LSTM-based Deep Learning Models for Non-factoid Answer Selection

TL;DR

This paper introduces LSTM-based deep learning models for non-factoid answer selection, combining CNNs and attention mechanisms to improve question-answer matching without manual features.

Contribution

It proposes novel LSTM-based models with CNN and attention mechanisms for answer selection, outperforming existing baselines on multiple datasets.

Findings

Models outperform strong baselines on TREC-QA and InsuranceQA datasets.

Combining CNNs with biLSTM improves representation quality.

Attention mechanisms enhance answer relevance modeling.

Abstract

In this paper, we apply a general deep learning (DL) framework for the answer selection task, which does not depend on manually defined features or linguistic tools. The basic framework is to build the embeddings of questions and answers based on bidirectional long short-term memory (biLSTM) models, and measure their closeness by cosine similarity. We further extend this basic model in two directions. One direction is to define a more composite representation for questions and answers by combining convolutional neural network with the basic framework. The other direction is to utilize a simple but efficient attention mechanism in order to generate the answer representation according to the question context. Several variations of models are provided. The models are examined by two datasets, including TREC-QA and InsuranceQA. Experimental results demonstrate that the proposed models substantially outperform several strong baselines.