O2O platform user portrait construction method based on deep learning

Abstract

The invention discloses an O2O platform user portrait construction method based on deep learning, and relates to the technical field of deep learning user portrait construction. According to the O2O platform user portrait construction method based on deep learning, group classification is performed on users through a classification and regression tree, user behavior features are identified, a user portrait decision tree model is constructed, user behavior data are predicted, user feature vectors are generated, and an O2O platform user portrait is constructed based on the user feature vectors. And analyzing the matching degree between the subsequent behavior data of the user and the O2O platform user portrait, and through comprehensive analysis with service indexes, obtaining a platform user portrait optimization index, establishing a real-time updating and iterative optimization mechanism, updating a user portrait decision tree model, optimizing the O2O platform user portrait, and improving the user portrait precision. The O2O platform is helped to better understand the behavior pattern and preference of the user, so that the marketing strategy is optimized, the platform is helped to better meet the user demand, and the user loyalty and the market share are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of deep learning to construct user portraits, and specifically to a method for constructing user portraits for an O2O platform based on deep learning. Background Art

[0002] With the popularization of the Internet and smartphones, users' lifestyles have changed significantly. More and more users are accustomed to shopping, restaurant reservations, travel arrangements, etc. through online platforms. The user portrait of the O2O platform needs to reflect this change in digital lifestyle in order to accurately capture user needs. Deep learning can achieve excellent performance in image recognition, natural language processing, speech recognition and other fields through a multi-layer neural network structure. These technological advances provide technical support for the accurate construction of user portraits on O2O platforms.

[0003] For example, a Chinese invention patent with announcement number CN116523545B discloses a method for constructing user portraits based on big data. The method collects a number of conditional features triggered by the user end at preset collection periods, calculates the conditional feature parameters corresponding to the portrait tag, determines whether to store the portrait tag and the number of conditional features triggered by the user end in the user portrait set corresponding to the user end, and determines the label type of the portrait label in the user portrait set. When the ratio of the number of fuzzy portrait labels to standard portrait labels in the user portrait set is higher than a preset ratio comparison threshold, determines the correlation between each fuzzy portrait label and the standard portrait label, optimizes the fuzzy portrait labels in the user portrait set, and adjusts the preset collection period when the number of continuously stored duplicate portrait labels is greater than the preset storage threshold, thereby improving the efficiency and accuracy of constructing portrait labels for the user end.

[0004] Traditional user portrait construction methods mostly rely on rule engines or simple machine learning algorithms, which are difficult to handle large-scale, multi-source, heterogeneous data, and have limitations in capturing dynamically changing user behaviors. They also rely on manual experience and simple statistical analysis, making it difficult to fully and accurately capture the complexity and dynamics of user behaviors. At the same time, when determining the correlation between each fuzzy portrait label and the standard portrait label, it is impossible to update and iterate the optimization in real time during the optimization process of the fuzzy portrait labels in the user portrait set. Summary of the invention

[0005] In view of the deficiencies in the prior art, the present invention provides a method for constructing a user portrait of an O2O platform based on deep learning, which solves the problems of the above-mentioned background technology.

[0006] To achieve the above objectives, the present invention is implemented through the following technical solutions: a method for constructing an O2O platform user portrait based on deep learning, comprising the following steps: S1. Collecting user behavior data from an O2O platform data source, and establishing a comprehensive user behavior data set through a clustering algorithm; S2. Based on the comprehensive user behavior data set, classifying users into groups through classification and regression trees, identifying user behavior characteristics, and constructing a user portrait decision tree model; S3. Using the user portrait decision tree model to predict user behavior data, generating user feature vectors, and constructing an O2O platform user portrait based on the user feature vectors; S4. Obtaining a platform user portrait optimization index through the matching degree and business indicator data of subsequent user behavior data in the O2O platform data source and the O2O platform user portrait, establishing a real-time update and iterative optimization mechanism, updating the user portrait decision tree model, and optimizing the O2O platform user portrait.

[0007] Furthermore, the specific process of establishing a comprehensive user behavior data set through a clustering algorithm is as follows: user behavior data is obtained through the O2O platform data source, and data cleaning and user behavior features are extracted. User behavior features include user activity, user purchasing power, and browsed product types. The K-means algorithm is used to perform cluster analysis on users with similar user behavior characteristics. According to the clustering results, users are divided into different clusters, and the average value of each cluster feature is calculated as the cluster center to generate a comprehensive user behavior data set.

