A Semantic Role-based Approach to Open-Domain Automatic Question Generation
Open accessThe DOI matches its Crossref record; 31 of the 34 references it cites were found in a registry or an index; none of the 15 sources its claims cite is checked for support yet.
Abstract
We present a novel rule-based system for automatic generation of factual questions from sentences, using semantic role labeling (SRL) as the main form of text analysis.The system is capable of generating both wh-questions and yes/no questions from the same semantic analysis.We present an extensive evaluation of the system and compare it to a recent neural network architecture for question generation.The SRL-based system outperforms the neural system in both average quality and variety of generated questions.
Overview
CoChat Summary
## TL;DR
The paper presents a rule-based open-domain question generation system that uses semantic role labeling to generate both wh-questions and yes/no questions from the same analysis. On 171 sentences, it outperformed an LSTM attention-based neural baseline in average quality, variety, and the share of questions judged potentially usable.
## Key findings
- The SRL-based system generated questions for 165 of 171 sentences, producing 890 questions in total: 236 yes/no questions and 654 constituent questions.
- Mean ratings favored the SRL system over the neural system on all three scales and on total score: SRL-YNQ 4.32/4.34/2.75/11.41, SRL-CQ 3.89/3.79/2.52/10.20, NN 3.75/2.61/1.65/8.01 for grammar/semantics/relevance/total.
- The best SRL constituent question per sentence had a mean total rating of 12.2 versus 8.1 for the neural system’s questions; the difference was statistically significant (t-test, p < .0001).
- Using the paper’s “potentially useful” criteria, 81% of SRL yes/no questions, 64% of SRL constituent questions, and 29% of neural questions had total rating ≥ 10; with the stricter per-scale threshold, the proportions were 71%, 50%, and 15%, respectively.
## Methods
- Rule-based AQG pipeline: tokenization, sentence boundary detection, POS tagging with OpenNLP, verbal-group detection, semantic role labeling with SENNA, postprocessing, and question generation.
- Data/sample for evaluation: five expository texts, yielding 171 declarative sentences (at least five words, no conditional clauses); the SRL and neural systems each generated questions for all 171 sentences, though the SRL system failed on 6 sentences and the neural system on 2.
- Human evaluation: 1,060 generated questions were rated by two annotators on grammar, semantics, and relevance; analysis used average ratings, ANOVA with Bonferroni-adjusted pairwise contrasts, and a t-test for best-per-sentence constituent questions.
## Limitations
- The paper states the SRL system falters on long and complicated sentences, especially those with subordinate clauses.
- The neural baseline cannot generate yes/no questions because it is trained on SQuAD, which contains only constituent-style questions; the paper notes that beam-search hypotheses were not fully formed and only small variants of the best question.
- Evident from the text: the evaluation is limited to 171 sentences from five expository texts, so the results are based on a relatively small and specific sample.
Model-written summary. Check it against the paper before relying on it.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
0 Valid · 1 Not valid · 8 Not checked · 51 Own results and interpretations
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“The SRL-based system outperforms the neural system in both average quality and variety of generated questions.”
plainstates a cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
The paper states this without a source or data behind it here, so there is nothing to check it against. The quote shows it is in the paper.
In the paper
“The system generates questions directly from semantic analysis, without templates.”
plainstates a cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: unclear — It is unclear why the paper says it holds.
- Rests on an assumption not shown here.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“The SRL-based system outperforms the neural system in both average quality and variety of generated questions.”
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A result the paper states — it cites no source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“The system is capable of generating both wh -questions and yes/no questions from the same semantic analysis.”
plainstates a cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Rests on an assumption not shown here.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“The SENNA system tends to assign the A1 role for subjects instead of A0. For example, for John arrived today , 'John' is assigned A1. This also often happens for copula sentences, e.g. SENNA produces: [ A 1 John] is [ A 1 a painter] .”
plainstates an association- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
24 more claims
Create a free account to see all 60 claims and to verify the 15 references still open.
- It may state two separate claims as one. 18
- Probably not a substantial claim. 14
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
The cited source says what the claim attributes to it.
