Dynamic bidding method based on intelligent decision model
By using a dynamic bidding method based on an intelligent decision-making model, bidding strategies are adjusted in real time, solving the problem of slow response of traditional bidding methods under market changes and competitive pressure, and achieving more efficient bidding decisions and resource optimization.
Patent Information
- Application Number
- CN202511642111.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-11-11
- Publication Date
- 2026-02-03
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
Traditional bidding methods are slow to respond to market changes and competitive pressures, lack flexibility, and lead to waste of resources and increased costs. They are unable to cope with complex market environments in real time, resulting in unsatisfactory bidding results.
A dynamic bidding method based on an intelligent decision-making model is adopted. Through real-time data collection, preprocessing, machine learning and reinforcement learning, a dynamic intelligent decision-making model is built to predict participant behavior and market trends in real time, generate and adjust bidding strategies, and optimize resource allocation.
It improves bidding efficiency and success rate, reduces response delay, lowers costs, enhances decision-making accuracy and flexibility, avoids resource waste, and is suitable for various bidding environments such as e-commerce platforms, advertising, and financial markets.
Smart Images
Figure CN121458352A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of dynamic bidding technology, specifically a dynamic bidding method based on an intelligent decision-making model. Background Technology
[0002] Dynamic bidding is a pricing strategy that adjusts product or service prices in real time based on market demand, supply, competitive environment, and other factors. It is widely used in various industries, including e-commerce, advertising, hotels, and airline ticketing. The core idea behind this technology is to acquire data in real time and automatically adjust prices based on that data to maximize profits or enhance competitiveness. Dynamic bidding involves a range of complex technologies, including real-time data analysis, machine learning algorithms, market forecasting, and optimization algorithms. Below are some of the core technologies and concepts related to dynamic bidding.
[0003] The existing technology has the following defects or problems: Traditional bidding methods often rely on manually set rules or simple algorithm models. However, this approach is often slow to react and lacks flexibility when facing market changes and competitive pressures. In particular, in bidding scenarios with multiple participants, dynamic competition, and frequent changes, traditional bidding methods cannot respond to complex market environments in real time, resulting in wasted resources, increased costs, and unsatisfactory bidding results.
[0004] It should be noted that the above content falls within the inventor's technical knowledge and does not necessarily constitute prior art. Summary of the Invention
[0005] To address the shortcomings of existing technologies, this invention provides a dynamic bidding method based on an intelligent decision-making model, which solves the current problems.
[0006] To achieve the above objectives, the present invention provides the following technical solution: a dynamic bidding method based on an intelligent decision-making model, comprising the following steps: Step 1: Collect relevant data in real time from multiple data sources during the bidding process; Step 2: Preprocess the collected data to ensure its accuracy and validity, providing high-quality input for subsequent model analysis; Step 3: Build a dynamic intelligent decision-making model through machine learning, combining historical bidding data and real-time market data, and learn and optimize through a simulated bidding environment; The dynamic intelligent decision-making model can predict participant behavior, market trends, and optimal bidding strategies in real time based on historical bidding data and the current market environment. Step 4: Analyze the current model using the decision model to generate a real-time bidding strategy; The bidding strategy includes adjusting the bidding amount, selecting the bidding timing, and predicting the strategies of bidding participants. The decision model can be continuously adjusted according to market feedback to ensure that the bid is kept at a reasonable distance from the strategies of competitors. Step 5: After the bidding process ends, collect the bidding results and update them. By analyzing the bidding results, adjust the model parameters and decision rules to further improve the accuracy and effectiveness of the decision-making. Step Six: Continuously optimize the decision-making model based on historical feedback data, and adaptively adjust the bidding strategy through incremental learning to ensure the long-term effectiveness of the model.
[0007] In some embodiments, the real-time data includes bidding product information, market demand data, historical bidding behavior of bidding participants, real-time behavior of competitors, and bidding period.
