An advertising management system based on big data

Abstract

The present invention relates to the field of data processing technology, and in particular to an advertising management system based on big data, the system comprising: an acquisition module, a prediction module, a first judgment module, a first determination module, a second judgment module, a second determination module, an adjustment module, and an output module. The present invention effectively determines advertising demand and determines advertising types by collecting key data of e-commerce platform advertisements in real time and predicting advertising delivery index thresholds; dynamically determines and grades investment risks based on real-time data, discovers potential risks and adjusts delivery index thresholds, and further optimizes advertising delivery effects; ultimately, provides targeted management suggestions, which helps e-commerce companies to rationally plan advertising budgets, enhance market competitiveness, achieve refined management of advertising delivery and maximize benefits, and effectively solve the problems of risk identification delays and optimization adjustment lags caused by over-reliance on static matching strategies and lagging feedback mechanisms.

Description

Technical Field

[0001] The present invention relates to the field of data processing technology, and in particular to an advertising management system based on big data. Background Art

[0002] In today's digital age, the advertising industry faces the dual challenges of massive data volumes and a complex market environment. With the increasing popularity of the internet and the widespread use of mobile devices, advertising channels are becoming increasingly diverse, and user behavior data is exploding. Advertisers and operators need to accurately grasp market dynamics within this massive data and adjust advertising strategies in real time to maximize advertising effectiveness. However, traditional advertising management systems often suffer from limited data processing capabilities, unpredictable advertising results, and delayed risk assessments, making them unable to meet the complex demands of modern advertising. Therefore, developing a highly adaptive advertising management system is particularly important.

[0003] Patent document CN112163901A discloses a big data-based advertising management system, comprising a central server and an advertising input module, an identity matching module, a data storage module, a classification and extraction module, an analysis and ranking module, a push statistics module, and a user feedback module connected to the central server. The central server is configured to receive and process data information from each module within the system. The advertising input module is configured for each merchant to provide advertising push data and send it to the central server. The identity matching module is configured to perform big data identity matching on each advertising push data with a user group and push the corresponding advertising push data to the user terminal according to a predetermined matching degree. The data storage module is configured to store advertising push data from all merchants and make it available for retrieval by the central server. The classification and extraction module is configured to separate advertising push data from all merchants and extract it according to different push frequencies. The analysis and ranking module is configured to sort all advertising push data according to the matching degree based on the identity matching data provided by the identity matching module. The push statistics module is configured to compile push statistics on all pushed advertising push data. The user feedback module is configured to provide feedback after receiving advertising push information and store the information for real-time review by staff.

[0004] It can be seen that the big data-based advertising management system has the following problems: excessive reliance on historical data matching and static push strategies, and inability to respond to changes in advertising effects in real time; lack of a prediction mechanism for advertising effects, relying only on matching ranking push, and unable to predict risks or optimize delivery strategies; merchants cannot perceive the actual ROI of advertising delivery (such as the problem of high investment and low conversion); mainly relying on user identity matching and advertising classification, ignoring the real-time effect data of the advertisement itself; relying on manual processing, and the feedback has no direct correlation with the adjustment of push strategy; the module is highly dependent on the central server, and manual intervention is required to adjust the strategy. Summary of the Invention

[0005] To this end, the present invention provides an advertising management system based on big data, which is used to overcome the problems of risk identification delay and optimization adjustment lag caused by over-reliance on static matching strategies and lagging feedback mechanisms in the prior art through real-time data monitoring and dynamic adjustment mechanisms.

[0006] To achieve the above objectives, the present invention provides an advertising management system based on big data, comprising:

[0007] The collection module is used to collect the real-time advertising input, real-time click volume, real-time order completion rate and real-time search frequency of the target on the e-commerce platform;

[0008] A prediction module, connected to the acquisition module, for predicting a delivery index threshold based on the real-time input amount, the real-time click volume, the real-time transaction rate, the real-time search times, and a preset big data prediction model;

[0009] a first determination module, connected to the acquisition module and the prediction module respectively, for determining whether there is advertising demand based on the real-time click volume, the real-time transaction rate, the real-time search times, and the delivery index threshold, and forming a first determination result;

[0010] a first determination module, connected to the acquisition module and the first determination module respectively, for determining an advertisement type according to the first determination result, the real-time transaction rate, and the real-time search times;

[0011] a second determination module, connected to the acquisition module and the first determination module respectively, for determining whether there is an investment risk based on the advertisement type, the real-time investment amount, and the real-time transaction rate, and forming a second determination result;

[0012] a second determination module, connected to the acquisition module and the second determination module respectively, for determining the risk level of the investment risk according to the second determination result, the real-time investment amount, and the real-time transaction rate;

[0013] an adjustment module, connected to the second determination module and the prediction module respectively, for adjusting the delivery index threshold according to the number of times the risk level is formed within a preset adjustment period to form an adjusted delivery index threshold;

[0014] An output module is connected to the adjustment module and is used to provide corresponding management prompts for the advertisement type corresponding to the risk level determined based on the adjustment delivery index threshold.

[0015] Furthermore, the first determination module includes:

[0016] A click volume fluctuation calculation unit, configured to calculate a standard deviation of the real-time click volume within a preset first determination time period to form a click volume fluctuation value;

[0017] a first transaction rate fluctuation calculation unit, configured to calculate a standard deviation of the real-time transaction rate within the preset first determination time period to form a first transaction rate fluctuation value;

[0018] A first search fluctuation calculation unit is used to calculate the standard deviation of the number of real-time searches within the preset first determination time period to form a first search fluctuation value;

[0019] The first judgment unit is respectively connected to the click volume fluctuation calculation unit, the first transaction rate fluctuation calculation unit and the first search fluctuation calculation unit, and is used to determine whether there is advertising demand based on the click volume fluctuation value, the first transaction rate fluctuation value, the first search fluctuation value and the delivery index threshold, and form a first judgment result.

[0020] Furthermore, the first determining unit includes:

[0021] a delivery index calculation subunit, configured to perform a weighted summation of the click volume fluctuation value, the first transaction rate fluctuation value, the first search fluctuation value, a preset click volume weight, a preset transaction rate weight, and a preset search number weight to form a delivery index;

[0022] The first determination subunit is connected to the delivery index calculation subunit and is used to determine that there is an advertisement demand when the delivery index is greater than the delivery index threshold, thereby forming a first determination result.

[0023] Furthermore, the first determining module includes:

[0024] A second transaction rate fluctuation calculation unit is used to calculate the standard deviation of the real-time transaction rate within a predetermined time period to form a second transaction rate fluctuation value;

[0025] A second search fluctuation calculation unit is used to calculate the standard deviation of the number of real-time searches within the preset determined time period to form a second search fluctuation value;

[0026] The first determining unit is connected to the second transaction rate fluctuation calculating unit and the second search fluctuation calculating unit respectively, and is used to determine the advertisement type according to the second transaction rate fluctuation value and the second search fluctuation value.

