Machine Learning for Sentiment Analysis on the Experience Project
This paper applies various machine learning algorithms to predict reader reaction to excerpts from the Experience Project to propose a system to process the documents and to predict human reactions, as well as provide results.
Abstract
The goal of sentiment analysis is to extract human emotions from text. This paper applies various machine learning algorithms to predict reader reaction to excerpts from the Experience Project. Metrics such as accuracy of prediction and precision/recall are presented to gauge the success of these different algorithms. We propose a system to process the documents and to predict human reactions, as well as provide results. We discuss various methods and their advantages and disadvantages in sentiment analysis for these documents. Finally, we comment on applying our findings to sentiment analysis in a more general sense.