[0008] In some embodiments, the data preprocessing process is as follows: 1) Check if there are duplicate records in the collected data, and delete redundant information; 2) Fill in missing values using the mean, median, and mode to avoid missing data affecting the accuracy of subsequent models; 3) Detect outliers in the data using the Z-score method and decide whether to delete or correct them; 4) Ensure that all data fields are formatted consistently, including date and time formats and currency unit formats, to avoid data parsing errors caused by inconsistent formats; 5) The data cleaned in the above steps is transformed and organized to make it suitable for model analysis, including: By using statistical analysis, we select features that have a significant impact on the model's prediction performance and extract features from high-dimensional data to reduce redundant features and improve computational efficiency. For categorical variables, they need to be converted into numerical data using hot-coding. 6) Map the data to a uniform scale to avoid certain features having a large numerical impact on the model, which could lead to weight imbalance during model training. Use min-max normalization to scale the data proportionally to the range {0,1}. The specific formula is as follows: ; 7) For features such as price and income, logarithmic transformation can change the data from a skewed distribution to an approximately normal distribution, thereby improving the model training effect. At the same time, new features can be constructed based on existing features and combined with domain knowledge or business rules.
[0009] In some embodiments, the dynamic intelligent decision-making model uses a decision tree regression method, and the specific steps are as follows: 1) Decision trees divide the input space into multiple regions and then assign a predicted value to each region. The construction of the tree is based on the influence of features on the target variable. The optimal split point is selected using the mean squared error. The splitting objective for each node is to reduce the prediction error of its child nodes. The specific formula is as follows: ; in, It represents the number of samples from the two child nodes. It is the predicted value of the child node; 2) In the dynamic intelligent decision-making model bidding environment, each action will affect the future state and reward. A reinforcement learning model is constructed through the Q-learning method to perform self-learning and strategy optimization. 3) By simulating bidding scenarios, the model is trained and the bidding strategy is optimized. First, the real-time bidding process is simulated, including participants' bids, bidding rules and market dynamics. The model is trained using historical bidding data and simulated bidding environment, and the model parameters are repeatedly optimized so that it has good predictive ability and strategy execution ability in practical applications. 4) After the model training is completed, based on the current market data, participant behavior and remaining bidding time, the model actually predicts the bidding behavior of each participant and market price fluctuations, and also adjusts the prediction results and changes in the market environment. 5) Evaluate the model's predictive accuracy, decision-making effectiveness, and deviation from actual results through cross-validation and A / B testing; 6) Deploy the trained model to the production environment to make bidding decisions in real time. It can monitor the bidding process in real time, predict the optimal bidding strategy based on real-time data, and adjust the behavior of participants.
[0010] In some embodiments, the bidding strategy includes: Adjustment of Bidding Amount: Bidders adjust their bidding amount based on market competition. The decision on bidding amount not only considers their own budget and expected returns, but also needs to be flexibly adjusted based on competitors' historical bidding patterns, market price fluctuations, and changes in market demand. Bidding Strategy Selection: Bidders need to choose an appropriate bidding method, which requires taking into account the uncertainties in the bidding scenario and the behavior of competitors; Bidding participant strategy prediction: Predicting the behavior of other bidders is key to developing a successful bidding strategy. By using historical data and participant behavior models, we can predict competitors' bidding behavior and then formulate the optimal bidding strategy based on the prediction results.
[0011] In some embodiments, the feedback analysis of the bidding results is accomplished by collecting various data related to the bidding process. The main indicators are as follows: whether the bid was won and the winning bid amount and related influencing factors, the difference between the actual bid and the budget, the effectiveness of the bidding strategy, the final bid of competitors, strategy changes and whether there is overreaction or overly conservative behavior, and whether the bidding results are significantly related to external factors such as market demand, price fluctuations and policy changes.
