Neural Question Generation from Text: A Preliminary Study (2017-04-06T00:00:00.000000Z)

TL;DR

A preliminary study on neural question generation from text with the SQuAD dataset is conducted, and the experiment results show that the method can produce fluent and diverse questions.

Abstract

Automatic question generation aims to generate questions from a text passage where the generated questions can be answered by certain sub-spans of the given passage. Traditional methods mainly use rigid heuristic rules to transform a sentence into related questions. In this work, we propose to apply the neural encoder-decoder model to generate meaningful and diverse questions from natural language sentences. The encoder reads the input text and the answer position, to produce an answer-aware input representation, which is fed to the decoder to generate an answer focused question. We conduct a preliminary study on neural question generation from text with the SQuAD dataset, and the experiment results show that our method can produce fluent and diverse questions.

Authors

M. Zhou

9 papers

Furu Wei

37 papers

Chuanqi Tan

8 papers

TL;DR

Abstract

Authors

References33 items

Machine Comprehension Using Match-LSTM and Answer Pointer

SQuAD: 100,000+ Questions for Machine Comprehension of Text

Pointing the Unknown Words

Generating Factoid Questions With Recurrent Neural Networks: The 30M Factoid Question-Answer Corpus

Abstractive Text Summarization using Sequence-to-sequence RNNs and Beyond

Effective Approaches to Attention-based Neural Machine Translation

Towards Topic-to-Question Generation

Adam: A Method for Stochastic Optimization

GloVe: Global Vectors for Word Representation

Neural Machine Translation by Jointly Learning to Align and Translate

Learning Phrase Representations using RNN Encoder–Decoder for Statistical Machine Translation

The Stanford CoreNLP Natural Language Processing Toolkit

Maxout Networks

On the difficulty of training recurrent neural networks

Understanding the difficulty of training deep feedforward neural networks

Bleu: a Method for Automatic Evaluation of Machine Translation

Foundations of Statistical Natural Language Processing

Book Reviews: Foundations of Statistical Natural Language Processing

2015) and simple SGD as our the optimizing algorithms. The training is separated into two phases, the first phase is optimizing the loss function with Adam and the second is with simple SGD

Dropout: a simple way to prevent neural networks from overfitting

A Fast and Accurate Dependency Parser using Neural Networks

NQG+Pretrain Based on NQG+, we initialize the word embedding matrix with pre-trained GloVe

Automatic factual question generation from text

PCFG-Trans The rule-based system 1 modiﬁed on the code released by Heilman

Model Training We initialize model parameters randomly using a Gaussian distribution with Xavier scheme

Measuring nominal scale agreement among many raters.

PCFG-Trans baseline and NQG++ with human judges

s2s+att We implement a seq2seq with attention as the baseline method

NQG+STshare Based on NQG+, we make the encoder and decoder share the same embedding matrix

NQG+ Based on NQG, we incorporate copy mechanism to deal with rare words problem

NQG We extend the s2s+att with our feature-rich encoder to build the NQG system

of 3rd International Conference for Learning Repre-sentations

NQG − NER Ablation test, the NER feature is removed from NQG model

Field of Study

Journal Information

Name

Volume

Venue Information

Name

Type

URL

Alternate Names