Search Result Blending Algorithm for Personalized Relevance
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current search engines fail to provide search results that are adequately personalized to individual users, often presenting items from multiple categories without considering the user's specific interests, leading to irrelevant results.
Innovation Solution
A system that uses clickstream data and affinity calculations to order search results based on user interests, incorporating a blending algorithm that weights items from categories most relevant to the user, while also promoting diversity in search results by considering global and personal navigational data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If search results are presented with minimal consideration of user interests, then the search engine can quickly retrieve and display results from multiple categories, but the relevance of search results to the particular user deteriorates
Solution Approach 1:
The system pre-calculates and stores affinity values between users and item categories based on historical clickstream data before search queries are submitted. This preliminary analysis of user behavior patterns enables the system to quickly retrieve pre-computed relevance scores during actual search operations, avoiding time-consuming real-time analysis while maintaining high personalization quality
Solution Approach 2:
The patent introduces an affinity value as an intermediary metric that mediates between user interests and search results. This affinity value, derived from historical clickstream data, serves as a pre-computed relevance score that bridges the gap between user preferences and item categories, enabling fast retrieval of personalized results without sacrificing relevance
2Reliability
If search results are personalized based on user interests, then the relevance of search results improves, but the system complexity increases due to affinity calculations and blending algorithms
Solution Approach 1:
The patent segments the search result generation process into distinct modular components: affinity value calculation module, diversity weight calculation module, and blending algorithm module. Each module handles a specific aspect of personalization independently, making the complex system more manageable, maintainable, and scalable while achieving high search result relevance
Solution Approach 2:
The affinity value serves as an intermediary data structure that simplifies the personalization process. By pre-computing and storing these affinity values between users and item categories, the system avoids complex real-time calculations during search operations, reducing system complexity while maintaining personalization quality
3Adaptability or versatility
If items from multiple categories are included in search results, then the diversity of search results improves, but the relevance to user interests deteriorates without proper weighting
Solution Approach 1:
The patent applies local quality by assigning different weights to items from different categories based on user-specific affinity values. Instead of treating all categories uniformly, the system dynamically adjusts the prominence of each category in search results according to the user's demonstrated interests, ensuring that highly relevant categories receive greater weight while maintaining diversity
Solution Approach 2:
The blending algorithm dynamically changes the weighting parameter for each category based on calculated affinity values. This parameter adjustment allows the system to optimize the balance between diversity and relevance by modulating the influence of different categories in the final search result ranking according to user preferences
Data Source
AI summary
Method and systems personalizing the search results presented to a user in an electronic marketplace. Data for a search query in an electronic marketplace is received. The query is executed and search results are received. Scores are calculated each item in the search result based at least in part on at least one of global navigational data and personal navigational data. Each item is allocated a position in an ordered search result list based at least in part on the scores. When at least some subset of the items in the search result list have been scored and allocated a position in the ordered search result list, the ordered search result list may be presented to the user.