Socially-Curated Brain Matrix for Search Relevance
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing data search technologies, such as Latent Semantic Analysis (LSA), face inefficiencies in processing large datasets and providing relevant results, as they analyze bulk uncurated data without considering socially-curated information which may be more valuable and relevant.
Innovation Solution
The method involves generating socially-curated 'brains' by filtering social objects with links to link objects, performing Singular Value Decomposition (SVD) on these objects to create a matrix, and using the resulting singular value representation to improve query results by focusing on socially-curated information.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If LSA analyzes bulk uncurated data to discover latent relationships, then comprehensive data coverage is achieved, but processing efficiency and result relevance deteriorate
Solution Approach 1:
The patent segments the bulk uncurated data into socially-curated subsets based on social media interactions (likes, shares, comments). By dividing the large dataset into smaller, socially-validated segments, the system processes only relevant portions while maintaining comprehensive coverage through aggregation of multiple curated subsets.
Solution Approach 2:
The patent performs preliminary filtering of data using social media metrics before applying LSA analysis. Socially-curated content is identified and selected in advance based on engagement levels, ensuring that only high-quality, relevant data enters the computational pipeline, thereby improving both efficiency and result relevance.
2Loss of information
If LSA processes large datasets using SVD decomposition, then latent relationships are discovered, but computational power and time consumption increase
Solution Approach 1:
The patent extracts and removes irrelevant or low-quality data from the bulk dataset before applying SVD decomposition. By taking out socially-uncurated or low-engagement content, the system reduces the dimensionality and size of the matrix requiring decomposition, thereby lowering computational power requirements while preserving latent relationship detection capability.
Solution Approach 2:
The patent changes the parameter of data quality by applying social media engagement metrics as filtering criteria. This parameter transformation converts raw uncurated data into socially-curated data with higher signal-to-noise ratio, enabling more efficient SVD processing with fewer iterations and reduced computational complexity.
3Measurement precision
If socially-curated filtering is applied to social objects, then result accuracy is improved, but data quantity and processing scope are reduced
Solution Approach 1:
The patent applies local quality enhancement by curating specific portions of data (social objects with links) based on their social media engagement characteristics. Rather than uniformly processing all data, the system identifies and enhances the quality of locally-relevant socially-curated content, achieving high accuracy in search results while maintaining adequate data volume through selective processing.
Data Source
AI summary
A method includes determining a plurality of social objects, each social object having a link to a link object on a network. The method further includes applying a filter to the determined social objects in order to determine a plurality of filtered social objects, retrieving a copy of each of the link objects linked to by the plurality of filtered social objects, and generating, using the retrieved copies of the link objects linked to by the plurality of filtered social objects, a matrix comprising a plurality of vectors. The method further includes generating a singular value representation of the matrix by performing Singular Value Decomposition (SVD) on the matrix and storing the singular value representation of the matrix in one or more memory units.