[0027] Furthermore, the first determining unit includes:

[0028] a transaction rate curve drawing subunit, configured to draw a change curve of the fluctuation value of the second transaction rate within the predetermined time period to form a transaction rate curve;

[0029] A search curve drawing subunit is used to draw a change curve of the second search fluctuation value within the preset determined time period to form a search curve;

[0030] a consistency calculation subunit, connected to the transaction rate curve drawing subunit and the search curve drawing subunit respectively, for calculating the cosine similarity between the transaction rate curve and the search curve to form a change consistency;

[0031] A first determination subunit is connected to the consistency calculation subunit, and is used to determine that the advertisement type is an active advertisement when the change consistency is greater than a preset consistency threshold; to determine that the advertisement type is a brand advertisement when the change consistency is less than or equal to the preset consistency threshold, and the second transaction rate fluctuation value is less than the preset second transaction rate fluctuation threshold, but the second search fluctuation value is greater than the preset second search fluctuation threshold; and to determine that the advertisement type is an effect advertisement when the change consistency is less than or equal to the preset consistency threshold, and the second transaction rate fluctuation value is greater than or equal to the preset second transaction rate fluctuation threshold, but the second search fluctuation value is less than or equal to the preset second search fluctuation threshold.

[0032] Furthermore, the second determination module includes:

[0033] A first input fluctuation value calculation unit is used to calculate the standard deviation of the real-time input amount within a preset second determination time period to form a first input fluctuation value;

[0034] a third transaction rate fluctuation value calculation unit, configured to calculate a standard deviation of the real-time transaction rate within the preset second determination time period to form a third transaction rate fluctuation value;

[0035] The second determination unit is connected to the first investment fluctuation value calculation unit and the third transaction rate fluctuation value calculation unit respectively, and is used to determine whether there is an investment risk according to the first investment fluctuation value and the third transaction rate fluctuation value to form a second determination result.

[0036] Furthermore, the second determining unit includes:

[0037] an investment comparison subunit, configured to compare the first investment fluctuation value with a preset investment fluctuation value threshold to form an investment comparison result;

[0038] a transaction rate comparison subunit, configured to compare the third transaction rate fluctuation value with a preset third transaction rate fluctuation threshold to form a transaction rate comparison result;

[0039] The second judgment sub-unit is connected to the investment comparison sub-unit and the transaction rate comparison sub-unit respectively, and is used to determine that there is an investment risk when the investment comparison result is that the first investment fluctuation value is greater than the preset investment fluctuation value threshold, or when the transaction rate comparison result is that the third transaction rate fluctuation value is less than the preset third transaction rate fluctuation threshold, to form a second judgment result.

[0040] Furthermore, the second determining module includes:

[0041] The high investment risk determination unit is used to determine that the investment risk is a high investment risk when the investment comparison result is that the first investment fluctuation value is greater than the preset investment fluctuation value threshold, and when the transaction rate comparison result is that the third transaction rate fluctuation value is less than the preset third transaction rate fluctuation threshold.

[0042] Furthermore, the adjustment module includes:

[0043] a quantity fluctuation value calculation unit, configured to calculate a standard deviation of the number of times the high investment risk occurs within the preset adjustment period to form a quantity fluctuation value;

[0044] An adjustment unit is connected to the quantity fluctuation value calculation unit and is used to adjust the delivery index threshold according to the quantity fluctuation value to form an adjusted delivery index threshold.

[0045] Furthermore, the adjustment unit includes:

[0046] a quantity fluctuation deviation calculation subunit, configured to calculate a relative deviation between the quantity fluctuation value and the preset quantity fluctuation threshold value when the quantity fluctuation value is greater than a preset quantity fluctuation threshold value, to form a quantity fluctuation deviation;

[0047] The adjustment subunit is connected to the quantity fluctuation deviation calculation subunit and is used to reduce the delivery index threshold according to the quantity fluctuation deviation and a preset adjustment coefficient to form an adjusted delivery index threshold when the quantity fluctuation deviation is greater than a preset quantity fluctuation deviation threshold.

[0048] Compared with the prior art, the beneficial effect of the present invention is that, by collecting key data of e-commerce platform advertisements in real time and combining it with a big data prediction model, it is possible to accurately predict the threshold of the advertising delivery index, thereby effectively judging the advertising demand and determining the type of advertisement. At the same time, by dynamically judging and grading the investment risk based on real-time data, it is possible to timely discover potential risks and adjust the delivery strategy. By adjusting the delivery index threshold, the advertising delivery effect is further optimized to ensure the scientificity and efficiency of advertising delivery. Ultimately, targeted management suggestions are provided for different risk levels and advertising types, which helps e-commerce companies to rationally plan their advertising budgets, improve the return on investment of advertising delivery, enhance market competitiveness, and achieve refined management and maximized benefits of advertising delivery, effectively solving the problems of risk identification delays and optimization adjustment delays caused by over-reliance on static matching strategies and lagging feedback mechanisms.

[0049] Furthermore, by calculating the click volume fluctuation value, the first conversion rate fluctuation value, and the first search fluctuation value within a preset first determination period, and combining this with the delivery index threshold for comprehensive determination, advertising demand can be accurately identified. This design captures fluctuations in key data during the advertising delivery process in real time, helping advertisers more accurately evaluate advertising effectiveness and adjust advertising strategies in a timely manner, thereby improving the accuracy and efficiency of advertising delivery.

[0050] Furthermore, by calculating the delivery index and determining advertising demand when it exceeds the delivery index threshold, the system comprehensively considers fluctuations in multi-dimensional data, avoiding the limitations of a single metric and enabling a more comprehensive and accurate identification of actual demand for advertising. This not only improves the accuracy of ad demand determination but also effectively reduces misjudgments, helping advertisers better optimize their advertising strategies, improve the efficiency and effectiveness of advertising, and ultimately enhance the scientific and economic benefits of advertising.

[0051] Furthermore, by calculating the second conversion rate fluctuation value and the second search fluctuation value within a preset time period to determine the ad type, the key data fluctuation characteristics of the ad delivery process can be accurately captured. Analysis based on data fluctuations can more accurately identify the performance differences between different ad types within a specific time period, helping advertisers select the most appropriate ad type based on fluctuations, optimize ad delivery strategies, improve the accuracy and effectiveness of ad delivery, and ultimately enhance the scientific and economic benefits of advertising.

[0052] Furthermore, by plotting the transaction rate curve and the search curve and calculating the cosine similarity of the two to obtain the change consistency, the advertising type can be accurately determined according to different conditions. When the change consistency is greater than the preset consistency threshold, it is determined to be an active advertisement, because the fluctuations in its transaction rate and search times are highly consistent, indicating that the advertising has a significant and synchronous impact on user behavior. When the change consistency is less than or equal to the preset consistency threshold, and the transaction rate fluctuates little but the search times fluctuate greatly, it is determined to be a brand advertisement, because brand advertisements focus more on enhancing brand awareness and prompting users to actively search. When the change consistency is less than or equal to the preset consistency threshold, and the transaction rate fluctuates greatly but the search times fluctuate little, it is determined to be an effect advertisement, because effect advertisements focus more on directly promoting sales. Analysis based on data fluctuation characteristics and correlation can effectively distinguish the effectiveness of different types of advertising, helping advertisers to choose advertising strategies more scientifically, thereby improving the accuracy and effectiveness of advertising.

[0053] Furthermore, by calculating the first investment fluctuation value and the third transaction rate fluctuation value within the preset second judgment period to determine the investment risk, it is possible to monitor the dynamic changes of investment and return during the advertising delivery process in real time, and can effectively identify situations where the investment fluctuation is large but the transaction rate fluctuation is small, and timely discover potential investment risks, helping advertisers make more informed decisions during the advertising delivery process, avoiding unnecessary waste of funds, and thus improving the economic benefits and return on investment of advertising.

