An approach to detect offence in Memes using Natural Language Processing(NLP) and Deep learning

Giri, Roushan Kumar; Gupta, Subhash Chandra; Gupta, Umesh

doi:10.1109/iccci50826.2021.9402406

Cited by 3 publications

(1 citation statement)

References 3 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Various social networks target different purposes: dissemination of ideas (Twitter or Facebook), professional contacts (LinkedIn), sharing of images (Instagram), dissemination of videos (YouTube), meetings (Tinder), etc. They all have one thing in common: they try to connect people [4,5,6].…”

Section: Introductionmentioning

confidence: 99%

Hate Speech Detection Using Decision Tree Algorithm

Lavanya,

Ramesh,

Kumar

et al. 2023

JAMCS

View full text Add to dashboard Cite

The advancement of the internet and social media, people has access to various platforms to freely share their thoughts and opinions on various topics. However, this freedom of expression is abused to incite hatred against individuals or groups of people based on race, religion, gender, etc. question. Therefore, to address this emerging problem on social media sites, recent studies have used various feature engineering techniques and machine learning algorithms to automatically detect hate speech posts on different datasets. Advances in machine learning have intrigued researchers seeking and implementing solutions to the problem of hate speech. Currently, we are using decision tree algorithm technique to detect hate speech using text data.

show abstract

Section: Introductionmentioning

confidence: 99%

Hate Speech Detection Using Decision Tree Algorithm

Lavanya,

Ramesh,

Kumar

et al. 2023

JAMCS

View full text Add to dashboard Cite

show abstract

An Abnormal Data Detection Method Based on Text Classification

Chen

Wang

2022

Digital TV and Wireless Multimedia Communications

View full text Add to dashboard Cite

A Deep Learning-Based Framework for Offensive Text Detection in Unstructured Data for Heterogeneous Social Media

Bacha,

Ullah,

Khan

et al. 2023

IEEE Access

View full text Add to dashboard Cite

Social media such as Facebook, Instagram, and Twitter are powerful and essential platforms where people express and share their ideas, knowledge, talents, and abilities with others. Users on social media also share harmful content, such as targeting gender, religion, race, and trolling. These posts may be in the form of tweets, videos, images, and memes. A meme is one of the mediums on social media which has an image and embedded text in it. These memes convey various views, including fun or offensiveness, that may be a personal attack, hate speech, or racial abuse. Such posts need to be filtered out immediately from social media. This paper presents a framework that detects offensive text in memes and prevents such nuisance from being posted on social media, using the collected KAU-Memes dataset 2582. The latter combines the "2016 U.S. Election" dataset with the newly generated memes from a series of offensive and non-offensive tweets datasets. In fact, this model uses the KAU-Memes dataset containing symbolic images and the corresponding text to validate the proposed model. We compare the performance of three proposed deep-learning algorithms to train and detect offensive text in memes. To the best of the authors knowledge and literature review, this is the first approach based on You Only Look Once (YOLO) for offensive text detection in memes. This framework uses YOLOv4, YOLOv5, and SSD MobileNetV2 to compare the model's performance on the newly labeled KAU-Memes dataset. The results show that the proposed model achieved 81.74%, 84.1%, mAP, and F1-score, respectively, for SSD-MobileNet V2, and 85.20%, 84.0%, mAP, and F1-score, respectively for YOLOv4. YOLOv5 had the best performance and achieved the highest possible mAP, F1-score, precision, and recall which are 88.50%, 88.8%, 90.2%, and 87.5%, respectively, for YOLOv5.

show abstract

An approach to detect offence in Memes using Natural Language Processing(NLP) and Deep learning

Cited by 3 publications

References 3 publications

Hate Speech Detection Using Decision Tree Algorithm

Hate Speech Detection Using Decision Tree Algorithm

An Abnormal Data Detection Method Based on Text Classification

A Deep Learning-Based Framework for Offensive Text Detection in Unstructured Data for Heterogeneous Social Media

Contact Info

Product

Resources

About