[0008] Furthermore, based on the comprehensive user behavior data set, the specific process of classifying users into groups through classification and regression trees is as follows: load the comprehensive user behavior data set, divide the comprehensive user behavior data set into a training set and a test set, use the classification and regression tree algorithm to traverse the user behavior characteristics of the training set, and divide the group characteristics according to the user behavior characteristics, calculate the Gini index of each feature after division, select the feature with the smallest Gini index as the division feature, and build a classification and regression tree layer by layer through the O2O platform product classification, each node represents a user group, and each leaf node represents a user group classification.

[0009] Furthermore, the specific process of identifying user behavior characteristics and constructing a user portrait decision tree model is as follows: select a combination of deep neural networks and classification and regression trees as the baseline model of the user portrait decision tree model, input the user behavior characteristics of the training set, use the deep neural network to perform multi-level abstraction and non-linear feature combination on the user behavior characteristics, extract high-level behavior characteristics, and use the user behavior characteristics of the test set to train the baseline model of the user portrait decision tree model to construct a user portrait decision tree model.

[0010] Furthermore, the user portrait decision tree model is used to predict user behavior data. The specific process of generating a user feature vector is as follows: traverse the user behavior data through the user portrait decision tree model, starting from the root node, and divide the nodes layer by layer according to the feature value until reaching the leaf node; at each node, compare the user behavior feature value with the node division feature, select the corresponding child node, and output the user feature vector.

[0011] Furthermore, through the matching degree of user subsequent behavior data and O2O platform user portrait in the O2O platform data source and the business indicator data, the specific process of obtaining the platform user portrait optimization index is as follows: similar features are extracted from the user subsequent behavior data and the O2O platform user portrait, and the cosine similarity between the user behavior feature vector and the user portrait feature vector is calculated to obtain the feature vector matching index; the user conversion rate and user click-through rate in the business indicator data are comprehensively calculated with the standard values ​​of the user conversion rate and the standard values ​​of the user click-through rate to obtain the business compliance index; the feature vector matching index and the business compliance index are comprehensively calculated to obtain the platform user portrait optimization index.

[0012] Furthermore, the specific process of establishing a real-time update and iterative optimization mechanism is as follows: conduct real-time analysis on the platform user portrait optimization index, compare the platform user portrait optimization index with the set platform user portrait optimization update level threshold, and obtain O2O platform user portrait optimization information; when the platform user portrait optimization index is less than the first-level update threshold, mark the user portrait as a high-update iterative optimization portrait; when the platform user portrait optimization index is less than the second-level update threshold and greater than the first-level update threshold, mark the user portrait as a medium-update iterative optimization portrait; when the platform user portrait optimization index is less than the third-level update threshold and greater than the second-level update threshold, mark the user portrait as a low-update iterative optimization portrait.

[0013] Furthermore, the specific process of updating the user portrait decision tree model and optimizing the user portrait of the O2O platform is as follows: through the O2O platform user portrait optimization information, identify the user portrait features that need to be improved, use the incremental learning algorithm to integrate the optimization information and feedback data into the user portrait decision tree model, retrain the model, and verify the updated user portrait decision tree model through cross-validation and model evaluation methods.

[0014] The present invention has the following beneficial effects:

[0015] (1) This method for constructing user portraits for an O2O platform based on deep learning uses classification and regression trees to classify users into groups, identify user behavior characteristics, and construct a user portrait decision tree model, thereby enhancing the O2O platform's understanding of user behavior and significantly improving the accuracy and efficiency of user analysis. It also enhances the ability to make personalized recommendations and optimize user experience, thereby improving the competitive advantage and market performance of the O2O platform. At the same time, through the matching degree of subsequent user behavior data in the O2O platform data source and the O2O platform user portrait and business indicator data, the platform user portrait optimization index is obtained, and a real-time update and iterative optimization mechanism is established to improve the user screen recognition accuracy.

[0016] (2) The method for constructing user portraits for O2O platforms based on deep learning uses a user portrait decision tree model to predict user behavior data, generate user feature vectors, and construct O2O platform user portraits based on user feature vectors. This can help O2O platforms better understand user behavior patterns and preferences, thereby optimizing marketing strategies. Having accurate user portraits can help platforms better meet user needs and improve user loyalty and market share.

