Method and system for predicting total number of active days of user based on step-by-step supervised learning

Abstract

The invention discloses a method and a system for predicting the total number of active days of a user based on step-by-step supervised learning, and the method comprises the steps: processing channel data in a set time window in a monthly grouping manner, and extracting a multi-dimensional feature set; performing initial training and optimization by using a supervised machine learning model; gradually expanding a training sample through an iterative prediction mechanism, and predicting a TAD value of a shortened period in stages until the whole stage is covered; and finally, a prediction model is constructed based on the complete data set, and accurate prediction of newly added channel data is realized. According to the method, recent data is utilized as a sample to the greatest extent in a distributed prediction mode, and the current state of a product and the active behavior mode of the user are reflected more accurately, so that the prediction accuracy is remarkably improved. Besides, a step-by-step prediction strategy enables the model to gradually adapt to the change trend of data, and high prediction performance can still be kept when product iteration is frequent and a channel structure changes greatly.

Description

Technical Field

[0001] The present invention relates to the field of machine learning technology, and in particular to a method and system for predicting total active days of users based on step-by-step supervised learning. Background Art

[0002] In the current Internet and mobile Internet fields, user behavior analysis is crucial for product optimization, marketing and user experience improvement. Total Active Days per User (TAD) is one of the important indicators to measure product user stickiness and activity. It refers to the total active days per user, that is, the average number of days each user uses an application during the statistical period. If the statistical period is more than one year, the total active days per user can basically reflect the number of days the user used the application before churn. For operation teams and product developers, accurately predicting the total active days of users can help formulate more effective operation strategies, optimize product functions and improve user experience.

[0003] However, traditional supervised TAD prediction methods face many challenges in practical applications. Traditional methods usually require the use of historical data from a longer period of time as samples, such as data from 360 days ago, for model training and prediction. However, this approach has several significant problems. First, over time, the product itself may have undergone multiple rounds of iterations, and the functions and interfaces have undergone major changes, which has led to changes in user behavior patterns and activity. Therefore, using earlier historical data as samples may not accurately reflect the actual situation of current users, resulting in deviations in prediction results.

[0004] Secondly, as the market environment changes, the channel type and structure of the product may also change significantly. These changes will also affect user acquisition and activity, and thus affect the TAD prediction results.

[0005] Furthermore, the TAD of the overall user base may also be significantly different from that of a year ago. This may be due to a variety of factors such as changes in user groups, evolution of market trends, and strategic adjustments by competitors. Therefore, using data from a long period of time as a sample for prediction may not be able to adapt to these changes, resulting in inaccurate prediction results.

[0006] In view of the above problems, there is an urgent need for a more accurate and efficient TAD prediction method to make full use of recent data that better reflects the product status as samples for prediction. This application proposes a user total active days prediction method and system based on step-by-step supervised learning, aiming to solve the problems existing in traditional methods and improve the accuracy of prediction. Summary of the invention

[0007] The purpose of the present invention is to provide a method and system for predicting the total active days of users based on step-by-step supervised learning, aiming to maximize the use of recent data that can better reflect the current status of the product as samples for prediction through distributed prediction, so as to effectively improve the accuracy of the prediction.

[0008] To achieve the above object, the present invention adopts the following technical solutions: In a first aspect, the present application provides a method for predicting the total number of active days of a user based on step-by-step supervised learning, comprising the following steps: S1. Group the channel data added within a set time window before the target forecast date (for example, a number of months before the target forecast date) by month, and extract a multidimensional feature set, including channel attributes, time attributes, TAD values ​​of each stage, and TAD ratios of each stage; S2, taking the longest cycle TAD as the target variable and the other features as input features, the supervised machine learning model is trained to obtain the initial prediction model, and the model parameters are optimized based on the prediction error rate indicator; S3. Adopt an iterative prediction mechanism, gradually expand the training samples to the data set containing historical prediction results, and predict the TAD value of the shortened period in stages until the full-stage TAD of the target prediction period is covered; the specific steps include: S3.1. When making the first prediction, use the initial sample data within the set time window to predict the longest period TAD and generate the first set of prediction results; S3.2, in each subsequent iteration, the prediction results of the previous step are incorporated into the training samples to expand the training data set; S3.3. Based on the expanded data set, gradually shorten the target TAD period and add multi-stage TAD predictions. The target period of each iteration is reduced by a set number of days, and at least one TAD prediction with a shortened period is added. S3.4, repeat steps S3.2-S3.3 until the training data covers all shortened stages TAD of the target prediction cycle; S4. Perform final modeling based on the complete cycle data set, extract the TAD value and TAD ratio of all stages, and train to generate the final prediction model; S5. Input the new channel data into the final prediction model and output the target period TAD prediction result.