[0012] In some embodiments, the decision-making model based on the collected bidding results and market feedback needs to be adjusted and optimized in real time. The specific steps are as follows: 1) Conduct error analysis based on the bidding results, including: Compare the differences between the model's predictions and the actual results, and analyze the reasons for the inaccuracy of the predictions; Analyze whether the bidding strategies generated by the analysis model are successful, and whether they exhibit overly conservative or overly aggressive behavior; Detect whether there is a strategy bias, and adjust the decision rules accordingly; 2) Based on the bidding results and error analysis, the model parameters need to be adjusted, including the bid range, budget allocation, and risk tolerance, among which: When the model is too conservative, the bid increment needs to be increased to make it more competitive; conversely, when the model is too aggressive, the bid increment needs to be reduced to avoid excessive competition. When the bidding results indicate that some strategies have high risks, the model needs to adjust its risk tolerance and adopt a more conservative bidding strategy. Based on the analysis of actual bidding results, adjust the cost-benefit balance in the bidding strategy to ensure that the bidding strategy can both increase the probability of winning and maximize profits. 3) When market demand and price volatility change, the sensitivity of bidding should be adjusted; at the same time, based on continuous learning of competitors' bidding behavior and historical data analysis, the accuracy of predicting competitors' strategies should be enhanced.
[0013] In some embodiments, the decision model optimization steps are as follows: 1) Collect accurate multi-dimensional information such as market trends, competitor behavior and price fluctuations to provide sufficient background for the model. Utilize multi-dimensional information to help the model better understand the reasons for past bidding successes and failures, thereby optimizing future bidding strategies; 2) Based on the feedback update mechanism of historical data, the model is trained and tested regularly. After each bidding, the bidding results are collected and fed back into the model as incremental data to optimize the next round of decision-making. 3) By extracting meaningful features, the model can better understand market dynamics. These features include bidders' historical bidding trends, changes in market supply and demand, and seasonal factors. 4) To ensure the continued effectiveness of the decision-making model in the long term, the model should be backtested periodically to evaluate its performance in different market environments. By comparing the model's predictions with the actual results, it is possible to detect whether the model is outdated or ineffective. 5) Based on the model's performance, conduct regular performance evaluations and updates. As the market environment and competitive landscape change, the model's parameters need to be adjusted regularly based on new data.
[0014] Compared with existing technologies, this invention provides a dynamic bidding method based on an intelligent decision-making model, which has the following beneficial effects: This paper presents a dynamic bidding method based on an intelligent decision-making model. Through machine learning and real-time data analysis, this method optimizes bidding decisions and automatically adjusts bidding strategies, significantly improving bidding efficiency and success rate. This method is applicable to various bidding environments such as e-commerce platforms, advertising, and financial markets, and has broad application prospects. Based on real-time market data and participant behavior, this method enables rapid bidding decisions compared to existing bidding methods, reducing response delays. Furthermore, through the training and optimization of the intelligent decision-making model, it can autonomously adjust bidding strategies, avoiding the limitations of manual operation and improving decision accuracy. Optimized bidding strategies reduce the risks of overbidding and missing bidding opportunities, thereby saving bidding costs. This dynamic bidding method can flexibly adjust strategies according to changes in the market environment and bidding situation, improving decision-making flexibility and responsiveness. Real-time adjustments by the intelligent decision-making model help optimize resource allocation, avoid resource waste, and improve overall bidding efficiency. Attached Figure Description
[0015] Figure 1 This is a schematic diagram of the dynamic bidding method based on an intelligent decision-making model according to the present invention. Detailed Implementation
[0016] The technical solutions of the present invention will be clearly and completely described below with reference to the embodiments and accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.
[0017] It should be understood that the step numbers used in the text are for ease of description only and are not intended to limit the order in which the steps are performed.
[0018] It should be understood that the terminology used in this specification is for the purpose of describing particular embodiments only and is not intended to limit the invention. As used in this specification and the appended claims, the singular forms “a,” “an,” and “the” are intended to include the plural forms unless the context clearly indicates otherwise.
[0019] The terms “comprising” and “including” indicate the presence of the described feature, whole, step, operation, element and / or component, but do not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components and / or collections thereof.
