On clusterization of "big data" streams

摘要: Current technology provides wonderful facilities for operating with extremely vast amounts of data. These are expanding due to capabilities "Cloud Computing." The developing situation gives rise the "Big Data" concept posing specific engineering and organizational challenges. Big data refers rising flood digital from many sources, including sensors, digitizers, scanners, software-based modeling, mobile phones, internet, videos, e-mails, social network communications. type could be texts, geometries, images, sounds, or their combination. Many such directly indirectly related geospatial information. In this paper, we suggest enhance available information processing resources a novel software/hardware technique on-the-fly clusterization amorphous diverse sources. presented approach is based on previously developed construction FuzzyFind Dictionary utilizing error-correction Golay Code. Realization requires intensive continuous streams, which can effectively implemented using multi-core pipelining forced interrupts. objective paper bring forward new simple efficacious tool one most demanding operations methodology --clustering items in stream mode. Improving our ability extract knowledge insights large complex collections promises solve some Nation's pressing Furthermore, reveals parallel between computational model integrating streams organization brain. uncertainties relation considered method moderated idea bounded rationality, an that does not require complete exact sensible decision-making.