[0009] In a preferred embodiment, in step S1, the channel attributes include a unique ID for identifying different channels, the channel type, and the channel addition amount; The time attribute indicates the date of the data record and identifies whether the date is a weekend or a holiday; The TAD value of each stage refers to the total number of active days in different time periods starting from the date the user is added, including but not limited to 30-day, 60-day, 90-day to 360-day TAD; The TAD ratio of each stage refers to the ratio between the TAD values ​​of different stages, including the ratio of TAD of adjacent periods and the combination of TAD ratios of the initial period and subsequent periods (such as 30 days / 60 days, 30 days / 90 days to 30 days / 360 days, etc.), which is used to reflect the changing trend of user activity.

[0010] In a preferred embodiment, in step S2, the supervised machine learning model is a LightGBM model or an XGBoost model.

[0011] In a preferred embodiment, in step S2, the prediction error rate indicator is the mean absolute percentage error MAPE.

[0012] In a preferred embodiment, in step S3.3, the target period is reduced by 30 days each time, and the number of TAD predictions for the shortened period added in each iteration increases by one each time.

[0013] In a preferred embodiment, in step S4, the full-stage TAD value includes the number of active days in the full cycle of 360 days from the date the user is added, and the TAD ratio includes the TAD growth ratio of consecutive dates.

[0014] In a preferred embodiment, in step S5, the newly added channel data must meet the data accumulation condition of at least 3 days before it can be input into the final prediction model for target period TAD prediction.

[0015] In a second aspect, the present application provides a user total active days prediction system based on step-by-step supervised learning, which is used to implement the user total active days prediction system based on step-by-step supervised learning as described in any one of the first aspects, and the system includes the following modules: A data preprocessing module is used to group the channel data newly added within a set time window before the target prediction date by month, and extract a multidimensional feature set, wherein the multidimensional feature set includes channel attributes, time attributes, TAD values ​​of each stage, and TAD ratios of each stage; The initial modeling module is used to train the longest cycle TAD as the target variable and the remaining features as input features through a supervised machine learning model to generate an initial prediction model and optimize the model parameters based on the prediction error rate indicator; An iterative prediction module is used to execute an iterative prediction mechanism, gradually expand the training samples to a data set containing historical prediction results, and predict the TAD value of the shortened period in stages until the full-stage TAD of the target prediction period is covered; the iterative prediction module includes: An initial prediction unit, used to predict the longest period TAD using the initial sample data within a set time window during the first prediction, and generate a first set of prediction results; The data expansion unit is used to incorporate the prediction results of the previous step into the training samples in each subsequent iteration to expand the training data set; A cycle adjustment unit, which is used to gradually shorten the target TAD cycle based on the expanded data set and add multi-stage TAD predictions, wherein the target cycle of each iteration is reduced by a set number of days and at least one TAD prediction with a shortened cycle is added; A loop control unit, used for repeatedly calling the data expansion unit and the cycle adjustment unit until the training data covers all shortened stages TAD of the target prediction cycle; The final modeling module is used to perform final modeling based on the complete cycle data set, extract the TAD value and TAD ratio of the entire stage, and train and generate the final prediction model; The prediction output module is used to input the final prediction model into the newly added channel data and output the target period TAD prediction result.

[0016] In a preferred embodiment, in the data preprocessing module: The channel attributes include a unique ID for identifying different channels, the channel type, and the amount of new channels added; The time attributes include the date of the data record and whether it is a weekend or holiday; The TAD value of each stage includes the total number of active days in different time periods starting from the date the user was added, including but not limited to 30-day, 60-day, 90-day to 360-day TAD; The TAD ratios of each stage include the ratios of TADs of adjacent periods and the combination of TAD ratios of an initial period and subsequent periods, which are used to reflect the changing trend of user activity.