[0017] Of course, any product implementing the present invention does not necessarily need to achieve all of the advantages described above at the same time. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 This is a flow chart of the method for constructing user portraits for an O2O platform based on deep learning in the present invention.

[0019] Figure 2 A flow chart of a user portrait decision tree model is constructed for the present invention. DETAILED DESCRIPTION

[0020] The embodiment of the present application solves the limitation of traditional methods that rely on rule engines or simple machine learning algorithms and cannot effectively process large-scale, multi-source, and heterogeneous data through a method for constructing user portraits of an O2O platform based on deep learning.

[0021] The overall idea of ​​the problem in the embodiment of this application is as follows:

[0022] Collect user behavior data from the O2O platform data source and establish a comprehensive user behavior data set through clustering algorithms.

[0023] Based on the comprehensive user behavior data set, users are classified into groups through classification and regression trees, user behavior characteristics are identified, and a user portrait decision tree model is constructed.

[0024] Use the user portrait decision tree model to predict user behavior data, generate user feature vectors, and build O2O platform user portraits based on user feature vectors.

[0025] Through the matching degree and business indicator data of the user's subsequent behavior data in the O2O platform data source and the O2O platform user portrait, the platform user portrait optimization index is obtained, a real-time update and iterative optimization mechanism is established, the user portrait decision tree model is updated, and the O2O platform user portrait is optimized.

[0026] See also Figure 1 , an embodiment of the present invention provides a technical solution: a method for constructing an O2O platform user portrait based on deep learning, comprising the following steps: S1. collecting user behavior data from an O2O platform data source, and establishing a comprehensive user behavior data set through a clustering algorithm; S2. based on the comprehensive user behavior data set, classifying users into groups through a classification and regression tree, identifying user behavior characteristics, and constructing a user portrait decision tree model; S3. using the user portrait decision tree model to predict user behavior data, generating a user feature vector, and constructing an O2O platform user portrait based on the user feature vector; S4. obtaining a platform user portrait optimization index through the matching degree and business indicator data of subsequent user behavior data in the O2O platform data source and the O2O platform user portrait, establishing a real-time update and iterative optimization mechanism, updating the user portrait decision tree model, and optimizing the O2O platform user portrait.

[0027] In this implementation scheme, in step S1, user behavior data is collected from multi-source data of the O2O platform, and users with similar user behavior characteristics are grouped using a clustering algorithm. Clustering algorithms help to discover potential patterns and groups in the data, thereby establishing a comprehensive data set containing diverse behavioral characteristics. In step S2, a classification and regression tree model is used to divide the user behavior comprehensive data set into different groups and identify the main behavioral characteristics of each group. This step helps to build a user portrait decision tree model, which effectively captures and characterizes the behavioral habits and preferences of different user groups through the hierarchical structure of the tree. In step S3, the constructed user portrait decision tree model is used to predict new user behavior data and generate user feature vectors. These feature vectors reflect the personalized characteristics and behavior patterns of users, and provide a basis for building user portraits of the O2O platform. In step S4, user feedback data collected from the O2O platform is used to establish a real-time update and optimization mechanism to continuously improve the user portrait decision tree model. Through incremental learning and model updates, user portraits are continuously optimized to ensure real-time matching and accuracy with user behavior.

[0028] Specifically, the specific process of establishing a comprehensive user behavior data set through a clustering algorithm is as follows: user behavior data is obtained through the O2O platform data source, and data cleaning and user behavior features are extracted. User behavior features include user activity, user purchasing power, and browsed product types. The K-means algorithm is used to perform cluster analysis on users with similar user behavior characteristics. According to the clustering results, users are divided into different clusters, and the average value of each cluster feature is calculated as the cluster center to generate a comprehensive user behavior data set.

