E-commerce platform consumer behavior prediction and decision optimization method based on attention iteration

Abstract

The invention relates to the technical field of e-commerce management, in particular to an attention iteration-based e-commerce platform consumer behavior prediction and decision optimization method. According to the method, through the steps of data collection, preprocessing, consumer classification and the like, consumer behavior data on an e-commerce platform are deeply analyzed, vectorization, classification and weighted summation are carried out on the consumer behavior data based on an attention mechanism, and a consumer behavior prediction model is established; performing iterative calculation on the click log sequence through a recurrent neural network; according to the method, the prediction model can be dynamically optimized, the prediction accuracy of future behaviors of consumers can be improved, an internal relation can be established between different click and purchase behaviors, different weights are allocated to the behaviors of each consumer through an attention mechanism, the purchase intention of the user is accurately captured, and the user experience is improved. And finally, personalized recommendation, accurate prediction and real-time decision optimization can be realized, so that the operation efficiency, the user experience and the commercial value of the e-commerce platform are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of e-commerce management, and in particular to an attention iteration-based consumer behavior prediction and decision optimization method for an e-commerce platform. Background Art

[0002] With the rapid development of e-commerce, e-commerce platforms have accumulated a massive amount of user behavior data, including user browsing, clicking, searching, shopping, etc. These data provide valuable insights for the platforms, helping them understand user needs, optimize product recommendations, and improve user experience. However, how to accurately predict consumer behavior from these massive amounts of data and achieve personalized recommendations and decision optimization is still a major challenge facing e-commerce platforms.

[0003] Most existing consumer behavior prediction methods rely on traditional statistical methods and rule-based models, such as collaborative filtering, content-based recommendation, and matrix factorization. These methods predict users' future preferences and needs by analyzing their historical behaviors and the behaviors of other similar users.

[0004] However, these traditional methods have the following limitations: 1. Data sparsity: Collaborative filtering methods are less effective when there is less interaction data between users and items. For new users or cold-start items, these methods are difficult to make effective recommendations.

[0005] 2. Long-tail problem: Most traditional recommendation algorithms do not work well when dealing with "long-tail products", that is, for those products with few views but potential markets, the recommendation accuracy of the recommendation system is low.

[0006] 3. Lack of dynamism: Traditional methods are usually unable to capture changes in user behavior in real time and cannot respond quickly to short-term changes in users.

[0007] In recent years, with the rapid development of deep learning technology, consumer behavior prediction methods based on deep learning have been widely used. Especially when dealing with large-scale, high-dimensional data, deep learning can effectively mine complex patterns in data and overcome some limitations of traditional methods.

[0008] Common consumer behavior prediction methods based on deep learning include convolutional neural networks (CNN), recurrent neural networks (RNN), long short-term memory networks (LSTM), etc. Although deep learning methods can significantly improve prediction accuracy, they still face challenges in processing multi-level and complex user behavior data. For example, how to effectively combine multi-dimensional data such as user click behavior, purchase behavior, and product information to further improve the model's prediction ability has become a hot topic in current deep learning research.

[0009] In order to overcome the limitations of traditional deep learning methods in processing complex data, the attention mechanism has been widely introduced into the field of consumer behavior prediction in recent years. The attention mechanism originally originated from the field of natural language processing (NLP). Its core idea is to give different weights to different parts of the input data so that the model can focus on the part most relevant to the task.

[0010] In the prediction of consumer behavior on e-commerce platforms, the attention mechanism can help the model automatically identify which user behaviors have an important impact on predicting consumer purchasing behavior. For example, the click behavior of some products may have more predictive value than the click behavior of other products, and the purchase history of some users may be more critical to predicting future behavior. Through attention weighting, the model can more accurately find these "key" behaviors in massive data, thereby improving the accuracy of prediction.

[0011] Although the consumer behavior prediction method based on deep learning and attention mechanism has high accuracy in theory, there are still many challenges in practical application: 1. Data quality issues: The data on e-commerce platforms often contain noise and omissions. How to achieve efficient prediction when the data quality is not high is an urgent problem to be solved.

[0012] 2. Real-time problem: Consumer behavior is highly dynamic. How to capture user behavior changes in real time and make quick decisions is still a difficult problem for most e-commerce platforms.