[0017] In a preferred embodiment, in the initial modeling module, the supervised machine learning model is a LightGBM model or an XGBoost model.

[0018] In a preferred embodiment, in the initial modeling module, the prediction error rate indicator is the mean absolute percentage error MAPE.

[0019] In a preferred embodiment, in the cycle adjustment unit, the target cycle is reduced by 30 days each time, and the number of TAD predictions for the shortened cycle added in each iteration increases by one each time.

[0020] In a preferred embodiment, in the final modeling module, the full-stage TAD value includes the number of active days in the 360-day full cycle from the date the user is added, and the TAD ratio includes the TAD growth ratio of consecutive dates.

[0021] In a preferred embodiment, in the prediction output module, the newly added channel data must meet the data accumulation condition of at least 3 days before it can be input into the final prediction model for target period TAD prediction.

[0022] Compared with the prior art, the technical solution of the present invention has the following beneficial effects: The present invention provides a method and system for predicting the total active days of users based on step-by-step supervised learning, the method comprising: processing channel data within a set time window by month and extracting a multidimensional feature set; using a supervised machine learning model for initial training and optimization; gradually expanding the training samples through an iterative prediction mechanism, predicting the TAD value of the shortened cycle in stages until all stages are covered; finally, building a prediction model based on a complete data set to achieve accurate prediction of newly added channel data. The present invention maximizes the use of recent data as samples through distributed prediction, more accurately reflecting the current status of the product and the active behavior patterns of users, thereby significantly improving the accuracy of the prediction. In addition, the step-by-step prediction strategy enables the model to gradually adapt to the changing trend of the data, and can still maintain a high prediction performance when product iterations are frequent and the channel structure undergoes major changes. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] In order to more clearly illustrate the specific implementation methods of the present invention or the technical solutions in the prior art, the drawings required for use in the specific implementation methods or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some implementation methods of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0024] Figure 1 It is a flow chart of a method for predicting the total number of active days of a user based on step-by-step supervised learning of the present invention; Figure 2 This is a user active days (TAD) stage-by-stage prediction data table shown in a preferred embodiment of the present invention. DETAILED DESCRIPTION

[0025] In order to make the above and other features and advantages of the present invention more clear, the present invention is further described below in conjunction with the accompanying drawings. It should be understood that the specific embodiments given herein are for the purpose of explaining to those skilled in the art and are only exemplary and not restrictive.

[0026] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" and "second" may explicitly or implicitly include at least one of the features. In the description of the present invention, the meaning of "plurality" is at least two, such as two, three, etc., unless otherwise clearly and specifically defined.

[0027] Embodiment 1: Referring to Figure 1, this embodiment provides a method for predicting the total number of active days of a user based on step-by-step supervised learning, which specifically includes the following steps: Step S1, group the channel data with newly added time within a set time window before the target prediction date (for example, several months before the target prediction date) by month, and extract a multidimensional feature set, including channel attributes, time attributes, TAD values ​​of each stage, and TAD ratios of each stage.

[0028] The channel attributes include a unique ID for identifying different channels, the channel type, and the amount of new channels added; The time attribute indicates the date of the data record and identifies whether the date is a weekend or a holiday; The TAD value of each stage refers to the total number of active days in different time periods starting from the date the user is added, including but not limited to 30-day, 60-day, 90-day to 360-day TAD; The TAD ratio of each stage refers to the ratio between the TAD values ​​of different stages, including the ratio of TAD of adjacent periods and the combination of TAD ratios of the initial period and subsequent periods (such as 30 days / 60 days, 30 days / 90 days to 30 days / 360 days, etc.), which is used to reflect the changing trend of user activity.

[0029] Step S2: Take the longest period TAD (such as 360-day TAD) as the target variable and the remaining features as input features, train through a supervised machine learning model (such as the LightGBM model or the XGBoost model) to obtain the initial prediction model, and optimize the model parameters based on the prediction error rate indicator (such as the mean absolute percentage error MAPE).

