Compliance source authentication technique for person adaptation networks utilizing deep learning-based patterns segmentation

Jamil Alsayaydeh, Jamil Abedalrahim and Oliinyk, Andrii O. and Yusof, Mohd Faizal and Shkarupylo, Vadym Viktorovych and Artemchuk, Volodymyr O. and Herawan, Safarudin Gazali (2024) Compliance source authentication technique for person adaptation networks utilizing deep learning-based patterns segmentation. IEEE Access, 12. pp. 99045-99057. ISSN 2169-3536

Text Compliance_Source_Authentication_Technique_for_Person_Adaptation_Networks_Utilizing_Deep_Learning-Based_Patterns_Segmentation.pdf Download (7MB)

Abstract

Due to delivering flexible access to apps and assisted support, individual systems with adaptive capabilities adjust their behaviour in response to input. The adaptation process relies on personal inputs, including contact interactions, notes of speech, and postures. Errors occur within these systems primarily as an outcome of insufficient mining information along with unfamiliar types of input during adjustment. This manuscript reduces recognition errors by introducing a Compliant Input Recognition with Pattern Classification (CIR-PC) system. The recommended strategy uses deep learning and statistical mining to avoid unstructured source handling and information deficiencies. Input sequence analysis and information deficit correction are the two stages of deep learning. Specific requirements for data, including extraction associated with each classified input type, are established in the first stage. The subsequent stage identifies the input pattern containing mining data to provide flexible user interaction. The machine learning model undergoes training with the input pattern and data parameters for categorization. The result improves categorizing, resource usage, and reactions from the system. On the contrary, it reduces misdetection and reactive latencies.

Item Type:	Article
Uncontrolled Keywords:	Data mining, Deep learning, Human adaptive systems, Input processing, Pattern classification
Divisions:	Faculty Of Electronics And Computer Technology And Engineering
Depositing User:	Sabariah Ismail
Date Deposited:	07 Oct 2024 11:33
Last Modified:	07 Oct 2024 11:33
URI:	http://eprints.utem.edu.my/id/eprint/27738
Statistic Details:	View Download Statistic

Actions (login required)

View Item