Comprehensive Browser Extension for Analysing YouTube User Engagement, Controversy, User Requirements, and Trending Keywords

In the dynamic realm of YouTube, addressing the diverse needs of its extensive audience while empowering content creators is of paramount importance. This research paper introduces a robust solution encapsulated in a Chrome extension, meticulously designed to enhance user engagement and fortify YouT...

Full description

Saved in:
Bibliographic Details
Main Authors: Peiris, Praveen, Herath, Thursha, Dissanayaka, Roshani, Saranga, Kanishka, Thelijjagoda, Samantha, Weerathunge, Ishara
Format: Conference Proceeding
Language:English
Subjects:
Audience Interaction Analysis
Chrome Extension
Controversial Detection
Dictionaries
Electron tubes
Entertainment industry
Keyword Suggestion
Machine learning algorithms
Natural language processing
NLP Methodologies
User Requirement Detection
Video on demand
Web sites
You Tube Comment Analysis
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:In the dynamic realm of YouTube, addressing the diverse needs of its extensive audience while empowering content creators is of paramount importance. This research paper introduces a robust solution encapsulated in a Chrome extension, meticulously designed to enhance user engagement and fortify YouTube content creators, specifically in the sphere of information technology-related videos. Our approach leverages state-of-the-art Natural Language Processing (NLP) methodologies and Machine Learning (ML) techniques. The extension provides a comprehensive suite of four critical functionalities: analysis of user engagement, extraction of user requirements, identification of controversial topics, and provision of invaluable keyword and title recommendations. To achieve these goals, we employed a range of techniques and algorithms, including Random Forest, VADER (Valence Aware Dictionary and Sentiment Reasoner) Lexicon, Multinomial Naive Bayes (MNB), Latent Dirichlet Allocation (LDA), Bag of Words (BOW), and the GPT-3 model. The system boasts an impressive overall accuracy of 89.72%, with individual components achieving accuracies of 83.040/0, 92.83 %, 94 %, and 89%, respectively. Beyond its direct application to information technology-themed YouTube content, the methodologies presented can be adapted for other platforms and a variety of categories, such as entertainment, gaming, cooking, travel, and tourism, among others. This research not only redefines the possibilities in YouTube analytics but also offers tangible tools and insights to help creators optimize engagement and foster deeper online dialogues.
ISSN:2474-154X
DOI:10.1109/ITNAC59571.2023.10368503