Mention Detection Using Pointer Networks for Coreference Resolution

Abstract

A mention has a noun or noun phrase as its head and constructs a chunk that defines any meaning, including a modifier. Mention detection refers to the extraction of mentions from a document. In mentions, coreference resolution refers to determining any mentions that have the same meaning. Pointer networks, which are models based on a recurrent neural network encoderdecoder, outputs a list of elements corresponding to an input sequence. In this paper, we propose mention detection using pointer networks. This approach can solve the problem of overlapped mention detection, which cannot be solved by a sequence labeling approach. The experimental results show that the performance of the proposed mention detection approach is F1 of 80.75%, which is 8% higher than rule-based mention detection, and the performance of the coreference resolution has a CoNLL F1 of 56.67% (mention boundary), which is 7.68% higher than coreference resolution using rule-based mention detection.

Neural Coreference Resolution for Turkish

“…On the other hand, deep learning-based approaches hinder feature engineering and capture both syntactic and semantic features of candidate mentions via word embeddings. Different neural architectures were explored for mention detection such as the Bidirectional long short-term memory (BiLSTM) with conditional random field (CRF) (Park and Lee, 2015), the pointer network (Park et al, 2017), and the stacked LSTM enhanced with stack pointer (Wang et al, 2018).…”