[0020] The term “and / or” refers to any combination of one or more of the associated listed items, as well as all possible combinations, and includes these combinations.
[0021] Please see Figure 1 In this implementation plan, a dynamic bidding method based on an intelligent decision-making model includes the following steps: Step 1: Collect relevant data in real time from multiple data sources during the bidding process; Real-time data includes information on bidding products, market demand data, historical bidding behavior of bidding participants, real-time behavior of competitors, and bidding periods; Step 2: Preprocess the collected data to ensure its accuracy and validity, providing high-quality input for subsequent model analysis; The data preprocessing process is as follows: 1) Check if there are duplicate records in the collected data, and delete redundant information; 2) Fill in missing values using the mean, median, and mode to avoid missing data affecting the accuracy of subsequent models; 3) Detect outliers in the data using the Z-score method and decide whether to delete or correct them; 4) Ensure that all data fields are formatted consistently, including date and time formats and currency unit formats, to avoid data parsing errors caused by inconsistent formats; 5) The data cleaned in the above steps is transformed and organized to make it suitable for model analysis, including: By using statistical analysis, we select features that have a significant impact on the model's prediction performance and extract features from high-dimensional data to reduce redundant features and improve computational efficiency. For categorical variables, they need to be converted into numerical data using hot-coding. 6) Map the data to a uniform scale to avoid certain features having a large numerical impact on the model, which could lead to weight imbalance during model training. Use min-max normalization to scale the data proportionally to the range {0,1}. The specific formula is as follows: ; 7) For features such as price and income, logarithmic transformation can change the data from a skewed distribution to an approximately normal distribution, thereby improving the model training effect. At the same time, new features can be constructed based on existing features and combined with domain knowledge or business rules. Step 3: Build a dynamic intelligent decision-making model through machine learning, combining historical bidding data and real-time market data, and learn and optimize through a simulated bidding environment; The dynamic intelligent decision-making model can predict participant behavior, market trends, and optimal bidding strategies in real time based on historical bidding data and the current market environment. The dynamic intelligent decision-making model uses decision tree regression, and its specific steps are as follows: 1) Decision trees divide the input space into multiple regions and then assign a predicted value to each region. The construction of the tree is based on the influence of features on the target variable. The optimal split point is selected using the mean squared error. The splitting objective for each node is to reduce the prediction error of its child nodes. The specific formula is as follows: ; in, It represents the number of samples from the two child nodes. It is the predicted value of the child node; 2) In a dynamic intelligent decision-making model bidding environment, every action affects the future state and reward. A reinforcement learning model is built using the Q-learning method to perform self-learning and strategy optimization. 3) By simulating bidding scenarios, the model is trained and the bidding strategy is optimized. First, the real-time bidding process is simulated, including participants' bids, bidding rules and market dynamics. The model is trained using historical bidding data and simulated bidding environment, and the model parameters are repeatedly optimized so that it has good predictive ability and strategy execution ability in practical applications. 4) After the model training is completed, based on the current market data, participant behavior and remaining bidding time, the model actually predicts the bidding behavior of each participant and market price fluctuations, and also adjusts the prediction results and changes in the market environment. 5) Evaluate the model's predictive accuracy, decision-making effectiveness, and deviation from actual results through cross-validation and A / B testing; 6) Deploy the trained model to the production environment to make bidding decisions in real time. It can monitor the bidding process in real time, predict the optimal bidding strategy based on real-time data, and adjust the behavior of participants.
