User Interaction Feature Processing for E-commerce Prediction Accuracy
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current e-commerce prediction models for user purchase behavior fail to accurately capture the influence of specific interaction objects, such as advertisements, and ignore time series features of user interactions, leading to inaccurate predictions and inefficient inventory management.
Innovation Solution
A method and apparatus for processing user interaction information that acquire and analyze user interaction data including category, brand, and time information, generating user interaction features, and using a pre-trained operation probability generation model to predict the likelihood of a user performing a target operation, such as purchasing, by incorporating interaction time and object-specific features.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional prediction models are used to predict user purchase behavior, then the model structure is simple, but the prediction accuracy is low because they fail to capture the influence of specific interaction objects and time series features
Solution Approach 1:
The patent segments the user interaction data into multiple feature dimensions including interaction object features (advertisements, search queries), temporal features (time intervals between interactions), and user attribute features. This segmentation allows the model to capture specific influences of different interaction objects and time series patterns, thereby improving prediction accuracy while maintaining manageable model complexity through structured feature organization
Solution Approach 2:
The patent introduces temporal dimension by incorporating time interval features between user interactions, and object dimension by including specific interaction object characteristics. This multi-dimensional feature space transformation enables the model to capture complex user behavior patterns that traditional single-dimensional models miss, resolving the contradiction between model complexity and prediction accuracy
2Measurement precision
If detailed user interaction information including time series features is incorporated, then prediction accuracy improves, but data processing complexity increases
Solution Approach 1:
The patent extracts key temporal features (time intervals between interactions) and object features (advertisement IDs, search query characteristics) from the raw user interaction data. By extracting only the most relevant features rather than processing all raw data, the model achieves high prediction accuracy while keeping data processing complexity manageable through selective feature extraction
Solution Approach 2:
The patent performs preliminary feature engineering by pre-processing user interaction data to extract meaningful temporal and object features before feeding them into the prediction model. This preliminary action includes calculating time intervals between interactions and encoding interaction object characteristics, which simplifies the subsequent modeling process while maintaining high prediction accuracy
Data Source
AI summary
Embodiments of the present disclosure disclose a method and a device for processing user interaction information. A specific embodiment of the method comprises: acquiring a set of user interaction information associated with a preset interaction operation, wherein the user interaction information comprises category information and brand information of an interaction object, user attribute information, and operation time information of interaction operations corresponding to a brand of the interaction object; generating a corresponding interaction feature of a user on the basis of the set of user interaction information; and determining, on the basis of the interaction feature of the user and a pre-trained preset operation probability generation model, a probability of the user executing a target operation associated with a brand of the interaction object in the corresponding user interaction information.