CS 224N Final Project: Unsupervised Clustering of People, Places, and Organizations in Wikileaks Cables with NLP Cues

Our goal is to extract the names of key entities from written U.S. diplomatic communications and then to apply natural-languageand sentiment-based clustering to these entities using contextual features extracted from the data. We extract entities using an off-the-shelf statistical NLP package, and then seek to generate meaningful clusters of entities in an unsupervised fashion. To do so, we experiment with different models, feature sets, and clustering algorithms. For the purposes of this project, we define key entities to be people, nations, and organizations that occur at least k times in the dataset. We determined k through empirical evaluation of the entities extracted from the data, and also the need of different scenarios.