[0054] Furthermore, by comparing the first investment fluctuation value with a preset investment fluctuation threshold, and the third closing rate fluctuation value with a preset third closing rate fluctuation threshold, potential investment risks during the advertising process can be accurately identified. When the first investment fluctuation value exceeds the threshold, it indicates significant fluctuations in advertising investment, potentially leading to the risk of uncontrolled costs. Conversely, when the third closing rate fluctuation value falls below the threshold, it indicates that advertising is not significantly boosting sales, potentially leading to a risk of insufficient return on investment. This dual comparison can provide timely warnings when investment is excessively high or returns are insufficient, helping advertisers adjust their advertising strategies and optimize advertising budget allocation in advance, effectively reducing investment risks and improving the economic benefits and return on investment of advertising.

[0055] Furthermore, through a multi-level risk assessment mechanism, the system can accurately identify high-investment risks based on different fluctuations. Advertisements with high investment risks may face the problem of large capital waste or unsatisfactory returns, requiring timely intervention to avoid greater losses. In contrast, advertisements with medium and low investment risks are usually in a more stable state, with their fluctuations within a controllable range and a relatively small impact on the overall delivery effect. Therefore, special prompts or interventions are not required for the time being. By focusing resources and attention on high-investment-risk ads, the system can more efficiently optimize advertising strategies and ensure that limited resources are prioritized for the ads that need the most attention, thereby improving the overall advertising effectiveness and return on investment.

[0056] Furthermore, by calculating the standard deviation of the number of high-risk events within a preset adjustment period, we can determine the quantity fluctuation value. This fluctuation value is then used to adjust the delivery index threshold. This allows us to automatically optimize ad delivery parameters based on the dynamic changes in risk levels. This adaptive adjustment effectively improves the flexibility and accuracy of ad delivery, helping advertisers maintain the effectiveness and efficiency of their advertising even under high-risk conditions, thereby improving the overall effectiveness of their advertising.

[0057] Furthermore, by calculating the relative deviation between the quantity fluctuation value and the preset quantity fluctuation threshold, and lowering the delivery index threshold according to the adjustment coefficient when the deviation exceeds the preset threshold, it is possible to dynamically respond to significant changes in the risk level and adjust the advertising delivery strategy in a timely manner, which can effectively avoid poor advertising delivery results due to excessive risk fluctuations. At the same time, by lowering the delivery index threshold, the investment and return of advertising delivery can be more carefully controlled, thereby improving the stability and economic benefits of advertising delivery and ensuring that the advertising delivery strategy maintains the best effect in a dynamic environment. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] Figure 1 This is a schematic diagram of the big data-based advertising management system of this embodiment;

[0059] Figure 2 This is a determination logic diagram for the first determination unit of this embodiment to determine whether there is an advertisement demand;

[0060] Figure 3 This is a decision logic diagram for determining the advertisement type by the first determination unit of this embodiment;

[0061] Figure 4 This is a determination logic diagram for the second determination unit of this embodiment to determine whether there is an investment risk. DETAILED DESCRIPTION

[0062] In order to make the objects and advantages of the present invention more clearly understood, the present invention is further described below in conjunction with embodiments; it should be understood that the specific embodiments described herein are merely used to explain the present invention and are not intended to limit the present invention.

[0063] The preferred embodiments of the present invention are described below with reference to the accompanying drawings. It should be understood by those skilled in the art that these embodiments are only used to explain the technical principles of the present invention and are not intended to limit the scope of protection of the present invention.

[0064] See also Figure 1 As shown, it is a schematic diagram of the advertising management system based on big data in this embodiment;

[0065] This embodiment provides an advertising management system based on big data, including:

[0066] The collection module is used to collect the real-time advertising input, real-time click volume, real-time order completion rate and real-time search frequency of the target on the e-commerce platform;

[0067] A prediction module, connected to the acquisition module, for predicting a delivery index threshold based on the real-time input amount, the real-time click volume, the real-time transaction rate, the real-time search times, and a preset big data prediction model;

[0068] a first determination module, connected to the acquisition module and the prediction module respectively, for determining whether there is advertising demand based on the real-time click volume, the real-time transaction rate, the real-time search times, and the delivery index threshold, and forming a first determination result;

[0069] a first determination module, connected to the acquisition module and the first determination module respectively, for determining an advertisement type according to the first determination result, the real-time transaction rate, and the real-time search times;

[0070] a second determination module, connected to the acquisition module and the first determination module respectively, for determining whether there is an investment risk based on the advertisement type, the real-time investment amount, and the real-time transaction rate, and forming a second determination result;

[0071] a second determination module, connected to the acquisition module and the second determination module respectively, for determining the risk level of the investment risk according to the second determination result, the real-time investment amount, and the real-time transaction rate;

[0072] an adjustment module, connected to the second determination module and the prediction module respectively, for adjusting the delivery index threshold according to the number of times the risk level is formed within a preset adjustment period to form an adjusted delivery index threshold;

[0073] An output module is connected to the adjustment module and is used to provide corresponding management prompts for the advertisement type corresponding to the risk level determined based on the adjustment delivery index threshold.

[0074] The acquisition module builds a complete evaluation chain from "exposure-interest-conversion-market feedback" by monitoring advertising investment, clicks, conversion rates, and search volume in real time. Real-time investment refers to the amount of advertising spent by merchants during a specific time period. It reflects the merchant's investment in advertising and is the fundamental data for measuring advertising effectiveness. Real-time clicks refer to the number of times an ad is clicked by users. The higher the click rate, the more attractive the ad is to users, which can drive more traffic to the store. The real-time order conversion rate refers to the ratio of the number of orders actually completed after clicking an ad to the number of clicks within a certain period of time. It reflects the conversion rate of the ad, that is, the proportion of the traffic generated by the ad is converted into actual sales. The real-time search volume of the target refers to the number of times users search for the target keyword on the platform. It reflects the market demand and popularity of the target product and provides a reference for advertising strategy.

[0075] These data are collected in real time through API interfaces, tracking technology and system logs to form dynamic associations: the amount of investment determines the scale of advertising coverage, the number of clicks reflects the attractiveness of the content, the conversion rate verifies the commercial value, and the number of searches reveals potential demand; it can not only comprehensively evaluate the performance of each link of the advertisement and avoid the limitations of traditional single indicators, but also quickly discover problems through real-time linkage analysis, providing an accurate data foundation for subsequent intelligent predictions and dynamic adjustments, thereby effectively improving the scientific nature and response speed of advertising management.

[0076] The output module, serving as the system's decision-making terminal, aggregates and categorizes all real-time data after receiving the latest data from the adjustment module. It then uses intelligent decision-making algorithms (such as reinforcement learning or Bayesian optimization) to comprehensively analyze operational data (click volume, conversion rate, etc.) for ad types corresponding to risk levels determined based on the adjusted delivery index threshold. It then simulates various optimization strategies, including budget reallocation, delivery time adjustment, creative optimization, and audience targeting correction. For campaign ads, it's recommended to shorten the delivery cycle, optimize creatives to highlight limited-time offers, and add a risk warning mechanism. For brand ads, it's recommended to strengthen brand monitoring and optimize creatives to highlight brand value. For performance ads, it's recommended to optimize bidding strategies and reduce invalid clicks and conversion costs. Finally, the output module integrates the analysis results and optimization suggestions to generate intelligent marketing management prompts that include real-time advertising performance, risk causes, grade assessment, precise optimization measures and expected effect predictions. All prompts are based on multi-dimensional data cross-analysis and come with historical adjustment effect references to ensure that merchants can quickly execute data-driven optimization decisions, minimize the risk of ineffective delivery, and provide advertisers with data-driven dynamic optimization solutions to maximize advertising efficiency, control risks and sustainably improve marketing ROI.