[0029] In this implementation scheme, user activity refers to the frequency and length of time that users are active on the platform, which can be measured by the number of logins and the duration of visits; purchasing power refers to the frequency and amount of purchases made by users on the platform, which can be measured by the number of orders and the amount of orders; browsed product types refer to the types or categories of products that users are interested in, which can be analyzed by the types of products browsed by users. The K-means algorithm is a commonly used clustering analysis method, which is used to group users with similar characteristics into different clusters. The clusters are divided according to the product types of the O2O platform. The K-means algorithm is used to perform cluster analysis on users with similar user behavior characteristics. Users are divided into different clusters based on the clustering results, and the average value of each cluster feature is calculated as the cluster center. An example of generating a comprehensive user behavior data set is as follows: The two feature tables of user activity A and purchasing power P are as follows

[0030]

[0031]

[0032] Use the K-means algorithm to cluster these users according to their activity and purchasing power, and randomly select two initial clusters 1 and 2. The corresponding cluster centers u A =(0.6, 0.8), u B =(0.2, 0.3), u A represents the cluster center of cluster 1, u B Represents the cluster center of cluster 2. The Euclidean distance is used to calculate the distance from the user to the cluster center. The Euclidean distance formula is as follows: d represents the distance from the user to the cluster center, and the distance from user 1 to cluster u is obtained. A The distance from the center is 0, to cluster u A The distance from the center is 0.64, and the user 2 is obtained from cluster u A The distance from the center is 0.22, to cluster u A The distance from the center is 0.72, and user 3 is obtained from cluster u A The distance from the center is 0.64, to cluster u A The distance from the center is 0, and user 4 is obtained from cluster u A The distance from the center is 0.22, to cluster u AThe center distance is 0.42. Users 1, 2, and 4 are closest to cluster 1 and are assigned to cluster 1. User 3 is assigned to cluster 2. The calculation formula for updating the cluster center is as follows: In the formula, ud represents the updated cluster center, C represents the number of cluster data points, and x i represents the i-th data point in the cluster, and the updated cluster center is u A1 =(0.633, 0.7), u B2 =(0.2, 0.3). A1 represents the updated cluster center of cluster 1, u B2 Represents the updated cluster center of cluster 2, assigns each user to the final cluster, and uses the cluster center as the user's comprehensive behavioral dataset.

[0033] Specifically, based on the comprehensive user behavior data set, the specific process of classifying users into groups through classification and regression trees is as follows: load the comprehensive user behavior data set, divide the comprehensive user behavior data set into a training set and a test set, use the classification and regression tree algorithm to traverse the user behavior features of the training set, and divide the group features according to the user behavior features, calculate the Gini index of each feature after division, select the feature with the smallest Gini index as the division feature, and build a classification and regression tree layer by layer through the 020 platform product classification. Each node represents a user group, and each leaf node represents a user group classification.

[0034] In this implementation scheme, group characteristics are divided according to user behavior characteristics, the Gini index after each feature division is calculated, and the feature with the smallest Gini index is selected as the division feature. The example of building a classification and regression tree layer by layer through the 020 platform product classification is as follows: through the two feature tables of user activity A and purchasing power P, the activity is used as the division feature, and the activity division point is set to 0.65. Cluster 1 in the user behavior comprehensive data set is used for division. Users 1 and 4 have an activity less than 0.65 and are divided into the left child node. User 2 is divided into the right child node. The calculation formula of the node Gini index is as follows: Gin = 1-(p1 2 -p2 2 ), where Gin represents the Gini index of the left child node, p1 represents the proportion of p1 in the node, and p2 represents the proportion of p2 in the node. The total node Gini index is 0.333.

[0035] See also Figure 2,Specifically, the specific process of identifying user behavior ,features and building a user portrait decision tree model is as follows: ,select a combination of a deep neural network and a ,classification and regression tree as the baseline model of the user ,portrait decision tree model, input the user behavior ,features of the training set, use the deep neural network to perform multi-level ,abstract and non-linear feature combination of the user behavior ,features to extract high-level behavior features, and use the user behavior ,features of the test set to train the baseline model of the ,user portrait decision tree model, and build the user portrait decision tree model.

[0036] In this embodiment, the deep neural network can learn complex user behavior features through multi-level abstraction and nonlinear feature combinations. This enables the model to handle more abstract and high-level feature representations, thereby better capturing the complex patterns behind user behavior. The classification and regression tree algorithm recursively divides the feature space into different regions and classifies or regresses the data in each region. It can provide intuitive decision rules and explain the relationship between features. The deep neural network processes the input user behavior features through multiple hidden layers, and through weight learning, it can extract higher-level and more abstract feature representations from the original data. These features can capture the nonlinear relationships and complex patterns of user behavior and improve the expressive power of the model. The test set is used to evaluate the generalization ability of the model, that is, the performance of the model on unseen data. During the training process, it is necessary to ensure that the model not only performs well on the training set, but also can be effectively extended to the test set, so as to build a user portrait decision tree model with good generalization ability.

