NL-FIIT at SemEval-2019 Task 9: Neural Model Ensemble for Suggestion Mining
This work addresses suggestion mining for natural language processing applications, but it is incremental as it builds on existing neural methods and achieved mid-tier competition results.
The paper tackled suggestion mining from online reviews and forums by proposing a neural model ensemble using Bi-LSTM and self-attention with ELMo embeddings, achieving official test scores of 0.6816 and 0.6850 in two subtasks.
In this paper, we present neural model architecture submitted to the SemEval-2019 Task 9 competition: "Suggestion Mining from Online Reviews and Forums". We participated in both subtasks for domain specific and also cross-domain suggestion mining. We proposed a recurrent neural network architecture that employs Bi-LSTM layers and also self-attention mechanism. Our architecture tries to encode words via word representations using ELMo and ensembles multiple models to achieve better results. We performed experiments with different setups of our proposed model involving weighting of prediction classes for loss function. Our best model achieved in official test evaluation score of 0.6816 for subtask A and 0.6850 for subtask B. In official results, we achieved 12th and 10th place in subtasks A and B, respectively.