[0013] 3. Computational complexity problem: Deep learning models, especially models with attention mechanisms and iterative optimization, have high computational complexity. How to reduce computational costs while ensuring model performance is an aspect that needs to be optimized. Summary of the invention

[0014] In view of the above problems, the purpose of the present invention is to propose: a method for predicting consumer behavior and optimizing decision-making on an e-commerce platform based on attention iteration, comprising the following steps: S1. Data collection: collect e-commerce information from e-commerce platforms as needed, including consumer behavior data, transaction data, and product data; S2. Data preprocessing: preprocessing the e-commerce information according to usage requirements to obtain a preprocessed data set; S3, consumer classification: construct a vectorized data set, a click log sequence and a purchase behavior data set based on the preprocessed data set, classify the vectorized data set and the vectorized initial weight and value vector set, and perform weighted summation and recursive iterative calculation on the classified vectorized data set based on the attention mechanism; S4. Establishing a consumer prediction model: inputting the vectorized data set after weighted summation and recursive iterative calculation into the classifier to determine the consumers to be predicted; S5. Behavior prediction: predicting the subsequent behavior of the consumer to be predicted based on the consumer prediction model.

[0015] Furthermore, step S3 is specifically as follows: S31. Build vectorized data set: Determine sequence data according to the preprocessed data set, the sequence data satisfies Each click behavior corresponds to one purchase behavior, and L is a positive integer and satisfies ; The vectorized data set is It indicates that, Indicates the first User's Click behavior, Indicates the first Users in their Purchases completed after clicks, of which , , U and I represent the total number of users and the total number of clicks, respectively; S32, determine the click log sequence: according to the vectorized data set, determine the vectorized data set that meets condition T1 as the click log sequence, and the condition T1 is: at least continuous Clicks constitute a purchase; S33. Determine the purchase behavior data set: vectorize the data set Based on at least one vectorized data set All vectors of form a set, construct a purchase behavior data set, purchase behavior data set satisfy:

[0016] in, to Represents the vectorized data sets from the first user to the last user respectively.

[0017] Further, in step S31, satisfy or or ,in, is a positive integer, Indicates at least continuous Clicks constitute a purchase, and the acceptable error is 2 times. The value of N is changed based on the need for accuracy.

[0018] Furthermore, in step S31, the click behavior and the purchase behavior are expressed in vector form.

[0019] Further, step S3 includes the following steps: S34, cyclic classification: classifying the click log sequence and the purchase behavior data set according to different users; S35, constructing an initial weight vector and an initial value vector: according to the cyclic classification result, determining all click log sequences of the same user to construct an initial weight vector, and determining all subsequent purchase behaviors under the same click log sequence of the same user to construct an initial value vector; S36, attention weighting: performing weighted sum calculation on the initial value vector based on the attention mechanism of the neural network, finding the target purchase behavior with the greatest probability for all subsequent purchase behaviors relative to the click log sequence, and constructing a target value vector. All target value vectors for the click log sequence constitute a target vector set; S37. Iterative recursion: Use a recursive neural network to iteratively calculate the click log sequence. The initial value vectors of two adjacent click log sequences jointly determine the target value vector of the first click log sequence as the output, and determine the initial value vector of the second click log sequence as the output. The input and output are jointly used as the input of the subsequent recursive click log sequence. The neural network optimizes the weight vector of the neural network through continuous iteration until convergence.

[0020] Further, in step S35, the initial weight vector is The initial value vector is represented by express, and Respectively expressed in Click logs for users The corresponding initial weights and initial values; The target value vector is Indicates a mapping relationship; The weighted target value vector is express, represents the initial weight vector with a certain initial value vector as a parameter; The final delivery vector after recursive calculation is express, Indicates that in the iteration process, for the same initial value vector The different initial weight vectors obtained The maximum value selected from The maximum weight of the click log among the users of the e-commerce platform indicates The click log has the highest probability of being clicked among all e-commerce platform users, and the corresponding click vector represents the user who is most likely to convert into purchasing behavior.

[0021] Furthermore, in step S36, the initial value vector is calculated as follows:

[0022] in, represents the initial weight vector With the initial value vector Multiply and weight the target value vector , all target value vectors constitute the target vector set.

[0023] Furthermore, in step S36, the recursive equation of the first click log sequence is:

[0024] in, Indicates the purchase probability of the current user progressing to the next click.

[0025] Furthermore, in step S37, the recursive equation of the second click log sequence is:

[0026] in, Indicates the purchase probability of the next user at the current click volume.

[0027] Furthermore, in step S37, the click log sequence includes clicks on product promotional words, product pictures and product titles.