[0037] Specifically, the user portrait decision tree model is used to predict user behavior data. The specific process of generating a user feature vector is as follows: the user behavior data is traversed through the user portrait decision tree model, starting from the root node, and node division is performed layer by layer according to the feature value until the leaf node is reached; at each node, the user behavior feature value is compared with the node division feature, the corresponding child node is selected, and the user feature vector is output.

[0038] In this embodiment, traversal refers to inputting the user's behavior data into the user portrait decision tree model for processing and prediction. The behavior data of each user can include multiple behaviors such as clicking, purchasing, and browsing. The user portrait decision tree model is a tree structure, starting from the root node, and dividing the nodes according to specific feature values. The basis for division is usually a specific feature in the user behavior data, such as the number of clicks and the purchase amount. The model will assign the user behavior data to the corresponding child nodes according to the division characteristics of each node. This process continues until it reaches the leaf node, which represents the final user group classification. At each node, the model compares the relationship between the user's specific behavior feature value and the node division feature. For example, if the division feature is that the purchase amount is greater than 100 yuan, and the user's purchase amount is 120 yuan, then the user is assigned to the eligible child node. Finally, by traversing the entire decision tree, a user feature vector can be obtained based on the user's behavior data. This vector usually contains features of multiple dimensions, such as the user's interests, purchase preferences, and behavioral activity, which are used to describe and characterize the user's attributes and behavior patterns.

[0039] Specifically, the specific process of constructing an O2O platform user portrait based on a user feature vector is as follows: map the user feature vector to a predefined user portrait dimension, generate the eigenvalue of the user feature vector, identify the group to which the user belongs based on the eigenvalue of the user feature vector, and construct an O2O platform user portrait. The O2O platform user portrait includes the user's key behavioral characteristics, preferences, and group information.

[0040] In this implementation scheme, first, the user feature vector is mapped to a predefined user portrait dimension. A user feature vector is usually a vector containing multiple eigenvalues, which can represent the user's behavior, preferences, and interests. For each user feature vector, a corresponding eigenvalue is generated according to the predefined user portrait dimension. The eigenvalue is a numerical and categorical label used to describe the user's behavioral characteristics and preferences. If the activity in the user feature vector is 0.6, the purchasing power is 0.8, and the product preference is electronic products, the generated eigenvalues ​​can be high activity, high purchasing power, and preference for electronic products.

[0041] Specifically, through the matching degree of user subsequent behavior data and 020 platform user portrait in the 020 platform data source and the business indicator data, the specific process of obtaining the platform user portrait optimization index is as follows: extract similar features of user subsequent behavior data and 020 platform user portrait, calculate the cosine similarity of user behavior feature vector and user portrait feature vector, and obtain the feature vector matching index; perform comprehensive calculation on user conversion rate and user click-through rate in the business indicator data with the standard values ​​of user conversion rate and user click-through rate to obtain the business compliance index; perform comprehensive calculation on the feature vector matching index and the business compliance index to obtain the platform user portrait optimization index.

[0042] In this implementation plan, the user's click, browse, purchase, evaluation and other behavioral data are collected, cleaned and encoded from the data source of the 020 platform, and converted into user behavior feature vectors. The calculation formula of the feature vector matching index is as follows: In the formula, Vdf represents the user behavior feature vector, Vrg represents the user portrait feature vector, εy represents the feature vector matching index, βGB represents the standard value of the cosine similarity between vectors, and the calculation formula of the business compliance index is as follows: In the formula, It represents the business compliance index, which is used to indicate whether the user's behavior meets the business standard. ER represents the user conversion rate, RG represents the standard value of the user's click-through rate, ΔFc represents the allowable deviation value of the user's click-through rate, QZ represents the user's click-through rate, DB represents the standard value of the user's click-through rate, and ΔBN represents the allowable deviation value of the user's click-through rate. The calculation formula of the platform user portrait optimization index is as follows: In the formula, ω represents the platform user portrait optimization index, ε1 represents the weight factor of the feature vector matching index, and ε2 represents the weight factor of the business compliance index.

