Search Result Redistributor Using Natural Distribution Index

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Conventional search systems face challenges in presenting the best search results among a vast number of potential results, leading to difficulties in identifying relevant information due to scoring inaccuracies and presentation bias.

Innovation Solution

The technology redistributes ranked search results based on a natural distribution index, which maps search query terms to search result attributes and represents interactions with these attributes, thereby minimizing the gap between the ranked results and the natural distribution.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If conventional search systems rank search results based on scoring algorithms, then search results are provided in a standardized format, but scoring inaccuracies and presentation bias make it difficult for users to identify relevant information

Engineering Contradiction:
Improvescoring accuracyVSAvoidrelevance identification
Core Design Contradiction:
Measurement precisionVSLoss of information

Solution Approach 1:

The patent introduces a blender as an intermediary component that sits between the conventional ranking system and the user. The blender receives ranked search results and redistributes them according to a natural distribution derived from historical user interactions. This intermediary layer corrects the biases and inaccuracies of the original scoring algorithm without requiring changes to the core ranking system, thereby improving relevance identification while maintaining system stability.

Inventive Principle:
Principle #24Intermediary (Mediator)

Solution Approach 2:

The system utilizes feedback from historical user interactions with search results to construct a natural distribution. By analyzing patterns in how users actually interact with search results (clicks, views, selections), the system creates a feedback loop that informs the redistribution process. This feedback mechanism allows the system to continuously improve its ability to identify relevant information by learning from actual user behavior rather than relying solely on theoretical scoring algorithms.

Inventive Principle:
Principle #23Feedback

2Quantity of substance

If search systems present a vast number of potential search results, then comprehensive coverage is achieved, but users face difficulties in identifying relevant information among the large volume of results

Engineering Contradiction:
Improvesearch results coverageVSAvoidrelevance identification
Core Design Contradiction:
Quantity of substanceVSEase of operation

Solution Approach 1:

The patent applies local quality by redistributing search results based on their specific attributes and historical interaction patterns. Instead of treating all search results uniformly, the system analyzes individual result characteristics and their performance in historical contexts, then positions them in the results list according to their local quality metrics. This allows highly relevant results to be prominently displayed while maintaining comprehensive coverage of the search topic.

Inventive Principle:
Principle #3Local quality

3Productivity

If conventional ranking algorithms are used to order search results, then processing efficiency is maintained, but presentation bias reduces the accuracy of result presentation

Engineering Contradiction:
Improveprocessing efficiencyVSAvoidresult presentation accuracy
Core Design Contradiction:
ProductivityVSReliability

Solution Approach 1:

The system performs preliminary action by pre-computing the natural distribution from historical interaction data and storing it for efficient retrieval. This pre-processing step allows the blender to quickly redistribute search results during query processing without performing complex calculations in real-time. The preliminary computation of interaction patterns enables the system to maintain high processing efficiency while applying bias-correction redistributions.

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentUS20250068637A1Search system ranking using a natural distribution
Publication Date: 2025.02.27 EBAY INC
  • US20250068637A1 patent drawing
  • US20250068637A1 patent drawing
  • US20250068637A1 patent drawing

AI summary

A search engine is provided that generates search results proportional to a natural distribution. Search results identified and ranked by the search engine for a search query are redistributed so that a fixed number of top ranked search results include search result attributes proportional to the natural distribution, as determined from interaction tracking of prior search results. A natural distribution index mapping prior search queries to search result attributes and a proportional representation of interactions with the search result attributes is generated. The natural distribution is determined from the proportional representation of interactions. The search engine redistributes the fixed number of top ranked search results by minimizing a gap between the ranked set of search results for the search query and the natural distribution.