Selective Indexing of View Item Pages Using Traffic Prediction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing search engine optimization systems fail to identify valuable view item pages (VIPs) for maximizing natural search traffic, leading to inefficient traffic driving strategies.
Innovation Solution
A machine learned (ML) model using Gradient Boosted Machine (GBM) techniques predicts the probability of future traffic based on item-level factors and SEO performance metrics to select VIPs for indexing or no-indexing, thereby optimizing search engine results.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If all view item pages are indexed to maximize search traffic, then search traffic coverage is improved, but storage resources and processing overhead increase significantly
Solution Approach 1:
The patent segments the view item pages into different categories based on their traffic potential and characteristics. By using machine learning models to predict future traffic for different segments of pages, the system selectively indexes only those pages with high predicted traffic value, rather than indexing all pages uniformly. This segmentation approach resolves the contradiction by maintaining storage efficiency while capturing the majority of valuable search traffic.
2Quantity of substance
If selective indexing is implemented to reduce storage overhead, then resource efficiency is improved, but the ability to capture valuable search traffic may deteriorate
Solution Approach 1:
The patent implements a feedback mechanism where machine learning models continuously predict future traffic for view item pages based on historical performance and current trends. This feedback loop allows the system to dynamically adjust which pages should be indexed, ensuring that pages with high predicted traffic value are selected for indexing. The feedback-driven approach resolves the contradiction by using intelligent prediction to maintain traffic capture effectiveness while reducing unnecessary storage overhead.
3Ease of operation
If traditional SEO optimization is used without machine learning prediction, then implementation simplicity is maintained, but traffic maximization effectiveness deteriorates
Solution Approach 1:
The patent enables the SEO system to automatically identify and prioritize valuable view item pages through machine learning-based traffic prediction, eliminating the need for manual analysis and selection. The system self-services by autonomously determining which pages to index based on predicted traffic potential, resolving the contradiction by automating the complex optimization process while maintaining ease of operation through integrated implementation.
Data Source
AI summary
Example embodiments involve a system and methods for identifying valuable view item pages for search engine optimization. The system and methods provide an improvement over existing systems, which do nothing to identify or select valuable view item pages for use in driving traffic from display sites. The system and methods described herein improve the earlier system by predicting the probability of future traffic for a given product based on a number of product level factors as input variables, and identifying a selection of view item pages corresponding to the products with the probability of the highest future traffic in order to maximize the driving natural search traffic to a linked site of the corresponding view item page.