# NL-FIIT at SemEval-2019 Task 9: Neural Model Ensemble for Suggestion Mining

## TL;DR

This paper introduces a neural ensemble model with Bi-LSTM and self-attention for suggestion mining, achieving top-15 results in SemEval-2019 tasks.

## Contribution

It proposes a novel neural architecture combining Bi-LSTM, self-attention, and ELMo embeddings, with ensemble techniques for improved suggestion mining performance.

## Key findings

- Achieved 0.6816 score in subtask A
- Achieved 0.6850 score in subtask B
- Ranked 12th and 10th in official results

## Abstract

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.