Software Path Prediction Using Segmentation and Ensemble Models
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional software path sequences often fail to engage all users, as they are one-size-fits-all, leading to disengagement, and existing customization methods, such as machine learning on segmented data, may not provide sufficient accuracy in predicting personalized software paths.
Innovation Solution
A computer system and method that employs a segmentation machine learning model and ensemble machine learning to determine a predicted software path by selecting the most accurate model from a plurality of models based on time segmented data, allowing for personalized software path prediction without requiring user devices to process this information, thereby enhancing user engagement.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If a one-size-fits-all software path sequence is used, then the software structure is simple and easy to implement, but user engagement decreases as the software cannot adequately engage all users
Solution Approach 1:
The patent segments users into different groups based on their behavior, preferences, and characteristics. By dividing the user base into segments, the system can apply different path sequences to different segments, achieving customization without requiring complete individual personalization for each user. This resolves the contradiction by enabling adaptability through segmentation while maintaining manageable complexity.
Solution Approach 2:
The patent changes parameters such as path selection criteria, engagement thresholds, and user segmentation attributes dynamically. By adjusting these parameters based on user feedback and performance metrics, the system adapts path sequences to different user groups without restructuring the entire software architecture, thus maintaining simplicity while achieving customization.
2Measurement precision
If machine learning is performed on segmented data to customize game parameters, then user engagement increases, but prediction accuracy is insufficient
Solution Approach 1:
The patent performs preliminary actions by pre-segmenting users and pre-processing data into meaningful categories before applying machine learning models. This preliminary segmentation and data preparation enables more accurate predictions with reduced data processing requirements during runtime, as the heavy lifting of data organization is done in advance.
Solution Approach 2:
The patent introduces intermediaries such as feature engineering layers and data transformation steps that convert raw segmented data into meaningful inputs for machine learning models. These intermediaries enhance prediction accuracy by extracting relevant patterns while managing data complexity, thus improving precision without proportionally increasing processing requirements.
3Adaptability or versatility
If software paths are customized for individual users, then user engagement increases, but the complexity of determining personalized paths becomes excessive
Solution Approach 1:
The patent applies segmentation to divide users into manageable groups based on shared characteristics and behaviors. Instead of creating unique paths for each individual user, the system creates paths for user segments, significantly reducing complexity while maintaining personalization benefits. This resolves the contradiction by balancing adaptability with manageable complexity through group-based customization.
Data Source
AI summary
Disclosed are a computer system, a software path prediction computer, non-transitory computer-readable medium, and method for determining a predicted software path that utilize segmentation machine learning in combination with ensemble machine learning to keep a most accurate model running on a server program that receives requests from and sends predicted software path(s) to a software client.