Video Dataset for Sentiment Analysis

Published: 12 March 2024| Version 1 | DOI: 10.17632/jrvj6rpnjd.1
SHANKAR MALI, Mayuri Bapat Mayuri


Videos from social media platforms like Facebook, Instagram, YouTube, Twitter, Telegram, and video libraries that meet the criteria are gathered and downloaded into the stated dataset. Six hundred videos representing a range of emotions, including fear, sadness, anger, disgust, neutral, happy, and surprised, comprise up the suggested dataset. The audio in each clip of the proposed dataset features the voice of a single speaker who appears in the movie. The videos in the dataset are of varying lengths and include people of all ages and ethnicities. US English language used in the videos. The video dataset is balanced and has labeled. For several types of research, such as sentiment analysis, opinion mining, trend analysis, and social behavior studies, among others, this dataset might be an invaluable resource.



Speech Analysis, Facial Recognition, Sentiment Analysis