Authorship Attribution Using Sentiment Analysis
License
This project is licensed under the MIT License.
Description
Authorship prediction refers to the association of an author to a specific document. In literature, it is very important to know about historical text who wrote it and when it was written. However, we can also predict the nationality of the author, characteristic styles of the author, characteristic styles of the author and genre of the text.
Here, I explored the problem of Authorship Prediction using Text Mining and Sentiment Analysis. This analysis predicts the authors based on a collection of text derived from their authored books. We have performed experiments on books that are extracted from Victorian Era Authors datasets by using different features such as bag of words, n-grams or TF-IDF models. Also implemented different classifier techniques to improve the success rates. This project is segmented into two phases, the first phase including the execution of a robust text mining task and sentiment analysis; the second phase involves the experimentation of different classification models.
Read our report here: Text_Mining_and_Authorship_Prediction