Online Behavioral Predictor Using Group-Specific Models
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods struggle to accurately predict the online behavior of web users, particularly in scenarios with limited data, as users have diverse objectives and preferences, making it challenging for merchants and web designers to tailor experiences effectively.
Innovation Solution
A system that classifies users into groups based on webpage experiences and actions, generating group-specific models and combining them to provide real-time predictions on user intentions and preferences, allowing for dynamic resource allocation and customization.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a single model is used to predict all user behaviors, then the system complexity is low, but the prediction accuracy deteriorates due to diverse user objectives and preferences
Solution Approach 1:
The patent segments users into distinct groups based on their behavior patterns, objectives, and preferences. Multiple group-specific models are created to predict behaviors for each segment separately, thereby improving overall prediction accuracy while managing complexity through structured segmentation.
Solution Approach 2:
The system dynamically selects and combines multiple group-specific models based on the characteristics of the user being predicted. This dynamic approach allows the system to adapt to diverse user types without requiring a single complex model, resolving the contradiction between accuracy and complexity.
2Measurement precision
If multiple group-specific models are used to improve prediction accuracy for different user segments, then the prediction accuracy improves, but the system complexity increases
Solution Approach 1:
The patent applies partial action by selectively using only the necessary group-specific models for each prediction scenario rather than deploying all possible models. This reduces the effective complexity while maintaining the accuracy benefits where needed.
Solution Approach 2:
The system introduces an intermediary mechanism (model selection and combination logic) that manages the complexity of multiple group-specific models. This intermediary layer simplifies the interaction between diverse user segments and multiple models, making the overall system more manageable.
3Productivity
If chat invitations are extended to all users, then the resource allocation is simple, but the resource utilization efficiency deteriorates due to users with low purchasing probability
Solution Approach 1:
The patent applies local quality by tailoring resource allocation (chat invitations) to the specific characteristics and predicted needs of each user segment. High-priority users receive chat invitations while low-priority users do not, optimizing resource utilization efficiency through localized decision-making.
Solution Approach 2:
The system changes the parameter of resource allocation from a uniform approach (all users receive invitations) to a differentiated approach based on predicted purchasing probability. This parameter change improves resource utilization by directing resources only where they are most likely to be effective.
Data Source
AI summary
In some embodiments, a set of user groups can be defined, with each group relating to a different webpage experience, user action, etc. Requests are assigned to one of the groups based on actual webpage presentation features and/or user actions. A group-specific model is generated for each group and translates user information to a preliminary result (e.g., a purchasing probability). A model combination includes a weighted combination of a set of available group-specific models. User information is processed using the model combination to generate a model result. The model result is evaluated to determine whether a requested webpage is to be customized in a particular manner and/or an opportunity is to be offered.