Where's My Head? Definition, Dataset and Models for Numeric Fused-Heads Identification and Resolution

26 May 2019  ·  Yanai Elazar, Yoav Goldberg ·

We provide the first computational treatment of fused-heads constructions (FH), focusing on the numeric fused-heads (NFH). FHs constructions are noun phrases (NPs) in which the head noun is missing and is said to be `fused' with its dependent modifier. This missing information is implicit and is important for sentence understanding. The missing references are easily filled in by humans but pose a challenge for computational models. We formulate the handling of FH as a two stages process: identification of the FH construction and resolution of the missing head. We explore the NFH phenomena in large corpora of English text and create (1) a dataset and a highly accurate method for NFH identification; (2) a 10k examples (1M tokens) crowd-sourced dataset of NFH resolution; and (3) a neural baseline for the NFH resolution task. We release our code and dataset, in hope to foster further research into this challenging problem.

PDF Abstract

Datasets


Introduced in the Paper:

Numeric Fused-Head
Task Dataset Model Metric Name Metric Value Global Rank Result Benchmark
Missing Elements Numeric Fused-Head (dev) Bi-LSTM + Scoring + Elmo Accuracy 77.2 # 1
Missing Elements Numeric Fused-Head (dev) Bi-LSTM + Scoring Accuracy 65.6 # 2
Missing Elements Numeric Fused-Head (test) Bi-LSTM + Scoring + Elmo Accuracy 74.0 # 1
Missing Elements Numeric Fused-Head (test) Bi-LSTM + Scoring Accuracy 60.8 # 2

Methods


No methods listed for this paper. Add relevant methods here