In the citing paper
“There are two major approaches for open-domain questions, which have been researched upon. The first approach makes use of the syntactic structure of sentences and other natural language processing operations to produce a question from a specified sentence, and the second approach makes use of an end-to-end method which uses an approach similar to machine language translation using neural networks”
From Automatic question generation: a review of methodologies, datasets, evaluation metrics, and applications (2023) · cites this paper as [10]
Citations
Cites
Glove: Global Vectors for Word Representation
Pennington, J., Socher, R., & Manning, C. (2014) · Proceedings of the 2014 Conference on Empirical Methods in Natural Language Processing (EMNLP) · doi:10.3115/v1/d14-1162 · Cited by 34,071
Effective Approaches to Attention-based Neural Machine Translation
Luong, T., Pham, H., & Manning, C. D. (2015) · Proceedings of the 2015 Conference on Empirical Methods in Natural Language Processing · doi:10.18653/v1/d15-1166 · Cited by 8,382
SQuAD: 100,000+ Questions for Machine Comprehension of Text
Rajpurkar, P., Zhang, J., Lopyrev, K., & Liang, P. (2016) · Proceedings of the 2016 Conference on Empirical Methods in Natural Language Processing · doi:10.18653/v1/d16-1264 · Cited by 6,208 · cited 1×
“We trained the network on the preprocessed version of the SQuAD dataset (Rajpurkar et al., 2016) provided by Du et al. (2017).”
MethodNatural Language Processing (almost) from Scratch
Collobert, R., Weston, J., Bottou, L., Karlen, M., Kavukcuoglu, K., & Kuksa, P. P. (2011) · arXiv (Cornell University) · doi:10.48550/arxiv.1103.0398 · Cited by 4,962
The Proposition Bank: An Annotated Corpus of Semantic Roles
Palmer, M., Gildea, D., & Kingsbury, P. (2005) · Computational Linguistics · doi:10.1162/0891201053630264 · Cited by 2,273
Learning to Ask: Neural Question Generation for Reading Comprehension
Du, X., Shao, J., & Cardie, C. (2017) · Proceedings of the 55th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) · doi:10.18653/v1/p17-1123 · Cited by 696 · cited 10×
“We use the sentence-oriented model5 from Du et al. (2017), where only the input sentence is encoded.”
BackgroundMethodInfluentialGood Question! Statistical Ranking for Question Generation
Heilman, M., & Smith, N. A. (2018) · Figshare · doi:10.1184/r1/6473390 · Cited by 366
HeidelTime: High Quality Rule-Based Extraction and Normalization of Temporal Expressions
Strötgen, J., & Gertz, M. (2010) · Cited by 293
Semeval-2010 task 13: Tempeval-2
Verhagen, M., Sauri, R. S., & Pustejovsky, J. E. (2010) · Cited by 280
Generating Factoid Questions With Recurrent Neural Networks: The 30M Factoid Question-Answer Corpus
Serban, I. V., García-Durán, A., Gulcehre, C., Ahn, S., Chandar, S., Courville, A., & Bengio, Y. (2016) · Proceedings of the 54th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) · doi:10.18653/v1/p16-1056 · Cited by 280 · cited 2×
“With recent advances in neural networks, some approaches forgo most linguistic analysis and train neural networks to generate questions from sequences of word tokens (Du et al., 2017; Serban et al., 2016).”
BackgroundGenerating Natural Language Questions to Support Learning On-Line
Lindberg, D., Popowich, F. P., Nesbit, J. C., & Winne, P. H. (2013) · Cited by 122
Automatic Gap-fill Question Generation from Text Books
Agarwal, M., & Mannem, P. (2011) · Cited by 109
Identifying Where to Focus in Reading Comprehension for Neural Question Generation
Du, X., & Cardie, C. (2017) · Proceedings of the 2017 Conference on Empirical Methods in Natural Language Processing · doi:10.18653/v1/d17-1219 · Cited by 100 · cited 1×
“Recently, Du and Cardie (2017) described a neural architecture for the content selection task in AQG.”