[0030] Step S3: adopt an iterative prediction mechanism to gradually expand the training samples to a data set containing historical prediction results, and predict the TAD value of the shortened period in stages until the full-stage TAD of the target prediction period is covered; the specific steps include: Step S3.1: During the first prediction, the longest period TAD is predicted using the initial sample data within the set time window, and the first set of prediction results is generated; Step S3.2, in each subsequent iteration, the prediction results of the previous step are incorporated into the training samples to expand the training data set; Step S3.3: Based on the expanded data set, gradually shorten the target TAD period (e.g., 330 days, 300 days), and add multi-stage TAD predictions. The target period of each iteration is reduced by a set number of days (e.g., 30 days), and at least one TAD prediction with a shortened period is added.

[0031] Step S3.4, repeat steps S3.2-S3.3 until the training data covers all shortened stages TAD of the target prediction cycle.

[0032] Step S4: Perform final modeling based on the complete cycle data set, extract the full-stage TAD value and TAD ratio, and train to generate the final prediction model. The full-stage TAD value includes the 360-day full-cycle active days from the date the user is added, and the TAD ratio includes the TAD growth ratio of consecutive days.

[0033] Step S5: Input the newly added channel data into the final prediction model and output the target period TAD prediction result. The newly added channel data must meet the data accumulation condition of at least 3 days before it can be input into the final prediction model for target period TAD prediction.

[0034] In order to more clearly illustrate the technical solution of the present application, a specific example is provided below for illustration.

[0035] Assume that the current date is January 1, 2025, and the total active days (TAD) of users in each stage throughout 2024 needs to be predicted. This embodiment uses a step-by-step supervised learning method and iterative modeling using recent data to improve the 360-day TAD prediction accuracy.

[0036] Step 1: Sample data preparation.

[0037] Time window setting: Select the channel new data added from October 2023 to the present as the sample, and group them by month (a total of 15 months). Figure 2 Data table shown.

[0038] Step 2: Predict the 360-day TAD under the January 2024 data grouping.

[0039] 2.1) Using the data from October 2023 to December 2023 as samples, the following indicators are processed and calculated: channel ID, channel type, channel additions, whether the current date is a weekend, whether the current date is a holiday, TAD value of each stage (30 days, 60 days, 90 days...360 days), TAD ratio of each stage (such as 30 days / 60 days, 30 days / 90 days...30 days / 330 days).

[0040] 2.2) After obtaining the above indicators, 360-day TAD (the longest period) is used as the target variable, and the other indicators are used as feature variables. LightGBM (lgb for short), XGBoost (xgb for short) and other supervised machine learning models are called for training. By comparing the prediction error rates (MAPE) of different models, the model and parameters with the smallest error rate are selected as the optimal model.

[0041] 2.3) Use the optimal model for prediction: Apply the optimal model and parameters obtained in step 2.2), input the model features under the January 2024 sample, and predict the output of the 360-day TAD in January 2024.

[0042] Step 3: Predict the 330-day TAD and 360-day TAD under the data grouping in February 2024.

[0043] 3.1) Using the prediction results of step 2, taking the data from October 2023 to January 2024 as samples, calculate the following indicators: channel ID, channel type, channel additions, whether the current date is a weekend, whether the current date is a holiday, TAD value of each stage (30 days, 60 days, 90 days...360 days), TAD ratio of each stage (such as 30 days / 60 days, 30 days / 90 days...30 days / 330 days).

[0044] 3.2) After obtaining the above indicators, take 330-day TAD and 360-day TAD as target variables, and the other indicators as feature variables. Call supervised machine learning models such as lgb and xgb, and select the model and parameters with the smallest prediction error rate (MAPE).

[0045] 3.3) Apply step 3.2) to obtain the model and parameters, input the model features under the February 2024 sample, and predict the output of 330-day TAD and 360-day TAD in February 2024.

[0046] Step 4: Predict the 300-day TAD, 330-day TAD, and 360-day TAD under the March 2024 data grouping.

