Big Data Management System for Dynamic Keyword Classification

The patent segments the monolithic data processing approach into a distributed architecture where data is divided into manageable chunks processed by multiple worker nodes. The master node coordinates these workers, allowing the system to handle large volumes of data by processing segments in parallel, thereby scaling productivity with data quantity.

The patent transitions from traditional single-dimension processing to multi-dimensional processing by implementing a distributed computing framework that processes data across multiple dimensions: spatial distribution across multiple nodes, temporal parallelism through concurrent processing, and hierarchical organization with master-worker relationships. This dimensional expansion enables the system to handle big data volumes that exceed traditional processing capabilities.

The patent implements preliminary action through data preprocessing and filtering at the source before data enters the main processing pipeline. Worker nodes perform initial data validation, formatting, and filtering operations, preparing data in advance for more efficient processing. This preliminary action reduces the burden on the master node and enables faster handling of high-velocity data streams.

The patent ensures continuous processing of data streams through persistent worker nodes that continuously receive and process incoming data without interruption. The system maintains continuous connections with data sources, continuously ingests data streams, and continuously processes information, ensuring no loss of real-time situational awareness even at high velocities.