Social Media Text Analysis

• Tech Stack: Python and various Python libraries: NLTK, Sklearn, Pandas, Seaborn, Matplotlib.
• Github URL: Project Link

In this project, I focused on processing and analyzing social media content to classify sentiment and stance using machine learning techniques. I pre-processed the dataset of tweets by applying both minimal and maximal cleaning methods, tokenized the data, and extracted features using POS tagging and polarity measurement. I calculated descriptive statistics to understand text length and sentiment distribution across different targets and stances. I implemented an SVM model to classify the data based on these features. My analysis highlighted challenges in POS tagging social media content and compared the effectiveness of TF-IDF and CountVectorizer. Overall, this project provided valuable insights into sentiment patterns and demonstrated the potential of machine learning in social media analysis.