AmazonQA: A Review-Based Question Answering Task

Every day, thousands of customers post questions on Amazon product pages. After some time, if they are fortunate, a knowledgeable customer might answer their question. Observing that many questions can be answered based upon the available product reviews, we propose the task of review-based QA. Given a corpus of reviews and a question, the QA system synthesizes an answer. To this end, we introduce a new dataset and propose a method that combines information retrieval techniques for selecting relevant reviews (given a question) and "reading comprehension" models for synthesizing an answer (given a question and review). Our dataset consists of 923k questions, 3.6M answers and 14M reviews across 156k products. Building on the well-known Amazon dataset, we collect additional annotations, marking each question as either answerable or unanswerable based on the available reviews. A deployed system could first classify a question as answerable and then attempt to generate an answer. Notably, unlike many popular QA datasets, here, the questions, passages, and answers are all extracted from real human interactions. We evaluate numerous models for answer generation and propose strong baselines, demonstrating the challenging nature of this new task.

[1]  Chris Dyer,et al.  The NarrativeQA Reading Comprehension Challenge , 2017, TACL.

[2]  Jing He,et al.  Summarization of Yes/No Questions Using a Feature Function Model , 2011, ACML.

[3]  Percy Liang,et al.  Know What You Don’t Know: Unanswerable Questions for SQuAD , 2018, ACL.

[4]  Jason Weston,et al.  Large-scale Simple Question Answering with Memory Networks , 2015, ArXiv.

[5]  Jeffrey Pennington,et al.  GloVe: Global Vectors for Word Representation , 2014, EMNLP.

[6]  Omer Levy,et al.  Annotation Artifacts in Natural Language Inference Data , 2018, NAACL.

[7]  Tat-Seng Chua,et al.  Answering Opinion Questions on Products by Exploiting Hierarchical Organization of Consumer Reviews , 2012, EMNLP.

[8]  Xinyan Xiao,et al.  DuReader: a Chinese Machine Reading Comprehension Dataset from Real-world Applications , 2017, QA@ACL.

[9]  Zachary C. Lipton,et al.  How Much Reading Does Reading Comprehension Require? A Critical Investigation of Popular Benchmarks , 2018, EMNLP.

[10]  Mengting Wan,et al.  Modeling Ambiguity, Subjectivity, and Diverging Viewpoints in Opinion Question Answering Systems , 2016, 2016 IEEE 16th International Conference on Data Mining (ICDM).

[11]  Hugo Zaragoza,et al.  The Probabilistic Relevance Framework: BM25 and Beyond , 2009, Found. Trends Inf. Retr..

[12]  Jianfeng Gao,et al.  A Human Generated MAchine Reading COmprehension Dataset , 2018 .

[13]  Kyunghyun Cho,et al.  SearchQA: A New Q&A Dataset Augmented with Context from a Search Engine , 2017, ArXiv.

[14]  W. P. Lehmann.  Subjectivity , 2015, Encyclopedia of Evolutionary Psychological Science.

[15]  Philip Bachman,et al.  NewsQA: A Machine Comprehension Dataset , 2016, Rep4NLP@ACL.

[16]  A. Agresti,et al.  Approximate is Better than “Exact” for Interval Estimation of Binomial Proportions , 1998 .

[17]  Salim Roukos,et al.  Bleu: a Method for Automatic Evaluation of Machine Translation , 2002, ACL.

[18]  Jian Zhang,et al.  SQuAD: 100,000+ Questions for Machine Comprehension of Text , 2016, EMNLP.

[19]  Julian J. McAuley,et al.  Addressing Complex and Subjective Product-Related Queries with Customer Reviews , 2015, WWW.

[20]  Guokun Lai,et al.  RACE: Large-scale ReAding Comprehension Dataset From Examinations , 2017, EMNLP.

[21]  Oren Etzioni,et al.  Think you have Solved Question Answering? Try ARC, the AI2 Reasoning Challenge , 2018, ArXiv.

[22]  Ming-Wei Chang,et al.  Natural Questions: A Benchmark for Question Answering Research , 2019, TACL.

[23]  Chin-Yew Lin,et al.  ROUGE: A Package for Automatic Evaluation of Summaries , 2004, ACL 2004.