From Texts to Networks
Jana Diesner, Kathleen M. Carley · 2008
This work is part of the Dynamic Networks project in the center for Computational Analysis of Social and Organizational Systems (CASOS) of the School of Computer Science (SCS) at Carnegie Mellon University (CMU). Support was provided, in part, by the National Science Foundation (NSF) Integrative Graduate Education and Research Traineeship (IGERT) program, 9972762, the Army Research Lab, and the Army Research Institute .The views and conclusions contained in this document are those of the authors and should not be interpreted as representing the official policies, either expressed or implied, of the Army Researc Lab, the Army Research Institute, the National Science Foundation, or the U.S. government. Development of Computational Solutions Utilize machinery from Machine Learning and Artificial Intelligence Deploy and develop supervised and semisupervised sequential stochastic learning techniques in order to train classifiers and build models that generalize to new data Construct a classifier h that for every sequence of (x, y) (joint probability) (where x = words per sequence and y = corresponding category) or (x|y) (conditional probability) predicts a sequence y = h (x) for any sequence of x, incl. new and unseen data We work with Generative (aka discriminative) models: P(x,y), such as Hidden Markov Model (HMM), and Conditional models: P(y|x), such as Maximum Entropy Markov Models (MEMM) and Conditional Random Fields (CRF)