Ensemble Neural Relation Extraction with Adaptive Boosting

TL;DR

This paper introduces an ensemble neural network model using adaptive boosting and attention mechanisms to improve relation extraction accuracy, effectively handling noisy data and label errors.

Contribution

The paper presents a novel ensemble neural network approach with adaptive boosting and attention for relation extraction, enhancing robustness and performance over existing models.

Findings

Improves F1-score by about 8% over state-of-the-art models.

Effectively handles noisy labels and data in relation extraction.

Demonstrates superior performance on real datasets.

Abstract

Relation extraction has been widely studied to extract new relational facts from open corpus. Previous relation extraction methods are faced with the problem of wrong labels and noisy data, which substantially decrease the performance of the model. In this paper, we propose an ensemble neural network model - Adaptive Boosting LSTMs with Attention, to more effectively perform relation extraction. Specifically, our model first employs the recursive neural network LSTMs to embed each sentence. Then we import attention into LSTMs by considering that the words in a sentence do not contribute equally to the semantic meaning of the sentence. Next via adaptive boosting, we build strategically several such neural classifiers. By ensembling multiple such LSTM classifiers with adaptive boosting, we could build a more effective and robust joint ensemble neural networks based relation extractor. Experiment results on real dataset demonstrate the superior performance of the proposed model, improving F1-score by about 8% compared to the state-of-the-art models.