[0077] The preset big data prediction model is a model built based on big data analysis and machine learning technology, which is used to predict the delivery index threshold based on real-time advertising data (including advertising investment, click volume, transaction rate, and search times).

[0078] 1. Model design ideas

[0079] Because advertising data has strong temporal dependencies and multivariate coupling characteristics (for example, click-through rates are influenced by historical exposure, and conversion rates are related to search trends), traditional statistical models (such as ARIMA) are difficult to accurately model. Therefore, the pre-defined big data prediction model uses a long short-term memory network (LSTM), which has the following advantages:

[0080] Time series modeling capabilities: LSTM memory units can capture long-term dependencies (such as the impact of seasonal promotions).

[0081] Multivariate collaborative analysis: supports simultaneous input of multi-dimensional features such as advertising investment, click volume, conversion rate, and number of searches.

[0082] Dynamic adaptability: Fine-tune the model through online learning to adapt to real-time data streams.

[0083] 2. Initial model: LSTM network structure

[0084] 2.1 Model Architecture

[0085] Input layer: receives multidimensional data in a time window ([batch_size, time_steps, input_dim]), where:

[0086] input_d im=4 (real-time input amount, click volume, transaction rate, number of searches).

[0087] time_steps = 7 (the default sliding window is 7 days, which can be adjusted).

[0088] LSTM layer:

[0089] A two-layer LSTM is used with 128 hidden units in each layer, and the information flow is controlled by the forget gate, input gate, and output gate.

[0090] Activation function: tanh (memory unit), sigmoid (gating).

[0091] Fully connected layer: maps the LSTM output to the delivery index threshold.

[0092] Output layer: Linear regression outputs predicted values ​​(ri sk_threshold).

[0093] 2.2 Core Parameters

[0094] Each layer of an LSTM contains multiple parameters that determine how it processes input data, maintains long-term memory, and outputs predictions.

[0095] batch_size: training batch size (default 32).

[0096] time_steps: time step (historical data window, default 7 days).

[0097] hidden_un its: LSTM hidden layer dimension (default 128).

[0098] Loss: Mean square error (MSE) + L2 regularization (to prevent overfitting).

[0099] Optimizer: Adam (initial learning rate 0.001).

[0100] Weights and biases of the LSTM layer:

[0101] The LSTM layer contains four gates: input gate, forget gate, output gate, and candidate memory unit. Each gate has a set of weights and biases:

[0102] Weight matrix (W): includes the input gate weight (W_i), forget gate weight (W_f), output gate weight (W_o), and candidate memory unit weight (W_c). The size of each matrix is ​​(input_d im, hidden_units), where hidden_units is the number of hidden units in the LSTM layer.

[0103] Bias vector (b): Each gate has a bias term, typically of size (h idden_un its,).

[0104] Activation function:

[0105] Forget gate (f): Sigmoid function (sigmoid), the output is between [0, 1], indicating the degree of forgetting of information in the memory unit.

[0106] Input gate (i): Sigmoid function that determines the degree of adoption of the input information of the current time step.

[0107] Candidate memory unit (c'): tanh function (tanh), used to generate candidate memory.

[0108] Output gate (o): Sigmoid function that determines the hidden state of the current time step.

[0109] 2.3 Hidden Layer and Output Layer

[0110] Hidden state (h_t): LSTM iteratively updates the hidden state through time steps. The hidden state stores the long-term memory of the time series.

[0111] Memory unit (C_t): The memory unit of LSTM is also updated as the time step advances, and it stores long-term information in the sequence.

[0112] Output layer: Ultimately, the output of the LSTM layer is passed to a fully connected layer or output layer to generate a prediction result.

[0113] 3. Training Process

[0114] 3.1 Data Preparation

[0115] Training data: historical advertising logs (investment, clicks, conversion rate, and number of searches over the past 90 days).

[0116] Label data: manually labeled delivery index threshold.

[0117] Data normalization: Min-Max normalization of input features (eliminating dimension effects).

[0118] 3.2 Model Training

[0119] Forward propagation: input time series data, the LSTM layer extracts features, and the fully connected layer outputs the predicted threshold.

[0120] Loss calculation: Compare the predicted value with the true label and calculate the MSE loss.

[0121] Backpropagation: Update weights through the Adam optimizer and adjust LSTM gating parameters.

[0122] Early stopping mechanism: If the validation set loss does not decrease for 5 consecutive rounds, the training is terminated.

[0123] 3.3 Training duration

[0124] Data size: 100,000 time series samples (training time: approximately 2 hours, NVIDIA T4 GPU).

[0125] Convergence criterion: validation set MSE < 0.01.

[0126] 4. Trained model

[0127] After training is complete, the model saves the following parameters:

[0128] LSTM layer weights: weight matrices of the input gate (W_i), forget gate (W_f), and output gate (W_o).

[0129] Fully connected layer parameters: weights (W_fc) and biases (b_fc) that map LSTM outputs to threshold predictions.

[0130] Normalization parameters: mean and variance of input features (used for real-time data normalization).

[0131] 5. Use the model to output the delivery index threshold

[0132] The trained LSTM model can be used to predict the delivery index threshold using the following steps:

[0133] Real-time data input: The collection module inputs the latest 7 days of advertising data (investment, click volume, conversion rate, number of searches).

[0134] Data preprocessing: normalization (using the mean / variance during training) and adjustment to (1, 7, 4) input dimensions.

[0135] Model reasoning: The LSTM layer calculates time series features, and the fully connected layer outputs the predicted delivery index threshold.

[0136] The pre-set big data prediction model in this embodiment is based on an LSTM network. By processing time series information from historical data, a deep learning model is developed that can predict delivery index thresholds. The model's core parameters include the LSTM layer's weight matrix, bias term, and the optimizer and loss function used during training. By training on historical data, the model can output accurate delivery index thresholds, assisting advertisers in managing their ads.

[0137] This system collects real-time advertising investment, real-time clicks, real-time product transaction rates, and real-time user search times from e-commerce platforms. Next, based on this real-time data and a pre-set big data prediction model, it predicts the delivery index threshold. Then, based on the real-time clicks, real-time transaction rate, real-time search times, and the delivery index threshold, it determines whether there is advertising demand and obtains a first determination result. Based on the first determination result, the real-time transaction rate, and real-time search times, it determines the ad type. Next, based on the determined ad type, real-time investment, and real-time transaction rate, it determines whether there is investment risk and obtains a second determination result. Based on the second determination result, the real-time investment, real-time transaction rate, and real-time clicks, it determines the investment risk level. Based on the number of times the risk level is reached within a pre-set adjustment period, it adjusts the delivery index threshold to obtain an adjusted delivery index threshold. Finally, it provides targeted management recommendations for the ad types corresponding to the risk levels determined based on the adjusted delivery index threshold.