[0028] Beneficial Effects This scheme proposes a method for predicting consumer behavior and optimizing decision-making on e-commerce platforms based on attention iteration, which can significantly improve the ability of e-commerce platforms in predicting consumer behavior, optimizing decisions, and improving user experience. Specifically, it has the following significant beneficial effects: 1. Improve the accuracy of consumer behavior prediction: By introducing the attention mechanism and iterative optimization algorithm, this solution can effectively identify and focus on the key factors in consumer behavior. The attention mechanism can automatically assign weights to different click behaviors, purchase behaviors and other data, thereby accurately capturing which behaviors have a greater impact on predicting future consumer behavior. For example, some clicks of users may directly affect purchase decisions, while other behaviors may only be noise during the browsing process. By dynamically adjusting the weights, the model can automatically optimize the prediction results according to different scenarios and data conditions, thereby improving the accuracy of the prediction.

[0029] 2. Personalized recommendations and accurate decision-making: The implementation of this solution can accurately predict the user's next action based on the user's historical behavior data and real-time behavior, thereby providing personalized recommendations and decision optimization for e-commerce platforms. For example, when a user browses a certain category of goods, the platform can adjust the product recommendations in real time and optimize the display strategy based on the predicted purchase behavior. This personalized recommendation can not only improve the user experience, but also effectively increase conversion rate and sales.

[0030] 3. Optimize data processing flow and improve processing efficiency: In traditional e-commerce platforms, user behavior data is usually huge and complex. How to extract valuable information from massive data has always been a technical challenge. This solution converts raw data into high-quality data sets for model learning through steps such as data preprocessing, behavior classification, and vectorization. At the same time, by adopting iterative optimization of recursive neural networks, it can more efficiently process multi-level and multi-dimensional data, and can continuously optimize the model to improve the efficiency and accuracy of data processing.

[0031] 4. Better adapt to dynamic changes in user behavior: Consumer behavior is highly dynamic, and user interests and needs may change significantly over time. Traditional models often cannot capture these changes in real time, but this solution can dynamically adjust user behavior through the combination of recursive neural networks and attention mechanisms. In practical applications, the platform can quickly update the prediction model and make corresponding decisions in a timely manner when user behavior changes, thereby improving the flexibility and real-time performance of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] Figure 1 It is a schematic diagram of the method flow of the present invention. DETAILED DESCRIPTION

[0033] In order to deepen the understanding of the present invention, the present invention will be further described in detail below in conjunction with examples. The examples are only used to explain the present invention and do not constitute a limitation on the protection scope of the present invention.

[0034] Embodiment 1 according to Figure 1 As shown, this embodiment provides an e-commerce platform consumer behavior prediction and decision optimization method based on attention iteration, including the following steps: S1. Data collection: collect e-commerce information from e-commerce platforms as needed, including consumer behavior data, transaction data, and product data; S2. Data preprocessing: preprocessing the e-commerce information according to usage requirements to obtain a preprocessed data set; S3, consumer classification: construct a vectorized data set, a click log sequence and a purchase behavior data set based on the preprocessed data set, classify the vectorized data set and the vectorized initial weight and value vector set, and perform weighted summation and recursive iterative calculation on the classified vectorized data set based on the attention mechanism; S4. Establishing a consumer prediction model: inputting the vectorized data set after weighted summation and recursive iterative calculation into the classifier to determine the consumers to be predicted; S5. Behavior prediction: predicting the subsequent behavior of the consumer to be predicted based on the consumer prediction model.

[0035] Furthermore, step S3 is specifically as follows: S31. Build vectorized data set: Determine sequence data according to the preprocessed data set, the sequence data satisfies Each click behavior corresponds to one purchase behavior, and L is a positive integer and satisfies ; The vectorized data set is It indicates that, Indicates the first User's Click behavior, Indicates the first Users in their Purchases completed after clicks, of which , , U and I represent the total number of users and the total number of clicks, respectively; S32, determine the click log sequence: according to the vectorized data set, determine the vectorized data set that meets condition T1 as the click log sequence, and the condition T1 is: at least continuous Clicks constitute a purchase; S33. Determine the purchase behavior data set: vectorize the data set Based on at least one vectorized data set All vectors of form a set, construct a purchase behavior data set, purchase behavior data set satisfy:

[0036] in, to Represents the vectorized data sets from the first user to the last user respectively.

