Deep learning and natural language processing have enjoyed widespread application in consumer and healthcare technology, but they have less often been used for analysis in the field of humanities. In this blog post, we take a look at one application of deep learning to analyze texts that detail the history of the Israeli-Palestinian conflict, perhaps “the most intractable conflict of our time.”
We find that given a text that was written by the conflict, we are able to use deep learning to detect whether it was a written by an Israeli or a Palestinian with 85–92% accuracy. The paper associated with this blog post is [here], and the data and code can be found [here].

#digital-humanities #machine-learning #text-classification #nlp #computational-linguistics