Search research articles

ABOUT JoVE

Overview Leadership Blog JoVE Help Center

AUTHORS

Publishing Process Editorial Board Scope & Policies Peer Review FAQ Submit

LIBRARIANS

Testimonials Subscriptions Access Resources Library Advisory Board FAQ

RESEARCH

JoVE Journal Methods Collections JoVE Encyclopedia of Experiments Archive

EDUCATION

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual Faculty Resource Center Faculty Site

Terms & Conditions of Use

Related Concept Videos

Classification of Signals

Classification of Signals

In signal processing, signals are classified based on various characteristics: continuous-time versus discrete-time, periodic versus aperiodic, analog versus digital, and causal versus noncausal. Each category highlights distinct properties crucial for understanding and manipulating signals.
A continuous-time signal holds a value at every instant in time, representing information seamlessly. In contrast, a discrete-time signal holds values only at specific moments, often denoted as x(n), where...

You might also read

Related Articles

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

Sort by

Same author

Image-Based Dietary Assessment Using the Swedish Plate Model: Evaluation of Deep Learning-Based You Only Look Once (YOLO) Models.

JMIR formative research·2025

Same author

Simulating Market Entry Rewards for Antibiotics Development.

The Journal of law, medicine & ethics : a journal of the American Society of Law, Medicine & Ethics·2018

Same author

The role of markup for enabling interoperability in health informatics.

Frontiers in physiology·2015

Same author

Semantically linking in silico cancer models.

Cancer informatics·2014

Same author

Web-based workflow planning platform supporting the design and execution of complex multiscale cancer models.

IEEE journal of biomedical and health informatics·2014

Same author

Dealing with diversity in computational cancer modeling.

Cancer informatics·2013

Search research articles

Related Experiment Video

Updated: Sep 17, 2025

Foreign Accent and Forensic Speaker Identification in Voice Lineups: The Influence of Acoustic Features Based on Prosody

Foreign Accent and Forensic Speaker Identification in Voice Lineups: The Influence of Acoustic Features Based on Prosody

Published on: September 27, 2024

Comparing machine learning models with a focus on tone in grooming chat logs.

Leonie Hamm¹, Steve McKeever¹

¹Department of Informatics and Media, Uppsala University, Uppsala, Sweden.

Frontiers in Pediatrics

|July 4, 2025

Summary

New deep learning models, like LLaMA 3.2 1B, significantly improve the detection of online child grooming conversations and predatory authors compared to traditional machine learning. Positive tones aid detection, while negative tones present challenges.

Keywords:

grooming detection large language models online predators predator tone sentiment analysis social exchange theory

More Related Videos

Eliciting and Analyzing Male Mouse Ultrasonic Vocalization USV Songs

Eliciting and Analyzing Male Mouse Ultrasonic Vocalization USV Songs

Published on: May 9, 2017

Asthma Detection Research Based on Voice Signal Processing and Machine Learning

Asthma Detection Research Based on Voice Signal Processing and Machine Learning

Published on: July 22, 2025

Related Experiment Videos

Last Updated: Sep 17, 2025

Foreign Accent and Forensic Speaker Identification in Voice Lineups: The Influence of Acoustic Features Based on Prosody

Foreign Accent and Forensic Speaker Identification in Voice Lineups: The Influence of Acoustic Features Based on Prosody

Published on: September 27, 2024

Eliciting and Analyzing Male Mouse Ultrasonic Vocalization USV Songs

Eliciting and Analyzing Male Mouse Ultrasonic Vocalization USV Songs

Published on: May 9, 2017

Asthma Detection Research Based on Voice Signal Processing and Machine Learning

Asthma Detection Research Based on Voice Signal Processing and Machine Learning

Published on: July 22, 2025

Area of Science:

Computer Science
Artificial Intelligence
Cybersecurity

Background:

Children face significant risks from sexual predators online.
Grooming involves predators contacting minors in online chat rooms for sexual abuse.

Purpose of the Study:

Compare deep learning models against traditional machine learning for detecting grooming conversations and authors.
Analyze predator tones and their impact on detection accuracy.
Enhance automatic grooming detection to protect children.

Main Methods:

Utilized the PAN12 chat logs dataset containing grooming conversations.
Classified chat sentiments using the DistilBERT classifier.
Trained and fine-tuned Support Vector Machines (SVMs) and the LLaMA 3.2 1B large language model.

Main Results:

LLaMA 3.2 1B demonstrated superior performance in grooming detection over traditional machine learning.
Positive sentiment in grooming conversations improved detection rates.
Negative toned grooming conversations exhibited complex patterns, making them harder to detect.
LLaMA 3.2 1B achieved high F1 scores (0.99) in author detection, outperforming other models.

Conclusions:

Large language models offer advanced capabilities for identifying online grooming behaviors.
Understanding predator communication strategies, including tonal variations, is crucial for effective detection.
LLaMA 3.2 1B represents a significant advancement in protecting minors from online exploitation.