Towards Understanding and Answering Multi-Sentence Recommendation Questions on Tourism
arXiv:1801.01825 · doi:10.1017/S1351324920000017
Abstract
We introduce the first system towards the novel task of answering complex multisentence recommendation questions in the tourism domain. Our solution uses a pipeline of two modules: question understanding and answering. For question understanding, we define an SQL-like query language that captures the semantic intent of a question; it supports operators like subset, negation, preference and similarity, which are often found in recommendation questions. We train and compare traditional CRFs as well as bidirectional LSTM-based models for converting a question to its semantic representation. We extend these models to a semisupervised setting with partially labeled sequences gathered through crowdsourcing. We find that our best model performs semi-supervised training of BiDiLSTM+CRF with hand-designed features and CCM(Chang et al., 2007) constraints. Finally, in an end to end QA system, our answering component converts our question representation into queries fired on underlying knowledge sources. Our experiments on two different answer corpora demonstrate that our system can significantly outperform baselines with up to 20 pt higher accuracy and 17 pt higher recall.
References in corpus (11)
- Efficient Estimation of Word Representations in Vector Space
- HuggingFace's Transformers: State-of-the-art Natural Language Processing
- Large-scale Simple Question Answering with Memory Networks
- LSTM-based Deep Learning Models for Non-factoid Answer Selection
- Know What You Don't Know: Unanswerable Questions for SQuAD
- DROP: A Reading Comprehension Benchmark Requiring Discrete Reasoning Over Paragraphs
- Learning Dependency-Based Compositional Semantics
- Task-Oriented Query Reformulation with Reinforcement Learning
- Question Answering on Freebase via Relation Extraction and Textual Evidence
- Medical Exam Question Answering with Large-scale Reading Comprehension
- Jointly Predicting Predicates and Arguments in Neural Semantic Role Labeling