Natural Language Information Extraction
We develop relation and attribute extraction techniques that are robust to limited labeled data and noise. Our work includes leveraging semantic information and prompts for zero-shot relation extraction, and visual semantic information for multimodal multifeature few-shot relation extraction. We also develop n-ary cross-sentence relation extraction methods for both supervised and unsupervised settings.
Funded Projects
- Heterogeneous Hypergraph Modeling for Zero-shot Product Aspect Identification – EBAY 2024
- Natural Language Processing for Teaching and Research in Engineering Education – National Science Foundation (NSF) 2022
- Content Analytics for Smart Search – Xerox Workplace Innovation Research 2018
- Insider Threat Detection for Document Content Security – Xerox Workplace Innovation Research Program 2017
Publications
2024
[t1] Few-Shot and Zero-Shot Learning for Information Extraction
J. Gong
PhD Dissertation
[j1] A Reinforcement Learning Framework for N-ary Document-Level Relation Extraction
C. Yuan, R. Rossi, A. Katz, and H. Eldardiry
IEEE Transactions on Big Data 2024
Impact Factor 7.2
[c1] Prompt-based Zero-Shot Relation Extraction with Semantic Knowledge Augmentation
J. Gong and H. Eldardiry
ACL International Conference on Computational Linguistics, Language Resources and Evaluation (COLING-LREC) 2024
Acceptance Rate 28%
[c2] Multilabel Zero-Shot Product Attribute-Value Extraction
J. Gong and H. Eldardiry
ACM International World Wide Web Conference (WWW) 2024
Acceptance Rate 20.2%
[c3] Multimodal Few-Shot Relation Extraction with Hybrid Visual Evidence
J. Gong and H. Eldardiry
ACL International Conference on Computational Linguistics, Language Resources and Evaluation (COLING-LREC) 2024
Acceptance Rate 28%
2023
[c4] Knowledge-Enhanced Multilabel Few-Shot Product Attribute-Value Extraction
J. Gong, W. T. Chen, and H. Eldardiry
International Conference on Information and Knowledge Management (CIKM) 2023
Acceptance Rate 24%
[i1] GradXKG: A Universal Explain-per-use Temporal Knowledge Graph Explainer
C. Yuan and H. Eldardiry
arXiv.2310.04889 2023
2022
[c5] Clustering-based Unsupervised Generative Relation Extraction
C. Yuan, R. A. Rossi, A. Katz, and H. Eldardiry
IEEE International Conference on Big Data 2022
Acceptance Rate 19.2%
2021
[t2] N-ary Cross-sentence Relation Extraction: From Supervised to Unsupervised Learning
C. Yuan
MS Thesis
[c6] Unsupervised Relation Extraction: a Variational Autoencoder Approach
C. Yuan and H. Eldardiry
Conference on Empirical Methods in Natural Language Processing (EMNLP) 2021
Acceptance Rate 32%
[c7] Zero-Shot Relation Classification from Side Information
J. Gong and H. Eldardiry
ACM International Conference on Information and Knowledge Management (CIKM) 2021
Acceptance Rate 21.7%
[c8] Multistage Hybrid Attentive Networks for Knowledge-Driven Stock Movement Prediction
J. Gong and H. Eldardiry
ACM International Conference on Neural Information Processing (ICONIP) 2021
Acceptance Rate 20.9%
[c9] Predicting Stock Price Movement Using Financial News Sentiment
J. Gong, B. Paye, G. Kadlec, and H. Eldardiry
International Conference on Engineering Applications of Neural Networks 2021
2019
[c10] Building Jarvis - A Learner-aware Conversational Trainer
S. Mohan, K. Ramea, B. Price, M. Shreve, and H. Eldardiry, and L. Nelson
ACM user2agent IUI Workshop on User-aware Conversational Agents 2019