Sentiment Analysis of Social Media Comments using Natural Language Procesing

Sentiment reasoning plays a pivotal role in sophisticated textual data analysis strategies like data processing and computer-aided learning. Data analysis frequently play a huge part in determining how individuals perceive products and make purchase decisions, which is a classic case of market basket analysis. Sentiment analysis uses the Natural Language Processing Toolkit, which is the backbone of the procedure, to gather valuable insights from digital views. It is an integral fragment of opinion mining and web content analysis. With a focus on the Facebook application, which has 3.04 billion active users, the approach to showcase the use of Python's Natural Language Toolkit (NLTK) library to conduct sentiment analysis wherein Graph API is an aid to gather necessary data. Open Refine is a pivotal tool that can be employed to refine the raw statistics to enhance the quality of information and transform the data, making it pertinent for utilization. Using Valence Aware Dictionary and sEntiment Reasoner (VADER) emotion analysis, the suggested method sorts Facebook comments into three categories: negative, neutral, and positive. As automation advances, social media platforms like Facebook become increasingly significant for speech, allowing companies to determine how people comprehend their services based on the responses acquired. It is seen how machine learning techniques such as Support Vector Machine, Random Forest, Gaussian Naïve Bayes, Decision Trees, and Voting Classifiers may be used to undertake the task of opinion mining. Amongst the used algorithms, it can be noted that the Voting Classifier and Decision Tree depict efficient results due to their ensemble and pliable nature respectively. Ultimately, the model affirms the use and efficacy of mood analysis in helping people comprehend and arrive at better decisions in the dynamic realm of Facebook interactions.