[0037] Further, in step S31, satisfy or or ,in, is a positive integer, Indicates at least continuous Clicks constitute a purchase, and the acceptable error is 2 times. The value of N is changed based on the need for accuracy.

[0038] Furthermore, in step S31, the click behavior and the purchase behavior are expressed in vector form.

[0039] Further, step S3 includes the following steps: S34, cyclic classification: classifying the click log sequence and the purchase behavior data set according to different users; S35, constructing an initial weight vector and an initial value vector: according to the cyclic classification result, determining all click log sequences of the same user to construct an initial weight vector, and determining all subsequent purchase behaviors under the same click log sequence of the same user to construct an initial value vector; S36, attention weighting: performing weighted sum calculation on the initial value vector based on the attention mechanism of the neural network, finding the target purchase behavior with the greatest probability for all subsequent purchase behaviors relative to the click log sequence, and constructing a target value vector. All target value vectors for the click log sequence constitute a target vector set; S37. Iterative recursion: Use a recursive neural network to iteratively calculate the click log sequence. The initial value vectors of two adjacent click log sequences jointly determine the target value vector of the first click log sequence as the output, and determine the initial value vector of the second click log sequence as the output. The input and output are jointly used as the input of the subsequent recursive click log sequence. The neural network optimizes the weight vector of the neural network through continuous iteration until convergence.

[0040] Further, in step S35, the initial weight vector is The initial value vector is represented by express, and Respectively expressed in Click logs for users The corresponding initial weights and initial values; The target value vector is Indicates a mapping relationship; The weighted target value vector is express, represents the initial weight vector with a certain initial value vector as a parameter; The final delivery vector after recursive calculation is express, Indicates that in the iteration process, for the same initial value vector The different initial weight vectors obtained The maximum value selected from The maximum weight of the click log among the users of the e-commerce platform indicates The click log has the highest probability of being clicked among all e-commerce platform users, and the corresponding click vector represents the user who is most likely to convert into purchasing behavior.

[0041] Furthermore, in step S36, the initial value vector is calculated as follows:

[0042] in, represents the initial weight vector With the initial value vector Multiply and weight the target value vector , all target value vectors constitute the target vector set.

[0043] Furthermore, in step S36, the recursive equation of the first click log sequence is:

[0044] in, Indicates the purchase probability of the current user progressing to the next click.

[0045] Furthermore, in step S37, the recursive equation of the second click log sequence is:

[0046] in, Indicates the purchase probability of the next user at the current click volume.

[0047] Furthermore, in step S37, the click log sequence includes clicks on product promotional words, product pictures and product titles. Specific embodiment 2 Suppose an e-commerce platform wants to optimize product recommendations and advertising strategies through in-depth analysis of consumer behavior. The platform has a large amount of user behavior data, including user click logs and purchase records. In order to improve the platform's operational efficiency, the platform hopes to accurately predict each user's future purchase behavior and make personalized recommendations based on these predictions.

[0049] Steps S1. Data Collection E-commerce platforms collect user behavior data on the platform, including user click logs (such as the pictures, titles, and promotional slogans of the products clicked), transaction data (such as detailed information on purchased products), and product data (such as product categories, prices, and inventory). Based on this data, the platform has established a complete database of users, products, and transactions.

[0050] S2. Data preprocessing After data collection, the platform cleans and preprocesses the raw data, removes invalid or missing data, and performs time series processing on user behavior data, sorting each user's click behavior in chronological order to generate a user behavior sequence. At the same time, the platform combines product information and transaction records with user behavior data to form a comprehensive preprocessed data set.

[0051] S3. Consumer Classification In this step, the platform first constructs a vectorized data set, which contains the click behavior vector and purchase behavior vector of each user. Taking user A as an example, user A's click log may contain the following data: click on product A for the first time, click on product B for the second time, and click on product C for the third time. On this basis, the platform determines whether these click behaviors will be converted into purchase behaviors according to certain rules. For example, if user A finally purchases product C after clicking on products A and B, this behavior is considered a purchase behavior.

[0052] The platform uses these vectorized data sets to perform weighted summation and recursive iterative calculations based on the attention mechanism to identify which behavior sequences are most important for predicting future purchase behavior. In this way, the platform can efficiently capture the key click log sequences that affect purchase decisions and weight them.