[0043] Specifically, the specific process of establishing a real-time update and iterative optimization mechanism is as follows: conduct real-time analysis of the platform user portrait optimization index, compare the platform user portrait optimization index with the set platform user portrait optimization update level threshold, and obtain 020 platform user portrait optimization information; when the platform user portrait optimization index is less than the first-level update threshold, mark the user portrait as a high-update iterative optimization portrait; when the platform user portrait optimization index is less than the second-level update threshold and greater than the first-level update threshold, mark the user portrait as a medium-update iterative optimization portrait; when the platform user portrait optimization index is less than the third-level update threshold and greater than the second-level update threshold, mark the user portrait as a low-update iterative optimization portrait.

[0044] In this implementation scheme, through real-time analysis and optimization of user portraits, it can be ensured that user portraits always reflect the latest user behaviors and preferences, thereby improving the accuracy and reliability of the portraits. The real-time update mechanism enables the platform to respond quickly to changes in user behavior, adjust marketing strategies and service content in a timely manner, and enhance the flexibility and competitiveness of the platform. Through real-time data analysis and optimization, the platform can make more scientific and efficient decisions based on real data, reducing blindness and decision-making risks.

[0045] Specifically, the specific process of updating the user portrait decision tree model and optimizing the user portrait of the O2O platform is as follows: based on the optimization information obtained from the user feedback data, identify the user portrait features that need to be improved, use the incremental learning algorithm to integrate the optimization information and feedback data into the user portrait decision tree model, retrain the model, and verify the updated user portrait decision tree model through cross-validation and model evaluation methods.

[0046] In this implementation scheme, user feedback data is used to analyze the current user portrait model and identify features that need to be improved. These features include user behavior characteristics, preferences, and group classifications. The identified user portrait features that need to be improved are integrated into the user portrait decision tree model with the optimization information in the user feedback data. The incremental learning algorithm allows the model to gradually learn and improve based on new data, rather than completely retraining the entire model. Retrain the user portrait decision tree model using the integrated data set. Ensure that the model can be updated with new feedback data and optimization information. Retraining involves adjusting model parameters, optimizing loss functions, and updating model structures to better reflect new features and behaviors of users. Use cross-validation techniques to evaluate the performance of the updated user portrait decision tree model. Cross-validation can help evaluate the generalization ability of the model and avoid overfitting problems.

[0047] In summary, this application has at least the following effects:

[0048] The O2O platform user portrait construction method based on deep learning uses the user portrait decision tree model to predict user behavior data, generate user feature vectors, and construct O2O platform user portraits based on user feature vectors. This can help the O2O platform better understand user behavior patterns and preferences, thereby optimizing marketing strategies. Having accurate user portraits can help the platform better meet user needs and improve user loyalty and market share.

[0049] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0050] The present invention is described with reference to flowcharts and / or block diagrams of systems, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0051] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0052] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0053] Although the preferred embodiments of the present invention have been described, those skilled in the art may make additional changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present invention.

[0054] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A method for constructing user portraits of an O2O platform based on deep learning, characterized in that: The following steps are involved: S1. Collect user behavior data from the O2O platform data source and establish a comprehensive user behavior data set through clustering algorithms; S2. Based on the comprehensive user behavior data set, classify users into groups through classification and regression trees, identify user behavior characteristics, and build a user portrait decision tree model; S3. Use the user portrait decision tree model to predict user behavior data, generate user feature vectors, and build O2O platform user portraits based on the user feature vectors; S4. Obtain the platform user portrait optimization index through the matching degree and business indicator data of the user's subsequent behavior data in the O2O platform data source and the O2O platform user portrait, establish a real-time update and iterative optimization mechanism, update the user portrait decision tree model, and optimize the O2O platform user portrait.

2. The method for constructing an O2O platform user portrait based on deep learning according to claim 1, characterized in that: The specific process of establishing a comprehensive user behavior data set through clustering algorithms is as follows: User behavior data is obtained through the O2O platform data source, and data cleaning and user behavior features are extracted. User behavior features include user activity, user purchasing power and browsed product types. The K-means algorithm is used to perform cluster analysis on users with similar user behavior features. According to the clustering results, users are divided into different clusters, and the average value of each cluster feature is calculated as the cluster center to generate a comprehensive user behavior data set.