[0138] By collecting key data from e-commerce platform ads in real time and integrating it with big data prediction models, we can accurately predict advertising placement index thresholds, effectively determining advertising demand and determining ad types. Furthermore, by dynamically assessing and grading investment risks based on real-time data, we can promptly identify potential risks and adjust advertising strategies. By adjusting the placement index thresholds, we can further optimize advertising effectiveness, ensuring scientific and efficient advertising. Ultimately, by providing targeted management recommendations for different risk levels and ad types, we can help e-commerce companies rationally plan their advertising budgets, improve their advertising return on investment, enhance their market competitiveness, and achieve refined advertising management and maximized benefits. This effectively addresses the issues of delayed risk identification and optimization adjustments caused by overreliance on static matching strategies and lagging feedback mechanisms.

[0139] Specifically, the first determination module includes:

[0140] A click volume fluctuation calculation unit, configured to calculate a standard deviation of the real-time click volume within a preset first determination time period to form a click volume fluctuation value;

[0141] a first transaction rate fluctuation calculation unit, configured to calculate a standard deviation of the real-time transaction rate within the preset first determination time period to form a first transaction rate fluctuation value;

[0142] A first search fluctuation calculation unit is used to calculate the standard deviation of the number of real-time searches within the preset first determination time period to form a first search fluctuation value;

[0143] The first determination unit is respectively connected to the click volume fluctuation calculation unit, the first transaction rate fluctuation calculation unit and the first search fluctuation calculation unit, and is used to determine whether there is an advertising demand based on the click volume fluctuation value, the first transaction rate fluctuation value, the first search fluctuation value and the delivery index threshold, and form a first determination result.

[0144] The preset first judgment duration refers to the time range used in the advertising system to calculate the fluctuation values ​​of real-time click volume, real-time transaction rate, and real-time search count. It depends on the volatility of the data, business needs, and the response speed of the system, and is generally set between 1 and 10 minutes. In this embodiment, it is set to 5 minutes. This can effectively capture data fluctuations in a short period of time, avoid frequent misjudgments due to an excessively short duration, and better adapt to the real-time and dynamic nature of advertising delivery.

[0145] By calculating the standard deviation of the real-time click volume, the click volume fluctuation value is obtained; then, the standard deviation of the real-time transaction rate is calculated to obtain the first transaction rate fluctuation value; then the standard deviation of the real-time search times is calculated to obtain the first search fluctuation value; finally, based on the click volume fluctuation value, the first transaction rate fluctuation value, the first search fluctuation value and the delivery index threshold, it is determined whether there is advertising demand, thereby obtaining the first judgment result.

[0146] By calculating the fluctuations in click volume, conversion rate, and search results within a preset first judgment period, and combining them with the delivery index threshold for comprehensive judgment, we can accurately identify advertising demand. This design captures fluctuations in key data during the delivery process in real time, helping advertisers more accurately evaluate the effectiveness of their advertising and adjust their advertising strategies in a timely manner, thereby improving the precision and efficiency of their delivery.

[0147] Please continue reading Figure 2 As shown, it is a determination logic diagram of the first determination unit of this embodiment for determining whether there is an advertisement demand;

[0148] The first determination unit includes:

[0149] a delivery index calculation subunit, configured to perform a weighted summation of the click volume fluctuation value, the first transaction rate fluctuation value, the first search fluctuation value, a preset click volume weight, a preset transaction rate weight, and a preset search number weight to form a delivery index;

[0150] The first determination subunit is connected to the delivery index calculation subunit and is used to determine that there is an advertisement demand when the delivery index is greater than the delivery index threshold, thereby forming a first determination result.

[0151] The preset click-through weight is used to measure the importance of clicks in evaluating the effectiveness of ad placement. It depends on the advertising objectives, market competition, and the stage of ad placement, and is typically set between 0.2 and 0.5. In this embodiment, it is set to 0.35, which can more effectively reflect the attractiveness and user attention of the ad, and helps to quickly evaluate the initial effectiveness of the ad placement.

[0152] The preset conversion rate weight is used to measure the importance of conversion rate in evaluating the effectiveness of advertising. It depends on the conversion goal of the advertisement, the product type, and the market environment, and is usually set between 0.3 and 0.6. In this example, it is set to 0.5, which can more directly reflect the actual contribution of advertising to sales and help accurately evaluate the conversion effect of advertising.

[0153] The preset search count weight is used to measure the importance of search counts in evaluating ad effectiveness. It depends on the target audience, brand awareness, and market competition, and is typically set between 0.2 and 0.4. In this example, it is set to 0.15, which reflects user interest and attention in the ad content while avoiding overreliance on a single metric and ensuring a balanced overall evaluation.

[0154] The delivery index is obtained by weighted summing the click volume fluctuation value, the first transaction rate fluctuation value, the first search fluctuation value, the preset click volume weight, the preset transaction rate weight and the preset search number weight; then, when the delivery index is greater than the delivery index threshold, it is determined that there is advertising demand, and a first determination result is obtained.

[0155] By calculating the delivery index and determining advertising demand when it exceeds the delivery index threshold, it comprehensively considers fluctuations in multi-dimensional data, avoiding the limitations of a single indicator and enabling a more comprehensive and accurate identification of actual demand for advertising. This not only improves the accuracy of ad demand determination but also effectively reduces misjudgments, helping advertisers better optimize their advertising strategies, improve the efficiency and effectiveness of advertising, and ultimately enhance the scientific and economic benefits of advertising.

[0156] Specifically, the first determining module includes:

[0157] A second transaction rate fluctuation calculation unit is used to calculate the standard deviation of the real-time transaction rate within a predetermined time period to form a second transaction rate fluctuation value;

[0158] A second search fluctuation calculation unit is used to calculate the standard deviation of the number of real-time searches within the preset determined time period to form a second search fluctuation value;

[0159] The first determining unit is connected to the second transaction rate fluctuation calculating unit and the second search fluctuation calculating unit respectively, and is used to determine the advertisement type according to the second transaction rate fluctuation value and the second search fluctuation value.

[0160] The preset duration is the time range used to calculate the standard deviation of the real-time conversion rate and the number of real-time searches. It depends on data volatility, business needs, and system response speed, and is typically set between 10 minutes and 1 hour. In this embodiment, it is set to 30 minutes. This not only captures short-term fluctuations in the conversion rate and the number of searches, avoiding frequent misjudgments due to too short a duration, but also ensures a certain degree of data stability and representativeness, providing a more reliable basis for accurately determining ad types.

[0161] The standard deviation of the real-time transaction rate is calculated to obtain a second transaction rate fluctuation value. The standard deviation of the real-time search count is also calculated to obtain a second search fluctuation value. The ad type is then determined based on the second transaction rate fluctuation value and the second search fluctuation value.

[0162] By calculating the second conversion rate fluctuation value and the second search fluctuation value within a preset time period to determine the ad type, we can accurately capture the key data fluctuation characteristics during the ad delivery process. Analysis based on data fluctuations can more accurately identify the performance differences between different ad types within a specific time period, helping advertisers select the most appropriate ad type based on fluctuations, optimize ad delivery strategies, improve the accuracy and effectiveness of ad delivery, and ultimately enhance the scientific and economic benefits of advertising.

