Recommendation Device With Browsing-Based Specification Matching
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional advertisement recommendation systems fail to accurately provide suitable product recommendations, often displaying unnecessary ads due to broad similarity measures or focusing on narrow product details, missing the user's true needs or latent demands, especially for infrequently purchased items or products with specific intentions.
Innovation Solution
A recommendation device that extracts candidate products based on user browsing and unpurchased product information, and utilizes product specification correlation to match product specifications, improving the accuracy of recommended product selection.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If broad similarity measures are used to select advertisements, then coverage is improved, but recommendation accuracy deteriorates
Solution Approach 1:
The patent segments the product recommendation process into multiple stages: first extracting candidate products based on broad similarity to purchased items, then refining the selection by extracting and comparing specific feature terms (e.g., color, material, function) to identify the most relevant features for accurate recommendation.
Solution Approach 2:
The patent applies local quality by focusing on specific feature terms within product specifications rather than treating all product attributes uniformly. It identifies and weights important features (such as color, material, or functional characteristics) that are most relevant to the user's preferences, allowing for accurate recommendations while maintaining broad coverage.
2Measurement precision
If narrow similarity measures are used to select advertisements, then recommendation accuracy is improved, but coverage deteriorates
Solution Approach 1:
The patent segments the recommendation process into two phases: an initial broad extraction phase that captures all potentially relevant products based on general similarity, followed by a refined selection phase that applies narrow feature-term comparison to achieve high accuracy. This segmentation allows the system to maintain both coverage and precision.
Solution Approach 2:
The patent employs partial action by extracting and comparing only the most relevant feature terms (such as color, material, or specific functions) rather than analyzing every possible product attribute. This selective approach enables accurate recommendations while avoiding the computational overhead and coverage limitations of exhaustive analysis.
3Device complexity
If only purchase history is used for recommendation, then simplicity is maintained, but recommendation accuracy deteriorates
Solution Approach 1:
The patent performs preliminary action by pre-processing product information to extract and organize feature terms (such as color, material, function) before the recommendation process. This preparation allows the system to accurately compare and match products based on their features without adding significant complexity to the overall system architecture.
Solution Approach 2:
The patent introduces feature term extraction as an intermediary step between raw product data and recommendation output. By transforming product specifications into standardized feature terms, the system creates a bridge that enables accurate matching based on user preferences while maintaining system simplicity through automated processing.
Data Source
AI summary
A product selecting unit extracts a candidate for recommended product on the basis of browsing information in a browsing information storing unit and unpurchased product information in an unpurchased product information storing unit. Product specifications that are information specific to each product are stored in a product information storing unit. A specification correlation calculating unit extracts a product having a product specification correlated with a product specification of the candidate for recommended product on the basis of the product specifications, and extracts the product as a recommended product.