[0047] 4.1) Using the prediction results of step 3, with the data from October 2023 to February 2024 as the sample, calculate the following indicators: channel ID, channel type, channel additions, whether the current date is a weekend, whether the current date is a holiday, TAD value of each stage (30 days, 60 days, 90 days...360 days), TAD ratio of each stage (such as 30 days / 60 days, 30 days / 90 days...30 days / 270 days).

[0048] 4.2) After obtaining the above indicators, take 300-day TAD, 330-day TAD, and 360-day TAD as target variables, and the remaining indicators as feature variables. Call supervised machine learning models such as lgb and xgb, and select the model and parameters with the smallest prediction error rate (MAPE).

[0049] 4.3) Apply step 4.2) to obtain the model and parameters, input the model features under the March 2024 sample, and predict the output of 300-day TAD, 330-day TAD, and 360-day TAD in March 2024.

[0050] Step 5: Repeat the modeling and forecasting to make up for the TAD of each stage under the full-year data of 2024 (see Figure 2 yellow part of the data table).

[0051] Through rolling forecasting, the TAD forecast values ​​for each month in 2024 at different stages (such as 30 days, 60 days...360 days) will be gradually supplemented.

[0052] Step 6: After completion, proceed to final modeling.

[0053] 6.1) Using the data from October 2023 to December 2024 as samples (including the predicted TAD value), we process and calculate the following more comprehensive indicators: channel ID, channel type, channel additions, whether the current date is a weekend, whether the current date is a holiday, TAD values ​​of each stage (1 day, 2 days, 3 days...360 days) and TAD ratios of each stage (such as 2 days / 1 day, 3 days / 1 day, etc.).

[0054] 6.2) After obtaining the above indicators, take 360-day TAD as the target variable and the other indicators as feature variables, call supervised machine learning models such as lgb and xgb, select the model and parameters with the smallest prediction error rate (MAPE), and train to generate the final prediction model.

[0055] Step 7: Add forecasts for new channels.

[0056] Apply the final prediction model and parameters obtained in step 6. When the number of newly added channels reaches 3 days, input the model features (including channel ID, channel type, channel additions, whether the current date is a weekend, whether the current date is a holiday, and the calculated TAD value, etc.) to predict the 360-day TAD of the channel (see Figure 2 (see blue part of the data table).

[0057] In summary, this application effectively integrates recent data and prediction results through an iterative mechanism. Each iteration incorporates the latest prediction results into the training samples, solving the problem of historical data lag, so that the model is always optimized based on the current product status (such as the prediction results in January 2024 are used in the February model construction). This application gradually shortens the target prediction cycle (reduced by 30 days each time) and increases the prediction stage (such as 360 days for the first prediction, 330 days and 360 days for the second prediction), so that the model can simultaneously capture short-term and long-term user activity trends.

[0058] Compared with traditional methods, the technical solution of this embodiment significantly reduces the reliance on obsolete data. Traditional methods usually rely on historical data from 360 days ago, while the technical solution of this embodiment only requires 3 months of initial data (October to December 2023) to start the prediction, and then expands the sample by iteration to reduce the reliance on obsolete data. In the embodiment, the prediction for January 2024 only uses 3 months of data, which will be gradually expanded to the full year of data, and the resource usage is more reasonable. For new channels, only 3 days of data need to be accumulated, and the final constructed model can be used to predict the 360-day user active days (TAD), thereby meeting the real-time needs of the business.

[0059] Embodiment 2: Based on the same design concept, this embodiment also provides a user total active days prediction system based on step-by-step supervised learning, which is used to execute the user total active days prediction method based on step-by-step supervised learning described in Example 1.

[0060] Specifically, the system includes: a data preprocessing module, an initial modeling module, an iterative prediction module, a final modeling module and a prediction output module.

[0061] The data preprocessing module is used to group the channel data newly added within a set time window before the target prediction date by month and extract a multidimensional feature set, which includes channel attributes, time attributes, TAD values ​​of each stage and TAD ratios of each stage.

[0062] Among them, the channel attributes include a unique ID for identifying different channels, the channel type and the number of new channels added; the time attributes include the date of data record and whether it is a weekend or holiday; the TAD value of each stage includes the total active days in different time periods starting from the date the user was added, including but not limited to 30-day, 60-day, 90-day to 360-day TAD; the TAD ratio of each stage includes the ratio of TAD of adjacent periods and the combination of TAD ratios of the initial period and subsequent periods, which are used to reflect the changing trend of user activity.