[0022] 5. A dynamic bidding method based on an intelligent decision-making model according to claim 1, characterized in that the bidding strategy in step four includes: Adjustment of Bidding Amount: Bidders adjust their bidding amount based on market competition. The decision on bidding amount not only considers their own budget and expected returns, but also needs to be flexibly adjusted based on competitors' historical bidding patterns, market price fluctuations, and changes in market demand. Bidding Strategy Selection: Bidders need to choose an appropriate bidding method, which requires taking into account the uncertainties in the bidding scenario and the behavior of competitors; Bidding participant strategy prediction: Predicting the behavior of other bidders is key to developing a successful bidding strategy. By using historical data and participant behavior models, we can predict competitors' bidding behavior and then develop the optimal bidding strategy based on the prediction results. Step 4: Analyze the current model using the decision model to generate a real-time bidding strategy; Bidding strategies include adjusting the bid amount, choosing the timing of bidding, and predicting the strategies of bidding participants. The decision-making model can be continuously adjusted based on market feedback to ensure that the bid is kept at a reasonable distance from the strategies of competitors. Step 5: After the bidding process ends, collect the bidding results and update them. By analyzing the bidding results, adjust the model parameters and decision rules to further improve the accuracy and effectiveness of the decision-making. Feedback analysis of bidding results is completed by collecting various data related to the bidding process. Its main indicators are as follows: whether the bid was won and the winning bid amount and related influencing factors, the difference between the actual bid and the budget, the effectiveness of the bidding strategy, the final bid of competitors, strategy changes and whether there is overreaction or overly conservative behavior, and whether the bidding results are significantly related to external factors such as market demand, price fluctuations and policy changes. Based on the collected bidding results and market feedback, the decision-making model needs to be adjusted and optimized in real time. The specific steps are as follows: 1) Conduct error analysis based on the bidding results, including: Compare the differences between the model's predictions and the actual results, and analyze the reasons for the inaccuracy of the predictions; Analyze whether the bidding strategies generated by the analysis model are successful, and whether they exhibit overly conservative or overly aggressive behavior; Detect whether there is a strategy bias, and adjust the decision rules accordingly; 2) Based on the bidding results and error analysis, the model parameters need to be adjusted, including the bid range, budget allocation, and risk tolerance, among which: When the model is too conservative, the bid increment needs to be increased to make it more competitive; conversely, when the model is too aggressive, the bid increment needs to be reduced to avoid excessive competition. When the bidding results indicate that some strategies have high risks, the model needs to adjust its risk tolerance and adopt a more conservative bidding strategy. Based on the analysis of actual bidding results, adjust the cost-benefit balance in the bidding strategy to ensure that the bidding strategy can both increase the probability of winning and maximize profits. 3) When market demand and price volatility change, the sensitivity of bidding should be adjusted; at the same time, based on continuous learning of competitors' bidding behavior and historical data analysis, the accuracy of predicting competitors' strategies should be enhanced. Step Six: Continuously optimize the decision-making model based on historical feedback data, and adaptively adjust the bidding strategy through incremental learning to ensure the long-term effectiveness of the model; The steps for optimizing the decision model are as follows: 1) Collect accurate multi-dimensional information such as market trends, competitor behavior and price fluctuations to provide sufficient background for the model. Utilize multi-dimensional information to help the model better understand the reasons for past bidding successes and failures, thereby optimizing future bidding strategies; 2) Based on the feedback update mechanism of historical data, the model is trained and tested regularly. After each bidding, the bidding results are collected and fed back into the model as incremental data to optimize the next round of decision-making. 3) By extracting meaningful features, the model can better understand market dynamics. These features include bidders' historical bidding trends, changes in market supply and demand, and seasonal factors. 4) To ensure the continued effectiveness of the decision-making model in the long term, the model should be backtested periodically to evaluate its performance in different market environments. By comparing the model's predictions with the actual results, it is possible to detect whether the model is outdated or ineffective. 5) Based on the model's performance, conduct regular performance evaluations and updates. As the market environment and competitive landscape change, the model's parameters need to be adjusted regularly based on new data.
[0023] The various embodiments in this specification are described in a progressive manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, the system embodiments are basically similar to the method embodiments, so the description is relatively simple; relevant parts can be referred to the descriptions in the method embodiments.
[0024] Finally, it should be noted that the above descriptions are merely preferred embodiments of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.
