Granular Cluster Generation for Real-Time Content Selection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The challenge in information technology infrastructure is efficiently and effectively selecting content items in real-time without introducing delay and latency, due to the large number of selection criteria processed from data packets received over a network.
Innovation Solution
A system that generates granular clusters using a metric-based model via machine learning to process tokens, reducing computing and memory utilization by applying filtering and processing techniques, allowing for real-time content selection.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a large set of tokens is processed for content item selection, then selection accuracy is improved, but processing latency and computing resource utilization increase
Solution Approach 1:
The patent segments the large set of tokens into multiple granular clusters, where each cluster contains a subset of tokens grouped by shared characteristics. This segmentation allows the system to process smaller, more manageable clusters in parallel, reducing overall processing latency while maintaining comprehensive coverage of all tokens for accurate content item selection.
Solution Approach 2:
The patent performs preliminary clustering and filtering of tokens before the actual content item selection process. By pre-organizing tokens into granular clusters and removing duplicates in advance, the system reduces the computational burden during real-time selection, thereby reducing latency without compromising selection accuracy.
2Measurement precision
If a large set of tokens is processed for content item selection, then selection accuracy is improved, but computing resource utilization increases
Solution Approach 1:
The patent segments the large token set into multiple granular clusters, enabling parallel processing of smaller subsets. This reduces the computational resources required at any given moment while maintaining comprehensive token coverage for accurate selection, as different clusters can be processed simultaneously across multiple computing units.
Solution Approach 2:
The patent merges multiple token clusters into a unified selection framework, where results from processing multiple smaller clusters are combined to achieve the same selection accuracy as processing the entire token set at once. This merging approach reduces peak computing resource utilization by distributing the workload across multiple smaller processing tasks.
3Quantity of substance
If granular clusters are generated from tokens, then memory utilization is reduced, but processing complexity increases
Solution Approach 1:
The patent segments the token set into granular clusters with clear defining characteristics, which reduces memory utilization by storing cluster representations rather than individual tokens. The segmentation process uses systematic criteria to group tokens, making the clustering operation manageable despite the increased processing steps required to create and manage multiple clusters.
Data Source
AI summary
Generating granular clusters for real-time processing is provided. The systems can identify tokens based on aggregating input from computing devices over a time interval. The systems can identify, based on metrics, a subset of tokens for cluster generation. The systems can generate, via a clustering technique, token clusters from the subset of the tokens, each of the token clusters comprising two or more tokens from the subset of the tokens. The systems can apply a de-duplication technique to each of the token clusters. The systems can apply a filtering technique to the token clusters to remove tokens erroneously grouped in a token cluster. The systems can assign, based on a selection process, a label for each of the token clusters. The systems can activate, based on a number of remaining tokens in each of the token clusters, a subset of the token clusters for real-time content selection.