[0053] S4. Build a consumer prediction model Next, the platform inputs the weighted summed vectorized data set into a classifier and trains a consumer behavior prediction model through a machine learning algorithm. The model can predict the user's future purchase behavior based on his or her historical behavior. For example, by analyzing user A's historical click records and purchase behavior, the model can predict the type of goods that user A may buy next, or may respond to certain promotions.

[0054] S5. Behavior prediction Based on the consumer prediction model constructed above, the platform can predict the subsequent behavior of each user. Taking user B as an example, the prediction model can analyze that user B is likely to click on a certain product or participate in a certain promotion in the future. The prediction results not only help the platform understand user needs, but also provide decision-making basis for the personalized recommendation system, thereby improving the platform's conversion rate and sales.

[0055] Detailed description of the specific steps S31. Constructing vectorized data sets Assume that the platform has 100 users, and each user's behavior sequence is converted into a vector. For example, the vectorized data set of user A may contain multiple click behaviors and subsequent purchase behaviors. Each click behavior and purchase behavior is represented in vector form as follows: Indicates user No. Click behavior; Indicates user No. Purchase behavior after clicks.

[0056] With these data, the platform can construct a complete set of user behavior vectors.

[0057] S32. Determine click log sequence Based on the preprocessed data, the platform extracts the click log sequence that meets condition T1. Condition T1 requires at least L consecutive clicks to constitute a purchase behavior. For example, if a user clicks on product A, product B, and product C in succession, and finally purchases product C, then this behavior sequence will be considered to meet condition T1.

[0058] S33. Determine the purchase behavior data set The platform further constructs a purchase behavior data set, which includes the click logs and purchase behaviors of all users. For each user, the platform combines their click behavior and purchase behavior into a set of vectors in order to form a purchase behavior data set. This set can help the platform understand the purchase patterns of different users on the platform, and then guide subsequent recommendations and advertising.

[0059] S34, Cycle Classification In this step, the platform classifies the click log sequences and purchase behavior data sets of all users. Each user's behavior sequence is processed independently to ensure that each user's behavior data can accurately reflect their unique consumption habits.

[0060] S35. Constructing initial weight vector and initial value vector According to the classification results of each user, the platform assigns an initial weight vector and an initial value vector to each user's click log sequence. The initial weight vector reflects the user's attention to each click behavior, while the initial value vector represents the purchase probability that the user may generate after the behavior.

[0061] S36, Attention Weighting With the help of the attention mechanism, the platform performs a weighted summation of the initial value vectors and calculates the target value vector that is most likely to be converted into a purchase behavior for each click log sequence. For example, the platform can identify that user A’s second click behavior (clicking on product B) is most likely to trigger a purchase behavior, so it assigns a higher weight to this behavior.

[0062] S37, Iterative recursion Finally, the platform iteratively calculates the click log sequence through a recursive neural network. The output of each click log sequence will be used as the input of the next sequence until the model converges. This iterative process helps the platform continuously optimize the prediction results and improve the accuracy of consumer behavior prediction.

[0063] Through the above steps, e-commerce platforms can use the attention iteration mechanism to efficiently predict users' purchasing behavior and optimize corresponding decisions. The platform can not only accurately identify which behaviors have a significant impact on purchasing decisions, but also provide personalized product recommendations and advertising strategies based on the prediction results, thereby improving the platform's conversion rate and user satisfaction.

[0064] The above shows and describes the basic principles, main features and advantages of the present invention. It should be understood by those skilled in the art that the present invention is not limited to the above embodiments. The above embodiments and descriptions are only for explaining the principles of the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, which fall within the scope of the present invention. The scope of protection of the present invention is defined by the attached claims and their equivalents.

Claims

1. An e-commerce platform consumer behavior prediction and decision optimization method based on attention iteration, characterized in that: The following steps are involved: S1. Data collection: collect e-commerce information from e-commerce platforms as needed, including consumer behavior data, transaction data, and product data; S2. Data preprocessing: preprocessing the e-commerce information according to usage requirements to obtain a preprocessed data set; S3, consumer classification: construct a vectorized data set, a click log sequence and a purchase behavior data set based on the preprocessed data set, classify the vectorized data set and the vectorized initial weight and value vector set, and perform weighted summation and recursive iterative calculation on the classified vectorized data set based on the attention mechanism; S4. Establishing a consumer prediction model: inputting the vectorized data set after weighted summation and recursive iterative calculation into the classifier to determine the consumers to be predicted; S5. Behavior prediction: predicting the subsequent behavior of the consumer to be predicted based on the consumer prediction model.

