Machine Learning Approaches to Sentiment Analysis in Classic and Modern Literature

Sentiment analysis is one of the fields of computer science in charge of solving this problem, you can see it as a sub-field of Natural Language Processing - that has practically turned the way computer models interpret literary texts upside down. In this paper we would explore sentiment analysis on classic and modern literature with the use of machine learning techniques to reveal emotional trends, narrative and sentiments in cultural texts over time. Since major literature creates sensitive human experiences subjectively, using ML implementations to sentiment analysis offers fascinating perspectives on the emotional moods of related texts, the psychological conditions of the authors, and the sociocultural influences in the art forms. That implies literary sentiment analysis is, in its own right, much more demanding than simple tasks on the other side. This is a unique domain for classic literature like old hinteriy, long sentence structures and sophisticated metaphors that are largely more complex than normal English text and therefore must have a higher sophistication level in terms of being the body of text on which the NLP models are being trained. This contrasts modern literature, which is often more directly and colloquially expressed and needs different computational requirements. tion to this field as they are tested against classic as well as contemporary literary texts. Lexicon Based approaches are the oldest techniques for literary sentiment analysis and depend on the predefined lexicon dictating a sentiment polarity. While these approaches effectively solve a basic polarity classification problem, they struggle making sense of more sophisticated phenomena such as context-dependent implications, irony, and new linguistic fashions. This was followed by the conventional ML models including SVM, Naïve Bayes, and Decision Trees being improvised with the linguistic attributes n-grams, part-of-speach, and syntactic dependencies that increased the performance to a higher extent.