Claims
1. A dynamic bidding method based on an intelligent decision-making model, characterized in that, Includes the following steps: Step 1: Collect relevant data in real time from multiple data sources during the bidding process; Step 2: Preprocess the collected data to ensure its accuracy and validity, providing high-quality input for subsequent model analysis; Step 3: Build a dynamic intelligent decision-making model through machine learning, combining historical bidding data and real-time market data, and learn and optimize through a simulated bidding environment; The dynamic intelligent decision-making model can predict participant behavior, market trends, and optimal bidding strategies in real time based on historical bidding data and the current market environment. Step 4: Analyze the current model using the decision model to generate a real-time bidding strategy; The bidding strategy includes adjusting the bidding amount, selecting the bidding timing, and predicting the strategies of bidding participants. The decision model can be continuously adjusted according to market feedback to ensure that the bid is kept at a reasonable distance from the strategies of competitors. Step 5: After the bidding process ends, collect the bidding results and update them. By analyzing the bidding results, adjust the model parameters and decision rules to further improve the accuracy and effectiveness of the decision-making. Step Six: Continuously optimize the decision-making model based on historical feedback data, and adaptively adjust the bidding strategy through incremental learning to ensure the long-term effectiveness of the model.
2. The dynamic bidding method based on an intelligent decision-making model according to claim 1, characterized in that, The real-time data mentioned in Step 1 includes bidding product information, market demand data, historical bidding behavior of bidding participants, real-time behavior of competitors, and bidding period.
3. The dynamic bidding method based on an intelligent decision-making model according to claim 1, characterized in that, The data preprocessing process described in step two is as follows: 1) Check if there are duplicate records in the collected data, and delete redundant information; 2) Fill in missing values using the mean, median, and mode to avoid missing data affecting the accuracy of subsequent models; 3) Detect outliers in the data using the Z-score method and decide whether to delete or correct them; 4) Ensure that all data fields are formatted consistently, including date and time formats and currency unit formats, to avoid data parsing errors caused by inconsistent formats; 5) The data cleaned in the above steps is transformed and organized to make it suitable for model analysis, including: By using statistical analysis, we select features that have a significant impact on the model's prediction performance and extract features from high-dimensional data to reduce redundant features and improve computational efficiency. For categorical variables, they need to be converted into numerical data using hot-coding. 6) Map the data to a uniform scale to avoid certain features having a large numerical impact on the model, which could lead to weight imbalance during model training. Use min-max normalization to scale the data proportionally to the range {0,1}. The specific formula is as follows: ； 7) For features such as price and income, logarithmic transformation can change the data from a skewed distribution to an approximately normal distribution, thereby improving the model training effect. At the same time, new features can be constructed based on existing features and combined with domain knowledge or business rules.
4. The dynamic bidding method based on an intelligent decision-making model according to claim 1, characterized in that, The dynamic intelligent decision-making model described in step three uses decision tree regression, and its specific steps are as follows: 1) Decision trees divide the input space into multiple regions and then assign a predicted value to each region. The construction of the tree is based on the influence of features on the target variable. The optimal split point is selected using the mean squared error. The splitting objective for each node is to reduce the prediction error of its child nodes. The specific formula is as follows: ； in, It represents the number of samples from the two child nodes. It is the predicted value of the child node; 2) In the dynamic intelligent decision-making model bidding environment, each action will affect the future state and reward. A reinforcement learning model is constructed through the Q-learning method to perform self-learning and strategy optimization. 3) By simulating bidding scenarios, the model is trained and the bidding strategy is optimized. First, the real-time bidding process is simulated, including participants' bids, bidding rules and market dynamics. The model is trained using historical bidding data and simulated bidding environment, and the model parameters are repeatedly optimized so that it has good predictive ability and strategy execution ability in practical applications. 4) After the model training is completed, based on the current market data, participant behavior and remaining bidding time, the model actually predicts the bidding behavior of each participant and market price fluctuations, and also adjusts the prediction results and changes in the market environment. 5) Evaluate the model's predictive accuracy, decision-making effectiveness, and deviation from actual results through cross-validation and A / B testing; 6) Deploy the trained model to the production environment to make bidding decisions in real time. It can monitor the bidding process in real time, predict the optimal bidding strategy based on real-time data, and adjust the behavior of participants.