2. The method for predicting consumer behavior and optimizing decision-making on an e-commerce platform based on attention iteration according to claim 1, characterized in that: Step S3 is specifically as follows: S31. Build vectorized data set: Determine sequence data according to the preprocessed data set, the sequence data satisfies Each click behavior corresponds to one purchase behavior, and L is a positive integer and satisfies ; The vectorized data set is It indicates that, Indicates the first User's Click behavior, Indicates the first Users in their Purchases completed after clicks; , , U and I represent the total number of users and the total number of clicks, respectively; S32, determine the click log sequence: according to the vectorized data set, determine the vectorized data set that meets condition T1 as the click log sequence, and the condition T1 is: at least continuous Clicks constitute a purchase; S33. Determine the purchase behavior data set: vectorize the data set Based on at least one vectorized data set All vectors of form a set, construct a purchase behavior data set, purchase behavior data set satisfy: ； in, to Represents the vectorized data sets from the first user to the last user respectively.

3. The method for predicting consumer behavior and optimizing decision-making on an e-commerce platform based on attention iteration according to claim 2, characterized in that: In step S31, satisfy or or ,in, is a positive integer, Indicates at least continuous Clicks constitute a purchase, and the acceptable error is 2 times. The value of N is changed based on the need for accuracy.

4. The method for predicting consumer behavior and optimizing decision-making on an e-commerce platform based on attention iteration according to claim 2, characterized in that: In step S31, the click behavior and the purchase behavior are expressed in vector form.

5. The method for predicting consumer behavior and optimizing decision-making on an e-commerce platform based on attention iteration according to claim 2, characterized in that: Step S3 includes the following steps: S34, cyclic classification: classifying the click log sequence and the purchase behavior data set according to different users; S35, constructing an initial weight vector and an initial value vector: according to the cyclic classification result, determining all click log sequences of the same user to construct an initial weight vector, and determining all subsequent purchase behaviors under the same click log sequence of the same user to construct an initial value vector; S36, attention weighting: performing weighted sum calculation on the initial value vector based on the attention mechanism of the neural network, finding the target purchase behavior with the greatest probability for all subsequent purchase behaviors relative to the click log sequence, and constructing a target value vector. All target value vectors for the click log sequence constitute a target vector set; S37. Iterative recursion: Use a recursive neural network to iteratively calculate the click log sequence. The initial value vectors of two adjacent click log sequences jointly determine the target value vector of the first click log sequence as the output, and determine the initial value vector of the second click log sequence as the output. The input and output are jointly used as the input of the subsequent recursive click log sequence. The neural network optimizes the weight vector of the neural network through continuous iteration until convergence.

6. The method for predicting consumer behavior and optimizing decision-making on an e-commerce platform based on attention iteration according to claim 5, characterized in that: In step S35, the initial weight vector is The initial value vector is represented by express, and Respectively expressed in Click logs for users The corresponding initial weights and initial values; The target value vector is Indicates a mapping relationship; The weighted target value vector is express, represents the initial weight vector with a certain initial value vector as a parameter; The final delivery vector after recursive calculation is express, Indicates that in the iteration process, for the same initial value vector The different initial weight vectors obtained The maximum value selected from The maximum weight of the click log among the users of the e-commerce platform indicates The click log has the highest probability of being clicked among all e-commerce platform users, and the corresponding click vector represents the user who is most likely to convert into purchasing behavior.

7. The method for predicting consumer behavior and optimizing decision-making on an e-commerce platform based on attention iteration according to claim 6, characterized in that: In step S36, the initial value vector is calculated as follows: ； in, represents the initial weight vector With the initial value vector Multiply and weight the target value vector , all target value vectors constitute the target vector set.

8. The method for predicting consumer behavior and optimizing decision-making on an e-commerce platform based on attention iteration according to claim 7, characterized in that: In step S36, the recursive equation of the first click log sequence is: ； in, Indicates the purchase probability of the current user progressing to the next click.

9. The method for predicting consumer behavior and optimizing decision-making on an e-commerce platform based on attention iteration according to claim 8, characterized in that: In step S37, the recursive equation of the second click log sequence is: ； in, Indicates the purchase probability of the next user at the current click volume.

10. The method for predicting consumer behavior and optimizing decision-making on an e-commerce platform based on attention iteration according to claim 9, characterized in that: In step S37, the click log sequence includes product promotion words, product pictures and product titles.

Images