Document Vector Grouping for Database Search Response Time
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Search engine systems face challenges in providing relevant search results within acceptable response times due to the exponential growth of information, leading to increased processing power and access requests on database servers, which can result in slower retrieval operations.
Innovation Solution
The method involves grouping documents using a Machine Learning Algorithm (MLA), specifically a Neural Network, to generate document vectors and query vectors based on relevance scores, allowing for the storage of documents in spatial proximity, thereby reducing the number of database servers accessed and distributing the load across multiple servers.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If the volume of information stored in the database system increases, then the comprehensiveness of search results improves, but the response time deteriorates
Solution Approach 1:
The patent divides the database system into multiple database servers (first database server, second database server, etc.) and groups documents into different collections based on their vectors. This segmentation allows the system to manage large volumes of information across multiple servers, reducing the response time by distributing the search workload rather than querying a single large database.
Solution Approach 2:
The patent transforms documents into vector representations and stores them in a vector space. By changing the parameter representation from traditional text-based indexing to vector-based spatial representation, the system can efficiently search through large volumes of information using proximity calculations, significantly improving response time while maintaining comprehensiveness.
2Reliability
If the number of database servers accessed increases, then the availability of search results improves, but the processing power required increases
Solution Approach 1:
The patent pre-calculates and stores document vectors in the database system before search operations. By performing this vectorization and grouping action in advance, the system avoids the need to process and analyze full documents during search operations, reducing the processing power required at query time while maintaining availability across multiple servers.
Solution Approach 2:
The patent extracts the essential features of documents by converting them into vectors and stores only these vector representations in the database servers. This extraction allows the system to maintain availability across multiple servers while significantly reducing the processing power needed for search operations, as vector comparisons are computationally less intensive than full document analysis.
3Measurement precision
If the access requests to database servers increase, then the relevance of search results improves, but the retrieval operation speed deteriorates
Solution Approach 1:
The patent segments documents into different collections based on their vector representations and distributes these collections across multiple database servers. This segmentation reduces the number of access requests needed by directing queries to specific relevant servers rather than broadcasting to all servers, thereby maintaining high relevance while improving retrieval speed.
Solution Approach 2:
The patent organizes documents into collections with specific thematic or contextual qualities based on their vectors. Each database server hosts specific collections with localized quality characteristics, allowing the system to quickly identify and access relevant servers based on query requirements, thus maintaining high relevance with fewer access requests and faster retrieval.
Data Source
AI summary
A method and server for storing documents in a database system is disclosed. The method includes acquiring document data associated with the documents. For each document, a Machine Learning Algorithm is employed to generate respective document vectors based on the respective document data. The method also includes storing the documents as groups of documents in the database system. Each group of documents is associated with a respective group vector. A given group of documents has documents associated with document vectors that are in a spatial proximity to the respective group vector.