[0063] The initial modeling module is used to use the longest period TAD as the target variable and the remaining features as input features, train through a supervised machine learning model, generate an initial prediction model, and optimize the model parameters based on the prediction error rate indicator. In a preferred embodiment, the supervised machine learning model is a LightGBM model or an XGBoost model; the prediction error rate indicator is the mean absolute percentage error MAPE.

[0064] The iterative prediction module is used to execute the iterative prediction mechanism, gradually expand the training samples to the data set containing historical prediction results, and predict the TAD value of the shortened period in stages until the full-stage TAD of the target prediction period is covered.

[0065] Specifically, the iterative prediction module includes: An initial prediction unit, used to predict the longest period TAD using the initial sample data within a set time window during the first prediction, and generate a first set of prediction results; The data expansion unit is used to incorporate the prediction results of the previous step into the training samples in each subsequent iteration to expand the training data set; A cycle adjustment unit is used to gradually shorten the target TAD cycle based on the expanded data set and add multi-stage TAD predictions. The target cycle of each iteration is reduced by a set number of days (such as 30 days), and at least one TAD prediction with a shortened cycle is added. A loop control unit is used to repeatedly call the data expansion unit and the cycle adjustment unit until the training data covers all shortened stages TAD of the target prediction cycle.

[0066] The final modeling module is used to perform final modeling based on the complete cycle data set, extract the full-stage TAD value and TAD ratio, and train to generate the final prediction model. The full-stage TAD value includes the 360-day full-cycle active days from the date the user is added, and the TAD ratio includes the TAD growth ratio of consecutive dates.

[0067] The prediction output module is used to input the final prediction model into the newly added channel data and output the target period TAD prediction result. The newly added channel data must meet the data accumulation condition of at least 3 days before it can be input into the final prediction model for target period TAD prediction.

[0068] It can be understood that the modules recorded in the user total active days prediction system based on step-by-step supervised learning are similar to Figure 1Therefore, the operations, features and beneficial effects described above for the method for predicting the total active days of users based on step-by-step supervised learning are also applicable to the system for predicting the total active days of users based on step-by-step supervised learning and the modules contained therein, and will not be repeated here.

[0069] The specific embodiments of the present invention are described in detail above, but they are only examples, and the present invention is not limited to the specific embodiments described above. For those skilled in the art, any equivalent modifications and substitutions made to the present invention are also within the scope of the present invention. Therefore, the equalization changes and modifications made without departing from the spirit and scope of the present invention should be included in the scope of the present invention.

Claims

1. A method for predicting the total number of active days of users based on step-by-step supervised learning, characterized in that: The steps include: S1. Group the channel data added within the set time window before the target forecast date by month, and extract the multidimensional feature set, including channel attributes, time attributes, TAD values ​​of each stage, and TAD ratios of each stage; S2, taking the longest cycle TAD as the target variable and the other features as input features, the supervised machine learning model is trained to obtain the initial prediction model, and the model parameters are optimized based on the prediction error rate indicator; S3, adopting an iterative prediction mechanism, gradually expanding the training samples to the data set containing historical prediction results, and predicting the TAD value of the shortened period in stages until the full-stage TAD of the target prediction period is covered; The specific steps include: S3.

1. When making the first prediction, use the initial sample data within the set time window to predict the longest period TAD and generate the first set of prediction results; S3.2, in each subsequent iteration, the prediction results of the previous step are incorporated into the training samples to expand the training data set; S3.

3. Based on the expanded data set, gradually shorten the target TAD period and add multi-stage TAD predictions. The target period of each iteration is reduced by a set number of days, and at least one TAD prediction with a shortened period is added. S3.4, repeat steps S3.2-S3.3 until the training data covers all shortened stages TAD of the target prediction cycle; S4. Perform final modeling based on the complete cycle data set, extract the TAD value and TAD ratio of all stages, and train to generate the final prediction model; S5. Input the new channel data into the final prediction model and output the target period TAD prediction result.

