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 Experiment Video

Updated: Oct 8, 2025

Author Spotlight: Addressing Technical and Subjective Challenges in Measuring Classroom Attention
06:37

Author Spotlight: Addressing Technical and Subjective Challenges in Measuring Classroom Attention

Published on: December 15, 2023

4.4K

Cyberbullying detection: advanced preprocessing techniques & deep learning architecture for Roman Urdu data.

Amirita Dewani1, Mohsin Ali Memon1, Sania Bhatti1

  • 1Institute of Information and Communication Technologies, Department of Software Engineering, Mehran University of Engineering & Technology, Jamshoro, Sindh Pakistan.

Journal of Big Data
|December 27, 2021
PubMed
Summary

Related Concept Videos

Bullying02:04

Bullying

8.5K
A modern form of aggression is bullying. As you learn in your study of child development, socializing and playing with other children is beneficial for children’s psychological development. However, as you may have experienced as a child, not all play behavior has positive outcomes. Some children are aggressive and want to play roughly. Other children are selfish and do not want to share toys. One form of negative social interactions among children that has become a national concern is...
8.5K

You might also read

Related Articles

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

Sort by
Same journal

CardiaTics: An explainable AI integrated heart disease diagnosis model with feature engineering and stacked ensemble approach.

Journal of big data·2026
Same journal

Comprehensive representation of health-related phenotypes in one million dogs using topic modelling of electronic health records.

Journal of big data·2026
Same journal

UniqueNOSD: a novel framework for NoSQL over SQL databases.

Journal of big data·2025
Same journal

<i>F</i>u<i>n</i>Da: scalable serverless data analytics and in situ query processing.

Journal of big data·2025
Same journal

Integrating Big Data, Artificial Intelligence, and motion analysis for emerging precision medicine applications in Parkinson's Disease.

Journal of big data·2024
Same journal

Interpolation-split: a data-centric deep learning approach with big interpolated data to boost airway segmentation performance.

Journal of big data·2024

Researchers developed methods to detect cyberaggression and hate speech in Roman Urdu social media text. Recurrent Neural Network-Long Short-Term Memory (RNN-LSTM) and Bidirectional Long Short-Term Memory (RNN-BiLSTM) models showed the best performance.

Area of Science:

  • Natural Language Processing
  • Computational Linguistics
  • Social Media Analysis

Background:

  • Social media facilitates communication but also enables hate speech and cyberbullying.
  • Existing research on cyberaggression detection primarily focuses on well-resourced languages.
  • Roman Urdu, a Roman-script version of Urdu, is increasingly used on social media, particularly in South Asia, creating a research gap.

Purpose of the Study:

  • To address the gap in cyberaggression and hate speech detection for resource-poor languages.
  • To preprocess and analyze Roman Urdu microtext for detecting cyberbullying patterns.
  • To evaluate the performance of different machine learning models for this task.

Main Methods:

  • Extensive preprocessing of Roman Urdu microtext, including slang dictionary creation and stop word removal.
Keywords:
Advanced preprocessingBig dataCyberbullyingDeep learningHate speech detection

Related Experiment Videos

Last Updated: Oct 8, 2025

Author Spotlight: Addressing Technical and Subjective Challenges in Measuring Classroom Attention
06:37

Author Spotlight: Addressing Technical and Subjective Challenges in Measuring Classroom Attention

Published on: December 15, 2023

4.4K
  • Handling of encoded text formats and non-linguistic features.
  • Implementation and experimentation with Recurrent Neural Network-Long Short-Term Memory (RNN-LSTM), Recurrent Neural Network-Bidirectional Long Short-Term Memory (RNN-BiLSTM), and Convolutional Neural Network (CNN) models, with hyperparameter tuning.

    • Main Results:

    • RNN-LSTM and RNN-BiLSTM models demonstrated superior performance in detecting cyberaggression.
    • RNN-LSTM achieved a validation accuracy of 85.5% and an F1 score of 0.7 for the aggression class.
    • RNN-BiLSTM achieved a validation accuracy of 85% and an F1 score of 0.67 for the aggression class.

    Conclusions:

    • The study successfully identified effective models for cyberaggression detection in Roman Urdu.
    • RNN-LSTM and RNN-BiLSTM are promising for analyzing textual patterns of cyberbullying in this under-resourced language.
    • The findings contribute to developing solutions for combating online harassment in diverse linguistic contexts.