[0163] Please continue reading Figure 3 As shown, it is a decision logic diagram of the first determination unit determining the advertisement type in this embodiment;

[0164] The first determining unit includes:

[0165] a transaction rate curve drawing subunit, configured to draw a change curve of the fluctuation value of the second transaction rate within the predetermined time period to form a transaction rate curve;

[0166] A search curve drawing subunit is used to draw a change curve of the second search fluctuation value within the preset determined time period to form a search curve;

[0167] a consistency calculation subunit, connected to the transaction rate curve drawing subunit and the search curve drawing subunit respectively, for calculating the cosine similarity between the transaction rate curve and the search curve to form a change consistency;

[0168] A first determination subunit is connected to the consistency calculation subunit, and is used to determine that the advertisement type is an active advertisement when the change consistency is greater than a preset consistency threshold; to determine that the advertisement type is a brand advertisement when the change consistency is less than or equal to the preset consistency threshold, and the second transaction rate fluctuation value is less than the preset second transaction rate fluctuation threshold, but the second search fluctuation value is greater than the preset second search fluctuation threshold; and to determine that the advertisement type is an effect advertisement when the change consistency is less than or equal to the preset consistency threshold, and the second transaction rate fluctuation value is greater than or equal to the preset second transaction rate fluctuation threshold, but the second search fluctuation value is less than or equal to the preset second search fluctuation threshold.

[0169] The preset consistency threshold is a standard value used to determine the degree of similarity between the conversion rate curve and the search curve. It depends on the correlation of data fluctuations and the requirements of the advertising delivery strategy, and is usually set between 0.7 and 0.9. In this embodiment, it is set to 0.8, which can effectively balance the consistency and difference of curve changes, avoid misjudgment, and ensure effective differentiation of ad types.

[0170] The preset second closing rate fluctuation threshold is a standard value used to determine whether closing rate fluctuations are significant. It depends on the normal range of closing rate fluctuations and the advertising delivery objectives, and is typically set between 10% and 20%. In this embodiment, it is set to 15%, which can effectively distinguish normal fluctuations from abnormal fluctuations and avoid misjudging the ad type due to small fluctuations.

[0171] The preset second search fluctuation threshold is a standard value used to determine whether the fluctuation in search counts is significant. It depends on the normal fluctuation range of search counts and the requirements of the advertising delivery strategy, and is typically set between 20% and 30%. In this embodiment, it is set to 25%, which can effectively distinguish normal from abnormal fluctuations and ensure accurate identification of brand ads and performance ads.

[0172] By plotting a conversion rate curve and a search curve over a preset duration and calculating their cosine similarity to obtain a change consistency, the system can accurately determine ad type based on different criteria. When the change consistency exceeds a preset consistency threshold, the ad is identified as a campaign ad. When the change consistency is less than or equal to the threshold, and the second conversion rate fluctuation value is less than the preset threshold but the second search fluctuation value is greater than the preset threshold, the ad is identified as a brand ad. When the change consistency is less than or equal to the threshold, and the second conversion rate fluctuation value is greater than or equal to the threshold but the second search fluctuation value is less than or equal to the threshold, the ad is identified as a performance ad. Campaign ads typically cause significant fluctuations in both conversion rates and search volume during their run. Therefore, when the change consistency of the conversion rate and search curves exceeds the preset consistency threshold, the ad is identified as a campaign ad. Brand ads primarily increase brand awareness and may significantly increase search volume, but have a minimal impact on conversion rates. Performance ads primarily promote sales and typically significantly increase conversion rates, but have a limited impact on search volume.

[0173] By plotting the conversion rate curve and the search curve and calculating their cosine similarity to obtain the variation consistency, the ad type can be accurately determined based on different criteria. When the variation consistency exceeds the preset consistency threshold, it is determined to be an active ad, as the fluctuations in its conversion rate and search count are highly consistent, indicating that the advertising has a significant and synchronized impact on user behavior. When the variation consistency is less than or equal to the preset consistency threshold, and the conversion rate fluctuates slightly but the search count fluctuates significantly, it is determined to be a brand ad, as brand ads focus more on increasing brand awareness and encouraging users to actively search. When the variation consistency is less than or equal to the preset consistency threshold, and the conversion rate fluctuates significantly but the search count fluctuates slightly, it is determined to be a performance ad, as performance ads focus more on directly driving sales. Analysis based on data fluctuation characteristics and correlations can effectively differentiate the effectiveness of different types of ads, helping advertisers more scientifically select advertising strategies, thereby improving the precision and effectiveness of advertising.

[0174] Specifically, the second determination module includes:

[0175] A first input fluctuation value calculation unit is used to calculate the standard deviation of the real-time input amount within a preset second determination time period to form a first input fluctuation value;

[0176] a third transaction rate fluctuation value calculation unit, configured to calculate a standard deviation of the real-time transaction rate within the preset second determination time period to form a third transaction rate fluctuation value;

[0177] The second determination unit is connected to the first investment fluctuation value calculation unit and the third transaction rate fluctuation value calculation unit respectively, and is used to determine whether there is an investment risk according to the first investment fluctuation value and the third transaction rate fluctuation value to form a second determination result.

[0178] The preset second judgment period is the time range used to assess the investment risk of advertising. It depends on the cyclical characteristics of advertising, the stability of data, and the timeliness requirements of risk assessment for the business, and is typically set between 15 minutes and 1 hour. In this embodiment, it is set to 30 minutes. This can effectively identify short-term fluctuations in investment and conversion rates while avoiding frequent misjudgments caused by too short a period, thereby more accurately assessing investment risk.

[0179] By calculating the standard deviation of the real-time investment amount, we obtain the first investment fluctuation value. Simultaneously, we calculate the standard deviation of the real-time transaction rate to obtain the third transaction rate fluctuation value. Then, based on the first investment fluctuation value and the third transaction rate fluctuation value, we determine whether there is investment risk and obtain the second determination result.

[0180] By calculating the first investment fluctuation value and the third transaction rate fluctuation value within the preset second judgment period to determine the investment risk, it is possible to monitor the dynamic changes of investment and return during the advertising delivery process in real time, and can effectively identify situations where investment fluctuations are large but transaction rate fluctuations are small, and timely discover potential investment risks, helping advertisers make more informed decisions during the advertising delivery process, avoid unnecessary waste of funds, and thus improve the economic benefits and return on investment of advertising.

[0181] Please continue reading Figure 4 As shown, it is a determination logic diagram of the second determination unit of this embodiment for determining whether there is an investment risk;

[0182] The second determining unit includes:

[0183] an investment comparison subunit, configured to compare the first investment fluctuation value with a preset investment fluctuation value threshold to form an investment comparison result;

[0184] a transaction rate comparison subunit, configured to compare the third transaction rate fluctuation value with a preset third transaction rate fluctuation threshold to form a transaction rate comparison result;

[0185] The second judgment sub-unit is connected to the investment comparison sub-unit and the transaction rate comparison sub-unit respectively, and is used to determine that there is an investment risk when the investment comparison result is that the first investment fluctuation value is greater than the preset investment fluctuation value threshold, or when the transaction rate comparison result is that the third transaction rate fluctuation value is less than the preset third transaction rate fluctuation threshold, to form a second judgment result.

[0186] The preset investment fluctuation threshold is a standard value used to measure the degree of real-time investment fluctuation during the advertising delivery process. It is determined by historical data fluctuations, advertising delivery strategies, and the advertiser's risk tolerance, and is typically set between 10% and 30%. In this embodiment, it is set to 20%, which can effectively identify excessive investment fluctuations and avoid investment risks caused by excessive investment, while also preventing over-sensitivity and frequent alarm triggering.

[0187] The preset third closing rate fluctuation threshold is a standard value used to measure the degree of real-time closing rate fluctuation during the advertising process. It depends on historical closing rate fluctuations, advertising delivery objectives, and market conditions, and is typically set between 5% and 15%. In this embodiment, it is set to 10% to effectively identify cases where closing rate fluctuations are too small, allowing timely adjustments to advertising delivery strategies to avoid wasted investment due to poor advertising performance.

