International Research Journal of Engineering and Technology (IRJET)
e-ISSN: 2395-0056
Volume: 11 Issue: 10 | Oct 2024
p-ISSN: 2395-0072
www.irjet.net
LEVERAGING MACHINE LEARNING TECHNIQUES FOR ANALYZING AND IDENTIFYING SENTIMENT IN SOCIAL MEDIA POSTS Kaneez Fatma1, Dipti Ranjan Tiwari2 1Master of Technology, Computer Science and Engineering, Lucknow Institute of Technology, Lucknow, India 2Assistant Professor, Department of Computer Science and Engineering, Lucknow Institute of Technology,
Lucknow, India ---------------------------------------------------------------------***---------------------------------------------------------------------
Abstract - Sentiment analysis is a field that examines
Machine learning methodologies have been extensively embraced to refine the precision and scalability of sentiment classification, rendering it an indispensable instrument for scrutinizing extensive quantities of online material.
people's opinions on various topics, from products to political and social events. It has gained widespread attention as it helps stakeholders make informed decisions based on public sentiment. Opinion mining is a key technique used to extract insights from platforms like search engines, blogs, Twitter, and social networks. However, manually analyzing large volumes of tweets, which are often in unstructured text form, can be challenging. To overcome this, researchers use computational techniques like the Bag-of-Words (BoW) model, which identifies sentiment-bearing words through machine learning. In this study, a lexicon-based method was employed to automatically detect sentiments in tweets collected from Twitter. Researchers applied three machine learning algorithms: Naive Bayes (NB), Maximum Entropy (ME), and Support Vector Machines (SVM), to assess their effectiveness in classifying tweets by sentiment. The results showed that both NB with Laplace smoothing and SVM were reliable classifiers, especially when using specific features like unigrams or Partof-Speech (POS) tags. Overall, sentiment analysis is a valuable tool for understanding public opinions shared on platforms like Twitter, allowing stakeholders to gauge public reactions to various subjects.
Figure-1: Social Media Sentiment
2.INTRODUCTION Sentiment identification plays a crucial role in various aspects such as understanding public opinions, customer feedback, and emerging trends. By delving into how people perceive products, services, events, or societal issues, businesses can gain valuable insights that drive actionable decisions. For instance, when a company analyzes customer feedback, it can pinpoint specific areas for improvement, leading to enhanced product quality and increased customer satisfaction. This, in turn, fosters brand loyalty and strengthens the company's market position.
Key Words: Sentiment Analysis, Social Media, Natural Language Processing (NLP), Deep Learning, Transformers, Supervised Learning, Text Classification.
1.SENTIMENT ANALYSIS IN SOCIAL MEDIA AND ONLINE POSTS
In the realm of public opinion, sentiment analysis serves as a powerful tool for governments, organizations, and media outlets. By monitoring reactions to policies, news events, or social movements, stakeholders can gauge public sentiment and make informed decisions. For example, during a political campaign, tracking sentiment trends can offer valuable insights into voter behavior and preferences, helping candidates tailor their strategies for maximum impact.
Sentiment analysis, also referred to as opinion mining, is a pivotal domain within natural language processing (NLP) that entails discerning the emotional undertones within a body of text. In the realm of social media and online discourse, it holds substantial significance in comprehending the sentiments articulated by users across various platforms such as Twitter, Facebook, and online forums. Given the proliferation of user-generated content, organizations, scholars, and governmental bodies are increasingly dependent on sentiment analysis to assess public perceptions pertaining to products, services, political occurrences, and societal matters. Nonetheless, owing to the informal and colloquial lexicon utilized in these posts— frequently incorporating slang, emoticons, and ambivalent sentiments—the task presents distinctive complexities.
© 2024, IRJET
|
Impact Factor value: 8.315
Moreover, sentiment analysis enables industries to anticipate consumer behavior, predict market shifts, and even forecast political outcomes. By leveraging sentiment data, businesses can stay ahead of the competition and make strategic decisions that align with evolving trends. This proactive approach not only enhances decision-making processes but also supports long-term planning efforts,
|
ISO 9001:2008 Certified Journal
|
Page 597