conftrace_

2018 EMNLP EMNLP 2018

Possessors Change Over Time: A Case Study with Artworks

Abstract

AbstractThis paper presents a corpus and experimental results to extract possession relations over time. We work with Wikipedia articles about artworks, and extract possession relations along with temporal information indicating when these relations are true. The annotation scheme yields many possessors over time for a given artwork, and experimental results show that an LSTM ensemble can automate the task.

🌉 Interdisciplinary Bridge - Artificial Intelligence and Natural Language Processing

🐣 Hot Topic Early Bird - knowledge extraction

Authors

Dhivya Chinnappa, Eduardo Blanco

Topics

Natural Language Processing > Applications > Information Extraction Deep Learning > Architectures > Recurrent Neural Networks Artificial Intelligence > Core AI > Information Extraction Natural Language Processing > Applications > Relation Extraction

Keywords

knowledge extraction relation extraction temporal information wikipedia article possession relation lstm ensemble artwork corpus

More from EMNLP 2018

All EMNLP 2018 papers →

Label-Free Distant Supervision for Relation Extraction via Knowledge Graph Embedding

Extracting Entities and Relations with Joint Minimum Risk Training

Large-scale Exploration of Neural Relation Classification Architectures

Using Lexical Alignment and Referring Ability to Address Data Sparsity in Situated Dialog Reference Resolution

Subgoal Discovery for Hierarchical Dialogue Policy Learning

Supervised Clustering of Questions into Intents for Dialog System Applications