Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Beyond REM: A New Approach to the Use of Image Classifiers for the Management of 6G Networks.

Sensors (Basel, Switzerland)·2023
Same author

A firebreak placement model for optimizing biodiversity protection at landscape scale.

Journal of environmental management·2023
Same author

Victim Detection and Localization in Emergencies.

Sensors (Basel, Switzerland)·2022
Same author

Microencapsulation of a Commercial Food-Grade Protease by Spray Drying in Cross-Linked Chitosan Particles.

Foods (Basel, Switzerland)·2022
Same author

WiFi FTM and UWB Characterization for Localization in Construction Sites.

Sensors (Basel, Switzerland)·2022
Same author

Kinetic Modeling, Thermodynamic Approach and Molecular Dynamics Simulation of Thermal Inactivation of Lipases from <i>Burkholderia cepacia</i> and <i>Rhizomucor miehei</i>.

International journal of molecular sciences·2022
Same journal

Enhancing Unsupervised Multi-Source Domain Adaptation for Person Re-Identification via Mixture of Experts and Graph-Based Relation.

Sensors (Basel, Switzerland)·2026
Same journal

Development of an Instrumented Glove for Palmar Pressure Assessment in Kayakers.

Sensors (Basel, Switzerland)·2026
Same journal

Development and Experimental Validation of an Autonomous IoT-Based Monitoring System for Real-Time Water Quality Assessment in the Amazon River.

Sensors (Basel, Switzerland)·2026
Same journal

Semi-Supervised Adversarial Learning Framework for Controller Area Network Bus Intrusion Detection.

Sensors (Basel, Switzerland)·2026
Same journal

Smart Optimization Method for Safety Signs in Innovative Manufacturing Environments Integrating Industrial Field IoT Sensors and Knowledge Graphs.

Sensors (Basel, Switzerland)·2026
Same journal

Three-Dimensional Modeling and Performance Analysis of Dynamic mmWave V2I Networks Based on Stochastic Geometry.

Sensors (Basel, Switzerland)·2026
See all related articles
  1. Home
  3. Active Learning Methodology For Expert-assisted Anomaly Detection In Mobile Communications.
  1. Home
  3. Active Learning Methodology For Expert-assisted Anomaly Detection In Mobile Communications.

Related Experiment Video

Evaluation of a Smartphone-based Human Activity Recognition System in a Daily Living Environment
06:49

Evaluation of a Smartphone-based Human Activity Recognition System in a Daily Living Environment

Published on: December 11, 2015

9.0K

Active Learning Methodology for Expert-Assisted Anomaly Detection in Mobile Communications.

José Antonio Trujillo1, Isabel de-la-Bandera1, Jesús Burgueño2

  • 1Instituto de Telecomunicación (TELMA), Universidad de Málaga, CEI Andalucía TECH E.T.S. Ingeniería de Telecomunicación, Bulevar Louis Pasteur 35, 29010 Málaga, Spain.

Sensors (Basel, Switzerland)
|January 8, 2023

View abstract on PubMed

Summary
This summary is machine-generated.

This article presents a hybrid framework that combines automated algorithms with human expert input to improve the identification of unusual patterns in mobile network traffic. By using an active learning approach, the system learns to prioritize the most informative data points for human review, resulting in more accurate detection of network anomalies compared to fully automated methods.

Keywords:
5Gactive learninganomaly detectionmachine learningself-organizing networksnetwork traffic analysismachine learning integrationhuman-in-the-loop systemstelecommunications security

Frequently Asked Questions

More Related Videos

Effective Analysis of Human Exposure Conditions with Body-worn Dosimeters in the 2.4 GHz Band
06:43

Effective Analysis of Human Exposure Conditions with Body-worn Dosimeters in the 2.4 GHz Band

Published on: May 2, 2018

7.1K
Microfluidic Platform with Multiplexed Electronic Detection for Spatial Tracking of Particles
11:54

Microfluidic Platform with Multiplexed Electronic Detection for Spatial Tracking of Particles

Published on: March 13, 2017

9.4K

Related Experiment Videos

Evaluation of a Smartphone-based Human Activity Recognition System in a Daily Living Environment
06:49

Evaluation of a Smartphone-based Human Activity Recognition System in a Daily Living Environment

Published on: December 11, 2015

9.0K
Effective Analysis of Human Exposure Conditions with Body-worn Dosimeters in the 2.4 GHz Band
06:43

Effective Analysis of Human Exposure Conditions with Body-worn Dosimeters in the 2.4 GHz Band

Published on: May 2, 2018

7.1K
Microfluidic Platform with Multiplexed Electronic Detection for Spatial Tracking of Particles
11:54

Microfluidic Platform with Multiplexed Electronic Detection for Spatial Tracking of Particles

Published on: March 13, 2017

9.4K

Area of Science:

