Article: Enhancing the performance assessment of network-based and machine learning for module availability estimation Journal: International Journal of System of Systems Engineering (IJSSE) 2024 Vol.14 No.1 pp.1 - 22 Abstract: Interpreting network telemetry data is difficult. Size and volume are network assets. Production rises. ML predicts traffic trends to help decision-making. Classification and monitoring enable data science, sensor fusion, diagnostic devices, and vulnerability assessment. Complex domains have algorithms. Researchers have not found a fast, reliable way to categorise a dataset. Most literature evaluates classifiers' accuracy and falsification rate. Classification constraints include model development time, false positive rate (FPR), and precision. AI can estimate network complexity. New technology expands and complicates network messages. First, send facts. Only key nodes send messages in conventional opportunistic networks. Overusing key nodes reduces network life. We provide energy-efficient message-based routing. We assess message relevance and node energy during forwarding. It fixes energy-hungry nodes and prioritises vital signals. We replace the cache when it's full. This hinders mobility aid. This study employs machine learning to improve traditional mobility management. It presents a realistic technique using path-based forwarding architectures to identify network links. Instead of destination-based routing, delivery path information is transmitted and advanced using a mandatory access test. Inderscience Publishers - linking academia, business and industry through research

Title: Enhancing the performance assessment of network-based and machine learning for module availability estimation

Authors: Aqeel Luaibi Challoob; Abdullah Hasan Hussein

Addresses: Department of Computer Techniques Engineering, Imam Al-Kadhum College (IKC), Baghdad, 10001, Iraq ' Department of Computer Techniques Engineering, Imam Al-Kadhum College (IKC), Baghdad, 10001, Iraq

Abstract: Interpreting network telemetry data is difficult. Size and volume are network assets. Production rises. ML predicts traffic trends to help decision-making. Classification and monitoring enable data science, sensor fusion, diagnostic devices, and vulnerability assessment. Complex domains have algorithms. Researchers have not found a fast, reliable way to categorise a dataset. Most literature evaluates classifiers' accuracy and falsification rate. Classification constraints include model development time, false positive rate (FPR), and precision. AI can estimate network complexity. New technology expands and complicates network messages. First, send facts. Only key nodes send messages in conventional opportunistic networks. Overusing key nodes reduces network life. We provide energy-efficient message-based routing. We assess message relevance and node energy during forwarding. It fixes energy-hungry nodes and prioritises vital signals. We replace the cache when it's full. This hinders mobility aid. This study employs machine learning to improve traditional mobility management. It presents a realistic technique using path-based forwarding architectures to identify network links. Instead of destination-based routing, delivery path information is transmitted and advanced using a mandatory access test.

Keywords: energy efficient; sensor network; machine learning; network security; data aggregation; module availability estimation; SwitchML; PyTorch; TensorFlow; network infrastructure.

DOI: 10.1504/IJSSE.2024.135910

International Journal of System of Systems Engineering, 2024 Vol.14 No.1, pp.1 - 22

Received: 13 Oct 2022
Accepted: 30 Nov 2022
Published online: 09 Jan 2024 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article

Title: Enhancing the performance assessment of network-based and machine learning for module availability estimation

Keep up-to-date