Two-Level Page And Sub-Strand Classification for Dynamic Websites
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for managing web pages in large-scale websites lack the ability to categorize at a granular level in real-time and adapt to dynamic changes, limiting their effectiveness in managing complex and dynamic websites.
Innovation Solution
A two-level categorization process using machine learning models to categorize web pages into strands and sub-strands based on first and second parameters, incorporating pre-processing steps and reinforcement learning for improved accuracy and efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If web pages are categorized at generic levels or individually, then the management process is simpler, but the granularity and real-time adaptability are insufficient
Solution Approach 1:
The patent divides the categorization task into two distinct levels: strand-level categorization (first level) and sub-strand-level categorization (second level). This segmentation allows the system to achieve fine-grained categorization by breaking down the complex task into manageable stages, each handled by specialized ML models trained on specific parameters, thereby resolving the contradiction between granularity and complexity.
Solution Approach 2:
The patent introduces a hierarchical dimension to categorization, transitioning from flat individual page categorization to a two-level structure (strands and sub-strands). This dimensional change enables the system to organize pages both broadly (by strand) and specifically (by sub-strand), achieving high granularity without overwhelming complexity by structuring the classification space hierarchically.
2Adaptability or versatility
If traditional retrospective categorization is used, then the system is simpler, but real-time computation and adaptation capability are limited
Solution Approach 1:
The patent pre-processes page data by extracting relevant parameters (first parameters for strands, second parameters for sub-strands) before categorization. This preliminary action prepares the data in advance, allowing the ML models to perform real-time categorization without complex computations during the actual classification moment, thus achieving real-time adaptability while managing system complexity through efficient data preparation.
Solution Approach 2:
The system uses ML models that automatically learn and adapt to new page structures and content patterns without requiring manual retraining or intervention. The models self-adjust to categorize pages in real-time based on their parameters, providing continuous adaptation to changing website structures while maintaining manageable complexity through automated learning processes.
3Productivity
If detailed granular categorization is implemented, then real-time management capability improves, but the computational requirements and system complexity increase
Solution Approach 1:
The patent segments the categorization process into two stages with different computational requirements. The first stage (strand categorization) uses a model trained on high-level parameters, which is computationally lighter. The second stage (sub-strand categorization) uses a model trained on specific parameters, also computationally efficient. This segmentation allows detailed granular categorization while managing overall computational resource consumption by avoiding a single complex monolithic model.
Solution Approach 2:
The patent applies categorization only where needed by identifying relevant parameters for each page type. Not all pages require the full two-level categorization process, and the system selectively applies the appropriate level of analysis based on page characteristics, reducing unnecessary computational overhead while maintaining high productivity for pages that do require detailed analysis.
Data Source
AI summary
This disclosure relates to a page management method and system thereof. The method includes categorizing, using a first ML model, a set of pages into at least one of a plurality of strands, based on a set of first parameters. Further, the method includes classifying each of the set of pages, using a second ML model, into one of a plurality of sub-strands, based on an associated set of second parameters and the associated strand from the plurality of strands. Further, the method includes determining, for each of the set of pages, a score, based on the associated sub-strand, the weight assigned to each of the subset of second parameters of the associated sub-strand, and values of each of the subset of second parameters of the associated sub-strand. Further, the method includes performing an action on at least one of the set of pages based on the determined score.