COREFERENCE RESOLUTION FOR THE ENGLISH LANGUAGE
ADRIEL GARCIA HERNANDEZ · 2017
One of the problems found in natural language processing systems, is the difficulty to identify textual elements referring to the same entity, this task is called coreference.Solving this problem is an integral part of discourse comprehension since it allows language users to connect the pieces of speech information concerning to the same entity.Consequently, coreference resolution is a key task in natural language processing.Despite the large efforts of existing research, the current performance of coreference resolution systems has not reached a satisfactory level yet.In this work, we describe a structure learning system for unrestricted coreference resolution that explores two techniques: latent coreference trees and automatic entropy-guided feature induction.The latent tree modeling makes the learning problem computationally feasible, since it incorporates a relevant hidden structure.Additionally, using an automatic feature induction method, we can efciently build enhanced non-linear models using linear model learning algorithms, namely, the structured and sparse perceptron algorithm.We evaluate the system on the CoNLL-2012 Shared Task closed track data set, for the English portion.The proposed system obtains a 62.24% value on the competition's official score.This result is below the 65.73%, the stateof-the-art performance for this task.Nevertheless, our solution significantly reduces the time to obtain the clusters of a document, since, our system takes 0.35 seconds per document in the testing set, while in the state-ofthe-art, it takes 5 seconds for each one.