Multi-Item Comparison Data Mining for E-Commerce Selection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing data mining methods limit item comparisons to pairs and fail to identify best sellers or popular items that users are unaware of, and often result in poor bundling suggestions by relying solely on purchase history without considering item viewing histories.
Innovation Solution
A system and method that analyze user activity data to generate comparison data for subsets of items, indicating how frequently users select one item over others, and incorporate this data into electronic catalog pages to assist users in making informed selection decisions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If data mining processes identify items based on pair-wise user activity patterns, then item relationships can be detected, but the ability to help users discriminate between multiple candidate items is insufficient
Solution Approach 1:
The patent segments the comparison function into modular components: (1) data collection module that gathers user activity data, (2) analysis module that processes the data to generate comparison information, and (3) presentation module that displays results. This segmentation allows the system to provide comprehensive multi-item comparisons while maintaining manageable system complexity through clear separation of concerns.
Solution Approach 2:
The patent transitions from traditional pair-wise item comparisons to multi-dimensional subset comparisons by analyzing user activity across multiple items simultaneously. This dimensional expansion enables users to discriminate between multiple candidate items in a single view rather than making sequential pair-wise decisions, thereby reducing information loss while controlling complexity through structured analysis.
2Productivity
If purchase-based item relationships are used to suggest bundled items, then automatic bundling suggestions can be provided, but the quality of bundling suggestions deteriorates
Solution Approach 1:
The patent merges multiple data sources including purchase history, item viewing patterns, and user activity sequences to generate bundled item suggestions. By combining these diverse indicators of user preference and behavior, the system maintains high automatic bundling efficiency while significantly improving suggestion quality, as the multi-factor analysis captures more nuanced relationships between items than purchase data alone.
Solution Approach 2:
The patent implements feedback mechanisms where user interactions with bundled suggestions are tracked and used to refine future recommendations. This feedback loop allows the system to learn from actual user behavior patterns, continuously improving bundling suggestion quality while maintaining automated operation, thereby resolving the contradiction between efficiency and reliability.
3Ease of operation
If users rely solely on item descriptions and ratings for selection, then simple presentation mechanisms can be used, but the effectiveness of item identification and selection deteriorates
Solution Approach 1:
The patent performs preliminary analysis of user activity data and generates comparison information before the user makes a selection decision. By pre-processing and organizing relevant comparison data across multiple items, the system maintains ease of operation during the actual selection process while ensuring that comprehensive preference information is already prepared and presented to the user, thereby reducing information loss without complicating the user interface.
Data Source
AI summary
Techniques for determining multiple item comparisons may be provided. For example, a system may monitor user interaction of a plurality of users that includes viewing and ordering items. The system may determine one or more items that compete, such that ordering a first item in the competing category of items lowers a probability that the user will also order a second item. The system may determine a subset of the competing items and providing information about the comparison and/or items for presentation to a user.