Quality enhanced hybrid youtube video recommendation based on user preference through sentiment analysis on comments – a study on natural remedy videos

With the rapid development of technologies, the Internet has become the primary source of information. In recent years, blogs and videos on the Internet have become the primary source of health-related information. About four out of five people trust the online information posted by anonymous people...

Full description

Saved in:

Bibliographic Details
Published in:	Multimedia tools and applications Vol. 83; no. 15; pp. 44217 - 44250
Main Authors:	A, Saravanan, S, Sathya Bama, P, Ramila Rajaleximi, D, Anandhi, M, Srividya
Format:	Journal Article
Language:	English
Published:	New York Springer US 01-05-2024 Springer Nature B.V
Subjects:	Computer Communication Networks Computer Science Content analysis Data mining Data Structures and Information Theory Internet Multimedia Information Systems Privacy Recommender systems Reliability analysis Sentiment analysis Special Purpose and Application-Based Systems Structural analysis Video quality analysis content relevancy content reliability user comments YouTube videos sentiment analysis
Online Access:	Get full text
Tags:	Add Tag No Tags, Be the first to tag this record!

Description
Summary:	With the rapid development of technologies, the Internet has become the primary source of information. In recent years, blogs and videos on the Internet have become the primary source of health-related information. About four out of five people trust the online information posted by anonymous people and access it for their own benefit. More specifically, the simple home remedies or natural remedies for minor health issues available on YouTube gain more attention among all age groups. Numerous viewers not only watch them; they even try or use these remedies personally for magical cures. However, the quality of the information about simple treatments and home remedies available on YouTube may not be accurate, reliable, useful, or trustworthy. Moreover, it is not simple to distinguish reliable information from misleading, false information. Thus, the main objective of this research work is to provide a quality enhanced recommendation of YouTube videos based on the user's preferences. In order to achieve the objective, the proposed work computes the scores for content relevancy and content reliability. The analysis of content reliability is two-fold, with implicit evaluation using structural analysis and explicit evaluation via sentimental analysis on YouTube video comments. The experimental and result analysis performed for the proposed work shows that the predicted ratings and the order of recommendations are more useful and have improved performance on reliability of the video content.
ISSN:	1573-7721 1380-7501 1573-7721
DOI:	10.1007/s11042-023-17391-6