  • Active Learning methodologies in telecommunications infrastructure
  • Network operations research within mobile communications

Background:

Modern mobile networks face escalating operational hurdles due to their inherent structural diversity and service variety. Identifying irregular traffic patterns remains a significant obstacle for maintaining service quality in these complex environments. Prior research has shown that automated systems often struggle to interpret the intricate relationships between various network parameters. That uncertainty drove the need for incorporating human intuition into the detection process. Experts frequently possess the only reliable understanding of the underlying causes behind specific network irregularities. No prior work had resolved how to effectively bridge the gap between machine processing power and human domain expertise. This study addresses the challenge of integrating these two distinct sources of information. The proposed framework aims to enhance detection accuracy by leveraging the strengths of both human and machine intelligence.

Purpose Of The Study:

The primary aim of this study is to develop an integrated framework for anomaly detection in mobile communications using active learning. This research seeks to address the limitations of fully automated algorithms when handling complex and heterogeneous network data. The authors intend to demonstrate that combining machine processing with human expert knowledge results in superior detection performance. The project explores how to effectively incorporate expert feedback into the operational workflow of modern networks. This motivation stems from the increasing difficulty of identifying abnormalities in rapidly evolving communication systems. The researchers investigate whether a hybrid model can bridge the gap between algorithmic speed and human analytical depth. They propose a methodology that selectively utilizes expert input to refine the detection process. This work aims to provide a practical solution for managing the growing variety of services in next-generation mobile infrastructures.

The researchers propose a hybrid framework where an active learning algorithm selects specific data points for human expert verification. This interaction allows the system to refine its detection capabilities by incorporating expert feedback, which outperforms methods relying exclusively on automated algorithmic outputs.

The study utilizes an active learning module to manage the interaction between the automated detection algorithm and the human expert. This component acts as a filter, identifying the most ambiguous or critical network events that require human intervention to ensure accurate classification.

The authors state that human expertise is necessary because automated algorithms often fail to interpret the complex, heterogeneous relationships between diverse network parameters. Human intuition provides the context required to correctly identify abnormalities that machines might otherwise misclassify or overlook entirely.

The active learning component serves as the decision-making engine that determines which data samples are presented to the human expert. By prioritizing uncertain or high-impact cases, this data type ensures that the expert's time is used efficiently to maximize the model's learning rate.

The researchers measured performance by comparing the hybrid model against a baseline method that relied solely on algorithm output. The hybrid approach consistently demonstrated increased detection accuracy, confirming that incorporating expert feedback significantly improves the system's ability to signal network abnormalities correctly.

The authors propose that their methodology offers a scalable solution for managing the increasing complexity of next-generation mobile networks. They claim that this hybrid approach provides a practical way to maintain high service standards despite the growing variety of services and network parameters.

Main Methods:

The research team designed a hybrid framework to evaluate the efficacy of expert-assisted anomaly detection. They implemented an online detection algorithm as the primary automated component within the system. The review approach involved comparing this new hybrid model against a standard method that functions without human input. Investigators utilized active learning to manage the flow of information between the machine and the human expert. This design ensures that the system only requests expert feedback for the most critical or ambiguous network events. The team conducted a series of controlled tests to measure the performance improvements provided by this integration. They focused on quantifying the accuracy of abnormality signaling in simulated network environments. This experimental setup allowed for a direct assessment of how expert knowledge influences the overall detection capability.

Main Results:

Key findings from the literature indicate that the hybrid model consistently achieves higher performance than fully automated solutions. The data show that incorporating expert knowledge leads to a more precise identification of network irregularities. Results confirm that the proposed methodology successfully automates a portion of the detection process while maintaining high accuracy. The study reveals that the hybrid approach effectively addresses the limitations of algorithms that lack human context. Quantitative comparisons demonstrate that the integration of expert feedback reduces the rate of misclassified abnormalities. The findings suggest that the active learning mechanism is instrumental in optimizing the expert's contribution to the system. These results highlight the clear advantage of combining machine processing capacity with human domain knowledge in mobile network operations. The evidence supports the conclusion that this hybrid framework is superior to traditional, non-expert-assisted detection methods.

Conclusions:

The authors demonstrate that integrating human expertise into automated detection frameworks yields superior performance. This synthesis suggests that hybrid models outperform systems relying solely on algorithmic outputs. The findings imply that active learning effectively optimizes the interaction between human reviewers and machine learning models. Researchers propose that this methodology reduces the burden on experts while maintaining high detection standards. The evidence indicates that automating parts of the process while retaining human oversight is a viable strategy for future network operations. This review of the literature confirms that human-in-the-loop systems provide a robust solution for complex communication environments. The study highlights the necessity of balancing computational efficiency with expert knowledge for improved anomaly identification. These results provide a clear pathway for developing more reliable and adaptive network monitoring tools.