[0188] An investment comparison result is obtained by comparing the first investment fluctuation value with a preset investment fluctuation value threshold; and a transaction rate comparison result is obtained by comparing the third transaction rate fluctuation value with a preset third transaction rate fluctuation threshold. When the investment comparison result shows that the first investment fluctuation value is greater than the preset investment fluctuation value threshold, or when the transaction rate comparison result shows that the third transaction rate fluctuation value is less than the preset third transaction rate fluctuation threshold, an investment risk is determined to exist, and a second determination result is obtained.

[0189] By comparing the first investment fluctuation value with the preset investment fluctuation threshold, and the third closing rate fluctuation value with the preset third closing rate fluctuation threshold, potential investment risks during the advertising process can be accurately identified. When the first investment fluctuation value exceeds the threshold, it indicates that advertising investment is fluctuating significantly, and there may be a risk of cost out-of-control. When the third closing rate fluctuation value is below the threshold, it indicates that advertising is not significantly promoting sales, and there may be a risk of insufficient return on investment. This dual comparison can provide timely warnings when investment is too high or returns are insufficient, helping advertisers to adjust their advertising strategies in advance and optimize advertising budget allocation, thereby effectively reducing investment risks and improving the economic benefits and return on investment of advertising.

[0190] Specifically, the second determining module includes:

[0191] a high investment risk determination unit, configured to determine that the investment risk is a high investment risk when the investment comparison result shows that the first investment fluctuation value is greater than the preset investment fluctuation value threshold, and when the transaction rate comparison result shows that the third transaction rate fluctuation value is less than the preset third transaction rate fluctuation threshold;

[0192] By judging the investment comparison result and the transaction rate comparison result, if the first investment fluctuation value is greater than the preset investment fluctuation value threshold, and the third transaction rate fluctuation value is less than the preset third transaction rate fluctuation threshold, the investment risk is determined to be high investment risk.

[0193] Through a multi-level risk assessment mechanism, the system can accurately identify high-investment risks based on different fluctuations. Ads with high investment risks may face the problem of large capital waste or unsatisfactory returns, requiring timely intervention to avoid greater losses. In contrast, ads with medium and low investment risks are usually in a more stable state, with their fluctuations within a controllable range and a relatively small impact on the overall delivery effect. Therefore, special prompts or interventions are not required for the time being. By focusing resources and attention on high-investment-risk ads, the system can more efficiently optimize advertising strategies and ensure that limited resources are prioritized for the ads that need the most attention, thereby improving the overall advertising effectiveness and return on investment.

[0194] Specifically, the adjustment module includes:

[0195] a quantity fluctuation value calculation unit, configured to calculate a standard deviation of the number of times the high investment risk occurs within the preset adjustment period to form a quantity fluctuation value;

[0196] An adjustment unit is connected to the quantity fluctuation value calculation unit and is used to adjust the delivery index threshold according to the quantity fluctuation value to form an adjusted delivery index threshold.

[0197] The preset adjustment period is the timeframe used to assess the frequency of high investment risk formation. It depends on data volatility, system response speed, and business needs, and is typically set between 1 and 24 hours. In this example, it is set to 6 hours, which effectively reflects the dynamic changes in risk levels while avoiding frequent adjustments caused by too short a period, thereby improving system stability and reliability.

[0198] By calculating the standard deviation of the number of times high investment risks are formed, the quantity fluctuation value is obtained; then, the investment index threshold is adjusted according to this quantity fluctuation value, thereby obtaining the adjusted investment index threshold.

[0199] By calculating the standard deviation of the number of high-risk events within a preset adjustment period, we determine the quantity fluctuation value. This fluctuation value is then used to adjust the delivery index threshold. This allows us to automatically optimize ad delivery parameters based on the dynamic changes in risk levels. This adaptive adjustment effectively improves the flexibility and accuracy of ad delivery, helping advertisers maintain effectiveness and efficiency even under high-risk conditions, thereby improving the overall effectiveness of advertising.

[0200] Specifically, the adjustment unit includes:

[0201] a quantity fluctuation deviation calculation subunit, configured to calculate a relative deviation between the quantity fluctuation value and the preset quantity fluctuation threshold value when the quantity fluctuation value is greater than a preset quantity fluctuation threshold value, to form a quantity fluctuation deviation;

[0202] The adjustment subunit is connected to the quantity fluctuation deviation calculation subunit and is used to reduce the delivery index threshold according to the quantity fluctuation deviation and a preset adjustment coefficient to form an adjusted delivery index threshold when the quantity fluctuation deviation is greater than a preset quantity fluctuation deviation threshold.

[0203] The preset quantity fluctuation threshold is a benchmark used to determine whether quantity fluctuations are abnormal. It depends on the data's fluctuation characteristics, business requirements, and the system's risk tolerance, and is typically set between 10% and 30%. In this embodiment, it is set at 20%, which allows for timely detection of abnormal fluctuations while avoiding frequent false alarms caused by a too low threshold.

[0204] The preset quantity fluctuation deviation threshold is a benchmark used to determine whether quantity fluctuation deviation is abnormal. It depends on the fluctuation characteristics of the data, business needs, and the adjustment sensitivity of the system, and is typically set between 5% and 15%. In this embodiment, it is set to 10%, which can both promptly identify situations requiring adjustment and avoid frequent adjustments due to a too low threshold.

[0205] The preset adjustment coefficient is a proportional factor used to adjust the delivery index threshold. It depends on the system's adjustment strategy, business needs, and risk appetite, and is typically set between 0.5 and 1. In this embodiment, it is set to 0.8. This allows for moderate adjustments to the delivery index threshold when abnormal fluctuations are detected, ensuring that it is neither too aggressive nor too conservative, helping the system maintain good adaptability in a dynamic environment.

[0206] By comparing the quantity fluctuation value within the preset adjustment period with the preset quantity fluctuation threshold, if the quantity fluctuation value is greater than the preset quantity fluctuation threshold, the relative deviation between the quantity fluctuation value and the preset quantity fluctuation threshold is calculated to obtain the quantity fluctuation deviation. If the quantity fluctuation deviation is greater than the preset quantity fluctuation deviation threshold, the delivery index threshold is lowered based on the quantity fluctuation deviation and the preset adjustment coefficient, thereby obtaining the adjusted delivery index threshold.

[0207] By calculating the relative deviation between the quantity fluctuation value and the preset quantity fluctuation threshold, and lowering the delivery index threshold according to the adjustment coefficient when the deviation exceeds the preset threshold, it is possible to dynamically respond to significant changes in the risk level and adjust the advertising delivery strategy in a timely manner, which can effectively avoid poor advertising delivery results caused by excessive risk fluctuations. At the same time, by lowering the delivery index threshold, it is possible to more carefully control the input and return of advertising delivery, thereby improving the stability and economic benefits of advertising delivery, and ensuring that the advertising delivery strategy maintains the best effect in a dynamic environment.

[0208] Thus far, the technical solutions of the present invention have been described in conjunction with the preferred embodiments shown in the accompanying drawings. However, it will be readily understood by those skilled in the art that the scope of protection of the present invention is obviously not limited to these specific embodiments. Without departing from the principles of the present invention, those skilled in the art may make equivalent changes or substitutions to the relevant technical features, and the technical solutions after such changes or substitutions will fall within the scope of protection of the present invention.

