Improving Real-Time Data Processing Performance Through Machine Learning

Actual-time information processing technology have become more and more vital for the efficient operation of diverse enterprise tactics. System gaining knowledge of generation can help improve the overall performance of those information processing structures via allowing them to conform to changing...

Full description

Saved in:
Bibliographic Details
Published in2023 IEEE International Conference on Paradigm Shift in Information Technologies with Innovative Applications in Global Scenario (ICPSITIAGS) pp. 282 - 288
Main Authors Sathya, S, Shewale, Chaitali R., Kar, Abhijeet, Pandey, Megha, Gosh, Srijani, Sohal, Jagmeet
Format Conference Proceeding
LanguageEnglish
Published IEEE 28.12.2023
Subjects
algorithms
information
Information processing
Machine learning
Machine learning algorithms
parameters
Performance evaluation
Prediction algorithms
predictions
progressed
Real-time systems
Scalability
Online AccessGet full text

Cover

More Information
Summary:Actual-time information processing technology have become more and more vital for the efficient operation of diverse enterprise tactics. System gaining knowledge of generation can help improve the overall performance of those information processing structures via allowing them to conform to changing conditions and styles. System gaining knowledge of algorithms may be used to discover tendencies and anomalies in information streams, detect patterns, and make predictions so that it will optimize the overall performance of the records processing gadget. The machine learning algorithms can be skilled and evaluated on ancient facts, or on streaming facts samples after which deployed on a records processing device. The overall performance of the statistics processing system may be progressed via adjusting parameters associated with information sampling frequency, facts granularity and the kind of features to be processed. Further, gadget mastering algorithms can be used to optimize facts preprocessing and publish-processing duties, together with feature extraction and information cleaning. Ultimately, gadget learning techniques can be utilized to optimize the scheduling and execution of records processing jobs, which includes identifying appropriate sources for a selected project and dynamically allocating sources and jobs as in line with the demand.
DOI:10.1109/ICPSITIAGS59213.2023.10527549