Capturing Curiosity : A Comparison of Machine Learning Algorithms for Classification of Curiosity in Text
Abstract: The amount of text available to us on a daily basis, in the form of blogs, news articles, and social media updates, is larger then it has ever been. Being able to analyze large amounts of text and being able to determine its overall sentiment by using machinelearning algorithms has been a large area of research during the last few decades. This thesis will attempt to build on that work by looking at three different algorithms – Naive Bayes, Support Vector Machine and J48 Decision Tree, and evaluating their performance on the special problem of identifying curiosity in text. It also examines differences in result depending on how the feature selection is performed. The results indicate that Naive Bayes performs the best at the task.
AT THIS PAGE YOU CAN DOWNLOAD THE WHOLE ESSAY. (follow the link to the next page)