Claims

1. An advertising management system based on big data, characterized in that: include: The collection module is used to collect the real-time advertising input, real-time click volume, real-time order completion rate and real-time search frequency of the target on the e-commerce platform; A prediction module, connected to the acquisition module, for predicting a delivery index threshold based on the real-time input amount, the real-time click volume, the real-time transaction rate, the real-time search times, and a preset big data prediction model; a first determination module, connected to the acquisition module and the prediction module respectively, for determining whether there is advertising demand based on the real-time click volume, the real-time transaction rate, the real-time search times, and the delivery index threshold, and forming a first determination result; a first determination module, connected to the acquisition module and the first determination module respectively, for determining an advertisement type according to the first determination result, the real-time transaction rate, and the real-time search times; a second determination module, connected to the acquisition module and the first determination module respectively, for determining whether there is an investment risk based on the advertisement type, the real-time investment amount, and the real-time transaction rate, and forming a second determination result; a second determination module, connected to the acquisition module and the second determination module respectively, for determining the risk level of the investment risk according to the second determination result, the real-time investment amount, and the real-time transaction rate; an adjustment module, connected to the second determination module and the prediction module respectively, for adjusting the delivery index threshold according to the number of times the risk level is formed within a preset adjustment period to form an adjusted delivery index threshold; An output module is connected to the adjustment module and is used to provide corresponding management prompts for the advertisement type corresponding to the risk level determined based on the adjustment delivery index threshold.

2. The big data-based advertising management system according to claim 1, characterized in that: The first determination module includes: A click volume fluctuation calculation unit, configured to calculate a standard deviation of the real-time click volume within a preset first determination time period to form a click volume fluctuation value; a first transaction rate fluctuation calculation unit, configured to calculate a standard deviation of the real-time transaction rate within the preset first determination time period to form a first transaction rate fluctuation value; A first search fluctuation calculation unit is used to calculate the standard deviation of the number of real-time searches within the preset first determination time period to form a first search fluctuation value; The first judgment unit is respectively connected to the click volume fluctuation calculation unit, the first transaction rate fluctuation calculation unit and the first search fluctuation calculation unit, and is used to determine whether there is advertising demand based on the click volume fluctuation value, the first transaction rate fluctuation value, the first search fluctuation value and the delivery index threshold, and form a first judgment result.

3. The big data-based advertising management system according to claim 2, characterized in that: The first determination unit includes: a delivery index calculation subunit, configured to perform a weighted summation of the click volume fluctuation value, the first transaction rate fluctuation value, the first search fluctuation value, a preset click volume weight, a preset transaction rate weight, and a preset search number weight to form a delivery index; The first determination subunit is connected to the delivery index calculation subunit and is used to determine that there is an advertisement demand when the delivery index is greater than the delivery index threshold, thereby forming a first determination result.

4. The big data-based advertising management system according to claim 3, characterized in that: The first determining module includes: A second transaction rate fluctuation calculation unit is used to calculate the standard deviation of the real-time transaction rate within a predetermined time period to form a second transaction rate fluctuation value; A second search fluctuation calculation unit is used to calculate the standard deviation of the number of real-time searches within the preset determined time period to form a second search fluctuation value; The first determining unit is connected to the second transaction rate fluctuation calculating unit and the second search fluctuation calculating unit respectively, and is used to determine the advertisement type according to the second transaction rate fluctuation value and the second search fluctuation value.

5. The big data-based advertising management system according to claim 4, characterized in that: The first determining unit includes: a transaction rate curve drawing subunit, configured to draw a change curve of the fluctuation value of the second transaction rate within the predetermined time period to form a transaction rate curve; A search curve drawing subunit is used to draw a change curve of the second search fluctuation value within the preset determined time period to form a search curve; a consistency calculation subunit, connected to the transaction rate curve drawing subunit and the search curve drawing subunit respectively, for calculating the cosine similarity between the transaction rate curve and the search curve to form a change consistency; A first determination subunit is connected to the consistency calculation subunit, and is used to determine that the advertisement type is an active advertisement when the change consistency is greater than a preset consistency threshold; to determine that the advertisement type is a brand advertisement when the change consistency is less than or equal to the preset consistency threshold, and the second transaction rate fluctuation value is less than the preset second transaction rate fluctuation threshold, but the second search fluctuation value is greater than the preset second search fluctuation threshold; and to determine that the advertisement type is an effect advertisement when the change consistency is less than or equal to the preset consistency threshold, and the second transaction rate fluctuation value is greater than or equal to the preset second transaction rate fluctuation threshold, but the second search fluctuation value is less than or equal to the preset second search fluctuation threshold.

6. The big data-based advertising management system according to claim 5, characterized in that: The second determination module includes: A first input fluctuation value calculation unit is used to calculate the standard deviation of the real-time input amount within a preset second determination time period to form a first input fluctuation value; a third transaction rate fluctuation value calculation unit, configured to calculate a standard deviation of the real-time transaction rate within the preset second determination time period to form a third transaction rate fluctuation value; The second determination unit is connected to the first investment fluctuation value calculation unit and the third transaction rate fluctuation value calculation unit respectively, and is used to determine whether there is an investment risk according to the first investment fluctuation value and the third transaction rate fluctuation value to form a second determination result.

7. The big data-based advertising management system according to claim 6, characterized in that: The second determining unit includes: an investment comparison subunit, configured to compare the first investment fluctuation value with a preset investment fluctuation value threshold to form an investment comparison result; a transaction rate comparison subunit, configured to compare the third transaction rate fluctuation value with a preset third transaction rate fluctuation threshold to form a transaction rate comparison result; The second judgment sub-unit is connected to the investment comparison sub-unit and the transaction rate comparison sub-unit respectively, and is used to determine that there is an investment risk when the investment comparison result is that the first investment fluctuation value is greater than the preset investment fluctuation value threshold, or when the transaction rate comparison result is that the third transaction rate fluctuation value is less than the preset third transaction rate fluctuation threshold, to form a second judgment result.

8. The big data-based advertising management system according to claim 7, characterized in that: The second determining module includes: The high investment risk determination unit is used to determine that the investment risk is a high investment risk when the investment comparison result is that the first investment fluctuation value is greater than the preset investment fluctuation value threshold, and when the transaction rate comparison result is that the third transaction rate fluctuation value is less than the preset third transaction rate fluctuation threshold.

9. The big data-based advertising management system according to claim 8, characterized in that: The adjustment module includes: a quantity fluctuation value calculation unit, configured to calculate a standard deviation of the number of times the high investment risk occurs within the preset adjustment period to form a quantity fluctuation value; An adjustment unit is connected to the quantity fluctuation value calculation unit and is used to adjust the delivery index threshold according to the quantity fluctuation value to form an adjusted delivery index threshold.

10. The big data-based advertising management system according to claim 9, characterized in that: The adjustment unit includes: a quantity fluctuation deviation calculation subunit, configured to calculate a relative deviation between the quantity fluctuation value and the preset quantity fluctuation threshold value when the quantity fluctuation value is greater than a preset quantity fluctuation threshold value, to form a quantity fluctuation deviation; The adjustment subunit is connected to the quantity fluctuation deviation calculation subunit and is used to reduce the delivery index threshold according to the quantity fluctuation deviation and a preset adjustment coefficient to form an adjusted delivery index threshold when the quantity fluctuation deviation is greater than a preset quantity fluctuation deviation threshold.

Images