Sentiment Analysis of a Social Network to Detect Political Propaganda

This paper focuses on the development of a sentiment analysis model to detect political propaganda using a convergent neural network, specifically BERT (Bidirectional Encoder Representations from Transformers). Available data in the form of comments are preprocessed and tokenized to train the model effectively. The model is then built and fine-tuned using these comments to classify sentiments into positive or negative categories. The effectiveness of the model is evaluated through practical experiments, and the results obtained are examined. The main focus is on understanding how this technique can help accurately identify and analyze political propaganda, which can have a significant impact on political campaigns and public opinion analysis. By utilizing advanced machine learning techniques, this paper aims to contribute to the growing field of natural language processing and sentiment analysis, providing insights that can aid in the detection and understanding of political messaging and its influence on the public. This research underscores the importance of developing sophisticated tools to parse and interpret the vast amounts of data generated by social media and other online platforms, ultimately aiming to foster a more informed and discerning public discourse.