

[0125] The computing device 500 may be any type of stationary or mobile computing device, including a mobile computer or mobile computing device (e.g., a tablet computer, a personal digital assistant, a laptop computer, a notebook computer, a netbook, etc.), a mobile phone (e.g., a smart phone), a wearable computing device (e.g., a smart watch, smart glasses, etc.), or other types of mobile devices, or a stationary computing device such as a desktop computer or a personal computer (PC). The computing device 500 may also be a mobile or stationary server.

[0126] The processor 520 is used to execute the following computer program / instructions, which implement the steps of the above-mentioned data processing method when executed by the processor.

[0127] Each embodiment in this specification is described in a progressive manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the computing device embodiment, since it is basically similar to the data processing method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the data processing method embodiment.

[0128] An embodiment of the present specification further provides a computer-readable storage medium storing a computer program / instruction, which implements the steps of the above-mentioned data processing method when executed by a processor.

[0129] Each embodiment in this specification is described in a progressive manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the computer-readable storage medium embodiment, since it is basically similar to the data processing method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the data processing method embodiment.

[0130] An embodiment of the present specification also provides a computer program product, including a computer program / instruction, which implements the steps of the above data processing method when executed by a processor.

[0131] The above is a schematic scheme of a computer program product of this embodiment. It should be noted that the technical scheme of the computer program product and the technical scheme of the above data processing method belong to the same concept, and the details not described in detail in the technical scheme of the computer program product can be referred to the description of the technical scheme of the above data processing method.

[0132] The above is a description of a specific embodiment of the specification. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recorded in the claims can be performed in an order different from that in the embodiments and still achieve the desired results. In addition, the processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0133] The computer instructions include computer program codes, which may be in source code form, object code form, executable files or some intermediate forms, etc. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, USB flash drive, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electric carrier signal, telecommunication signal and software distribution medium, etc. It should be noted that the content contained in the computer-readable medium may be appropriately increased or decreased according to the requirements of patent practice. For example, in some regions, according to patent practice, computer-readable media do not include electric carrier signals and telecommunication signals.

[0134] It should be noted that, for the above-mentioned method embodiments, for the sake of simplicity of description, they are all expressed as a series of action combinations, but those skilled in the art should be aware that the embodiments of this specification are not limited by the order of the actions described, because according to the embodiments of this specification, some steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required by the embodiments of this specification.

[0135] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0136] The preferred embodiments of this specification disclosed above are only used to help explain this specification. The optional embodiments do not describe all the details in detail, nor do they limit the invention to only the specific implementation methods described. Obviously, many modifications and changes can be made according to the content of the embodiments of this specification. This specification selects and specifically describes these embodiments in order to better explain the principles and practical applications of the embodiments of this specification, so that technicians in the relevant technical field can well understand and use this specification. This specification is only limited by the claims and their full scope and equivalents.

Claims

1. A data processing method, comprising: Determine resource usage data for target projects; Perform resource prediction on the resource usage data using a resource prediction model to obtain a plurality of predicted resource thresholds corresponding to the target project and predicted resource usage data corresponding to the target project within a target time range; The target project resources of the target project are determined, and resource detection is performed according to the target project resources, the predicted resource usage data, and the multiple predicted resource thresholds to determine a resource detection result corresponding to the target project.

2. The data processing method according to claim 1, wherein the resource usage data is resource usage time series data; Using a resource prediction model to perform resource prediction on the resource usage data, obtaining a plurality of predicted resource thresholds corresponding to the target project, and predicted resource usage data corresponding to the target project within a target time range, including: The resource prediction model is used to perform resource prediction on the resource usage time series data to obtain multiple predicted resource thresholds corresponding to the target project and predicted resource usage data corresponding to the target project within the target time range, wherein each of the multiple predicted resource thresholds corresponds to a different preset time range.

3. The data processing method according to claim 1, wherein the performing resource detection according to the target project resources, the predicted resource usage data and the plurality of predicted resource thresholds to determine the resource detection result corresponding to the target project comprises: Calculate according to the target project resources and the predicted resource usage data to obtain the remaining resource value of the project corresponding to the target time range; Determining a target predicted resource threshold by comparing the remaining resource value of the project with each predicted resource threshold, wherein the remaining resource value of the project is less than the target predicted resource threshold; The resource detection result corresponding to the target project is determined according to the target predicted resource threshold and the remaining resource value of the project.

4. The data processing method according to claim 3, wherein the resource detection result is resource warning information; The determining, according to the target predicted resource threshold and the remaining resource value of the project, a resource detection result corresponding to the target project includes: Determine a resource usage alarm level corresponding to the target predicted resource threshold; Based on the resource usage alarm level and the remaining resource value of the project, resource alarm information corresponding to the target project is generated.

5. The data processing method according to any one of claims 3 or 4, wherein the target prediction resource threshold is multiple; The determining, according to the target predicted resource threshold and the remaining resource value of the project, a resource detection result corresponding to the target project includes: Determine a resource usage alarm level corresponding to each target prediction resource threshold among multiple target prediction resource thresholds; Determine, according to the resource usage alarm level, a selected predicted resource threshold from the multiple target predicted resource thresholds, wherein the resource usage alarm level of the selected predicted resource threshold is greater than or equal to the resource usage alarm levels of other predicted resource thresholds in the multiple target predicted resource thresholds; The resource detection result corresponding to the target project is determined according to the selected predicted resource threshold and the remaining resource value of the project.

6. The data processing method according to any one of claims 1 to 4, wherein after performing resource detection according to the target project resources, the predicted resource usage data and the multiple predicted resource thresholds and determining the resource detection result corresponding to the target project, the method further comprises: Determine unit information of a detection result receiving unit corresponding to the target item; The resource detection result is sent to the detection result receiving unit according to the unit information.

7. The data processing method according to claim 6, wherein sending the resource detection result to the detection result receiving unit according to the unit information comprises: Determining a target resource usage alarm level corresponding to the resource detection result, wherein the target resource usage alarm level is used to generate the resource detection result; Determine a result sending strategy corresponding to the target resource usage alarm level; The resource detection result is sent to the detection result receiving unit according to the unit information by utilizing the result sending strategy.

8. The data processing method according to any one of claims 1 to 4, wherein determining the resource usage data of the target project comprises: Determine resource usage data to be processed corresponding to multiple types of service items, and perform data preprocessing on the resource usage data to be processed to obtain resource usage data of various types of service items; A target project is determined from the various types of service projects, and resource usage data of the target project is determined from the resource usage data, wherein the target project is any type of service project among the multiple types of service projects.

9. The data processing method according to any one of claims 1 to 4, further comprising: Determine the project execution end time of the target project and the preset time threshold corresponding to the target project; Calculate the time difference between the current time and the end time of the project execution; In the case where the time difference is less than the preset time threshold, generating a time detection result corresponding to the target item based on the time difference; The unit information of the detection result receiving unit corresponding to the target item is determined, and the time detection result is sent to the detection result receiving unit according to the unit information.

10. A computing device comprising: Memory and processor; The memory is used to store computer programs / instructions, and the processor is used to execute the computer programs / instructions. When the computer program / instructions are executed by the processor, the steps of the method described in any one of claims 1 to 9 are implemented.

11. A computer-readable storage medium storing a computer program / instruction, wherein the computer program / instruction, when executed by a processor, implements the steps of the method according to any one of claims 1 to 9.

12. A computer program product, comprising a computer program / instruction, which, when executed by a processor, implements the steps of the method according to any one of claims 1 to 9.

Images