2. The method for predicting total active days of users based on step-by-step supervised learning according to claim 1, characterized in that: In step S1, the channel attributes include a unique ID for identifying different channels, the channel type, and the amount of new channels added; The time attribute indicates the date of the data record and identifies whether the date is a weekend or a holiday; The TAD value of each stage refers to the total number of active days in different time periods starting from the date the user was added; The TAD ratio of each stage refers to the ratio between TAD values ​​in different stages, including the ratio of TADs of adjacent periods and the combination of TAD ratios of an initial period and subsequent periods, and is used to reflect the changing trend of user activity.

3. The method for predicting total active days of users based on step-by-step supervised learning according to claim 1, characterized in that: In step S2, the supervised machine learning model is a LightGBM model or an XGBoost model.

4. The method for predicting total active days of users based on step-by-step supervised learning according to claim 1, characterized in that: In step S2, the prediction error rate indicator is the mean absolute percentage error MAPE.

5. The method for predicting total active days of users based on step-by-step supervised learning according to claim 1, characterized in that: In step S3.3, the target period is reduced by 30 days each time, and the number of TAD predictions for the shortened period added in each iteration increases by one each time.

6. The method for predicting total active days of users based on step-by-step supervised learning according to claim 1, characterized in that: In step S4, the full-stage TAD value includes the number of active days in the 360-day full cycle from the date the user is added, and the TAD ratio includes the TAD growth ratio of consecutive dates.

7. The method for predicting total active days of users based on step-by-step supervised learning according to claim 1, characterized in that: In step S5, the newly added channel data must meet the data accumulation condition of at least 3 days before it can be input into the final prediction model for target period TAD prediction.

8. A user total active days prediction system based on step-by-step supervised learning, characterized in that: Includes the following modules: A data preprocessing module is used to group the channel data newly added within a set time window before the target prediction date by month, and extract a multidimensional feature set, wherein the multidimensional feature set includes channel attributes, time attributes, TAD values ​​of each stage, and TAD ratios of each stage; The initial modeling module is used to train the longest cycle TAD as the target variable and the remaining features as input features through a supervised machine learning model to generate an initial prediction model and optimize the model parameters based on the prediction error rate indicator; Iterative prediction module, used to execute the iterative prediction mechanism, gradually expand the training samples to the data set containing historical prediction results, and predict the TAD value of the shortened period in stages until the full-stage TAD of the target prediction period is covered; The iterative prediction module comprises: An initial prediction unit, used to predict the longest period TAD using the initial sample data within a set time window during the first prediction, and generate a first set of prediction results; The data expansion unit is used to incorporate the prediction results of the previous step into the training samples in each subsequent iteration to expand the training data set; A cycle adjustment unit, which is used to gradually shorten the target TAD cycle based on the expanded data set and add multi-stage TAD predictions, wherein the target cycle of each iteration is reduced by a set number of days and at least one TAD prediction with a shortened cycle is added; A loop control unit, used for repeatedly calling the data expansion unit and the cycle adjustment unit until the training data covers all shortened stages TAD of the target prediction cycle; The final modeling module is used to perform final modeling based on the complete cycle data set, extract the TAD value and TAD ratio of the entire stage, and train and generate the final prediction model; The prediction output module is used to input the final prediction model into the newly added channel data and output the target period TAD prediction result.

9. The user total active days prediction system based on step-by-step supervised learning according to claim 8, characterized in that: In the data preprocessing module: The channel attributes include a unique ID for identifying different channels, the channel type, and the amount of new channels added; The time attributes include the date of the data record and whether it is a weekend or holiday; The TAD value of each stage includes the total number of active days in different time periods starting from the date when the user was added; The TAD ratios of each stage include the ratios of TADs of adjacent periods and the combination of TAD ratios of an initial period and subsequent periods, which are used to reflect the changing trend of user activity.

10. The user total active days prediction system based on step-by-step supervised learning according to claim 8, characterized in that: In the final modeling module, the full-stage TAD value includes the 360-day full-cycle active days from the date the user is added, and the TAD ratio includes the TAD growth ratio of consecutive dates.

Images