3. The method for constructing an O2O platform user portrait based on deep learning according to claim 2, characterized in that: Based on the comprehensive user behavior data set, the specific process of group classification of users through classification and regression trees is as follows: Load the comprehensive user behavior dataset and divide it into a training set and a test set. Use the classification and regression tree algorithm to traverse the user behavior features of the training set, divide the group features according to the user behavior features, calculate the Gini index of each feature after division, select the feature with the smallest Gini index as the division feature, and build a classification and regression tree layer by layer through the O2O platform product classification. Each node represents a user group, and each leaf node represents a user group classification.

4. The method for constructing an O2O platform user portrait based on deep learning according to claim 3 is characterized in that: The specific process of identifying user behavior characteristics and building a user portrait decision tree model is as follows: A combination of deep neural network and classification and regression tree is selected as the benchmark model of the user portrait decision tree model. The user behavior characteristics of the training set are input, and the deep neural network is used to perform multi-level abstraction and nonlinear feature combination on the user behavior characteristics to extract high-level behavior features. The user behavior characteristics of the test set are used to train the benchmark model of the user portrait decision tree model to construct the user portrait decision tree model.

5. The method for constructing an O2O platform user portrait based on deep learning according to claim 4, characterized in that: The specific process of using the user portrait decision tree model to predict user behavior data and generate user feature vectors is as follows: Traverse the user behavior data through the user portrait decision tree model, starting from the root node, and divide the nodes layer by layer according to the feature values ​​until reaching the leaf node; At each node, the user behavior feature value is compared with the node partition feature, the corresponding child node is selected, and the user feature vector is output.

6. The method for constructing an O2O platform user portrait based on deep learning according to claim 5, characterized in that: The specific process of constructing O2O platform user portraits based on user feature vectors is as follows: Map the user feature vector to the predefined user portrait dimension, generate the eigenvalue of the user feature vector, identify the group to which the user belongs based on the eigenvalue of the user feature vector, and build the O2O platform user portrait. The O2O platform user portrait includes the user's key behavioral characteristics, preferences and group information.

7. The method for constructing an O2O platform user portrait based on deep learning according to claim 6, characterized in that: The specific process of obtaining the platform user portrait optimization index through the matching degree and business indicator data of the user's subsequent behavior data in the O2O platform data source and the O2O platform user portrait is as follows: Extract similar features from the user's subsequent behavior data and the O2O platform user portrait, calculate the cosine similarity between the user behavior feature vector and the user portrait feature vector, and obtain the feature vector matching index; Perform comprehensive calculation on the user conversion rate and user click-through rate in the business indicator data and the standard values ​​of the user conversion rate and the standard values ​​of the user click-through rate to obtain the business target index; The feature vector matching index and the business compliance index are comprehensively calculated to obtain the platform user portrait optimization index.

8. The method for constructing an O2O platform user portrait based on deep learning according to claim 7, characterized in that: The specific process of establishing a real-time update and iterative optimization mechanism is as follows: Conduct real-time analysis on the platform user portrait optimization index, compare the platform user portrait optimization index with the set platform user portrait optimization update level threshold, and obtain O2O platform user portrait optimization information; When the platform user portrait optimization index is less than the first-level update threshold, the user portrait is marked as a high-update iterative optimization portrait; When the platform user portrait optimization index is less than the second-level update threshold and greater than the first-level update threshold, the user portrait is marked as a medium-update iterative optimization portrait; When the platform user portrait optimization index is less than the third-level update threshold and greater than the second-level update threshold, the user portrait is marked as a low-update iterative optimization portrait.

9. The method for constructing an O2O platform user portrait based on deep learning according to claim 8, characterized in that: The specific process of updating the user portrait decision tree model and optimizing the user portrait of the O2O platform is as follows: Through the O2O platform user portrait optimization information, the user portrait features that need to be improved are identified, and the optimization information and feedback data are integrated into the user portrait decision tree model using the incremental learning algorithm. The model is retrained, and the updated user portrait decision tree model is verified through cross-validation and model evaluation methods.

Images