5. The dynamic bidding method based on an intelligent decision-making model according to claim 1, characterized in that, The bidding strategy described in step four includes: Adjustment of Bidding Amount: Bidders adjust their bidding amount based on market competition. The decision on bidding amount not only considers their own budget and expected returns, but also needs to be flexibly adjusted based on competitors' historical bidding patterns, market price fluctuations, and changes in market demand. Bidding Strategy Selection: Bidders need to choose an appropriate bidding method, which requires taking into account the uncertainties in the bidding scenario and the behavior of competitors; Bidding participant strategy prediction: Predicting the behavior of other bidders is key to developing a successful bidding strategy. By using historical data and participant behavior models, we can predict competitors' bidding behavior and then formulate the optimal bidding strategy based on the prediction results.
6. The dynamic bidding method based on an intelligent decision-making model according to claim 1, characterized in that, The feedback analysis of the bidding results described in step five is completed by collecting various data related to the bidding process. Its main indicators are as follows: whether the bid was won and the winning bid amount and related influencing factors, the difference between the actual bid and the budget, the effectiveness of the bidding strategy, the final bid of competitors, strategy changes and whether there is overreaction or overly conservative behavior, and whether the bidding results are significantly related to external factors such as market demand, price fluctuations and policy changes.
7. The dynamic bidding method based on an intelligent decision-making model according to claim 6, characterized in that, Step five describes a decision-making model that needs to be adjusted and optimized in real time based on the collected bidding results and market feedback. The specific steps are as follows: 1) Conduct error analysis based on the bidding results, including: Compare the differences between the model's predictions and the actual results, and analyze the reasons for the inaccuracy of the predictions; Analyze whether the bidding strategies generated by the analysis model are successful, and whether they exhibit overly conservative or overly aggressive behavior; Detect whether there is a strategy bias, and adjust the decision rules accordingly; 2) Based on the bidding results and error analysis, the model parameters need to be adjusted, including the bid range, budget allocation, and risk tolerance, among which: When the model is too conservative, the bid increment needs to be increased to make it more competitive; conversely, when the model is too aggressive, the bid increment needs to be reduced to avoid excessive competition. When the bidding results indicate that some strategies have high risks, the model needs to adjust its risk tolerance and adopt a more conservative bidding strategy. Based on the analysis of actual bidding results, adjust the cost-benefit balance in the bidding strategy to ensure that the bidding strategy can both increase the probability of winning and maximize profits. 3) When market demand and price volatility change, the sensitivity of bidding should be adjusted; at the same time, based on continuous learning of competitors' bidding behavior and historical data analysis, the accuracy of predicting competitors' strategies should be enhanced.
8. The dynamic bidding method based on an intelligent decision-making model according to claim 1, characterized in that, The decision model optimization steps described in step six are as follows: 1) Collect accurate multi-dimensional information such as market trends, competitor behavior and price fluctuations to provide sufficient background for the model. Utilize multi-dimensional information to help the model better understand the reasons for past bidding successes and failures, thereby optimizing future bidding strategies; 2) Based on the feedback update mechanism of historical data, the model is trained and tested regularly. After each bidding, the bidding results are collected and fed back into the model as incremental data to optimize the next round of decision-making. 3) By extracting meaningful features, the model can better understand market dynamics. These features include bidders' historical bidding trends, changes in market supply and demand, and seasonal factors. 4) To ensure the continued effectiveness of the decision-making model in the long term, the model should be backtested periodically to evaluate its performance in different market environments. By comparing the model's predictions with the actual results, it is possible to detect whether the model is outdated or ineffective. 5) Based on the model's performance, conduct regular performance evaluations and updates. As the market environment and competitive landscape change, the model's parameters need to be adjusted regularly based on new data.