Predicting the Popularity of Trending Videos in Youtube Using Sentimental Analysis
G. Mohana Prabha1, B. Madhumitha2, R. P. Ramya3
1Dr. G. Mohana Prabha, Associate Professor, Department of Information Technology, M. Kumarasamy College of Engineering, Karur, (TamilNadu), India.
2B. Madhumitha, UG Student, Department of Information Technology, M. Kumarasamy College of Engineering, Karur, (TamilNadu), India.
3R. P. Ramya, UG Student, Department of Information Technology, M. Kumarasamy College of Engineering, Karur, (TamilNadu), India.
Manuscript received on 05 April 2019 | Revised Manuscript received on 14 April 2019 | Manuscript Published on 24 May 2019 | PP: 215-220 | Volume-8 Issue-6S3 April 2019 | Retrieval Number: F10430486S319/19©BEIESP
Open Access | Editorial and Publishing Policies | Cite | Mendeley | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open-access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: The recognition of social media is growing hastily because it is simple in use and easy to create and percentage snap shots, video even from the ones customers who’re technically ignorant of social media. In social media particularly in YOUTUBE, detection of sentiment polarity is a completely hard task due to a few obstacles in current sentiment dictionaries. In present dictionaries there are not any proper sentiments of terms created by way of network. For the past several years YouTube has been via some distance the largest user-pushed online video provider. While a lot of these movies comprise a tremendous wide variety of user comments, little paintings has been executed so far in extracting tendencies from those feedback due to their low information consistency and nice. In this venture, we can perform sentiment analysis of the YouTube feedback related to popular topics the use of Natural Language Processing using machine getting to know techniques. The Natural language processing is the discipline that research the way to make the machines study and interpret the language that the humans use, the herbal language. But within the machines global, the phrases no longer exist and they’re represented by using sequences of numbers that the device represents with a character whilst displaying them on display. The Sentiment Analysis is the call of the hassle that with a sentence or textual content the device receives succesful to analyze and are expecting with the maximum precision viable the sentiment with a purpose to be received with the aid of a person while it or the contextual opinion reads associated with something. We display that an evaluation of the emotions to perceive their developments, seasonality and forecasts can offer a clean image of the affect of real international activities on user sentiments. Support Vector Machine algorithm can implement as system studying set of rules to improve the accuracy in sentiment evaluation to categorize the remarks as advantageous, poor and impartial. Experimental outcomes suggests that the proposed system enhance the overall performance of the machine in actual world environments.
Keywords: Social Media, Natural Language Processing, Sentiment Evaluation, Machine Studying, Popularity Prediction.
Scope of the Article: Community Information Systems