BackgroundQuestion Generation from Paragraphs at UPenn: QGSTEC System Description
Mannem, P., Prasad, R., & Joshi, A. K. (2010) · Cited by 82
Mind the Gap: Learning to Choose Gaps for Question Generation
Becker, L. A., Basu, S., & Vanderwende, L. (2012) · North American Chapter of the Association for Computational Linguistics · Cited by 77
G-Asks: An Intelligent Automatic Question Generation System for Academic Writing Support
Liu, M., Calvo, R. A., & Rus, V. (2012) · Dialogue & Discourse · doi:10.5087/dad.2012.205 · Cited by 77
Semantics-based Question Generation and Implementation
Yao, X., Bouma, G., & Zhang, Y. (2012) · Dialogue & Discourse · doi:10.5087/dad.2012.202 · Cited by 71
Automatic question generation from text - an aid to independent study
Wolfe, J. H. (1976) · Proceedings of the ACM SIGCSE-SIGCUE technical symposium on Computer science and education - · doi:10.1145/800107.803459 · Cited by 66
Generating Questions and Multiple-Choice Answers using Semantic Analysis of Texts.
Araki, J., Rajagopal, D., Sankaranarayanan, S., Holm, S., Yamakawa, Y., & Mitamura, T. (2016) · International Conference on Computational Linguistics · Cited by 48
Rating Computer-Generated Questions with Mechanical Turk
Heilman, M., & Smith, N. A. (2010) · KiltHub Repository · doi:10.1184/r1/6473645 · Cited by 39
Infusing NLU into Automatic Question Generation
Mazidi, K., & Tarau, P. (2016) · Proceedings of the 9th International Natural Language Generation conference · doi:10.18653/v1/w16-6609 · Cited by 36 · cited 2×
“Template-based methods allow asking questions that can go beyond the explicit information in a text (Mazidi and Tarau, 2016; Lindberg et al., 2013).”
BackgroundThe Importance of Being Important: Question Generation
Vanderwende, L. (2008) · Cited by 35
Leveraging Multiple Views of Text for Automatic Question Generation
Mazidi, K., & Nielsen, R. D. (2015) · Lecture notes in computer science · doi:10.1007/978-3-319-19773-9_26 · Cited by 34
Automatic generation of short answer questions for reading comprehension assessment
HUANG, Y., & HE, L. (2016) · Natural Language Engineering · doi:10.1017/s1351324915000455 · Cited by 30 · cited 3×
“Huang and He (2016) present an AQG system that uses the Lexical Functional Grammar representation, including syntactic and semantic layers.”
BackgroundMethodAutomatic question generation from text - an aid to independent study
Wolfe, J. H. (1976) · ACM SIGCSE Bulletin · doi:10.1145/952989.803459 · Cited by 26
Pedagogical Evaluation of Automatically Generated Questions
Mazidi, K., & Nielsen, R. D. (2014) · Lecture notes in computer science · doi:10.1007/978-3-319-07221-0_36 · Cited by 16
Ranking Automatically Generated Questions Using Common Human Queries
Chali, Y., & Golestanirad, S. (2016) · International Conference on Natural Language Generation · doi:10.18653/v1/w16-6635 · Cited by 16
A Syntactic Approach to Domain-Specific Automatic Question GenerationPreprint
Danon, G., & Last, M. (2017) · arXiv (Cornell University) · doi:10.48550/arxiv.1712.09827 · Cited by 15
SUTime: Evaluation in TempEval-3
Chang, A. L. S., & Manning, C. D. (2013) · Cited by 14
Collecting Reliable Human Judgements on Machine-Generated Language: The Case of the QG-STEC Data
Godwin, K., & Piwek, P. (2016) · International Conference on Natural Language Generation · doi:10.18653/v1/w16-6634 · Cited by 9
QGASP: a Framework for Question Generation Based on Different Levels of Linguistic Information
Rodrigues, H. P., Coheur, L., & Nyberg, E. (2016) · International Natural Language Generation Conference (INLG) · doi:10.18653/v1/w16-6640 · Cited by 6
Research works printed but not linked: 1
- Frank R. Palmer. 1987. The English Verb , 2nd edition. Longman, London, UK.
Cited by
A Systematic Review of Automatic Question Generation for Educational Purposes
Kurdi, G., Leo, J., Parsia, B., Sattler, U., & Al-Emari, S. (2020) · International Journal of Artificial Intelligence in Education · doi:10.1007/s40593-019-00186-y · Cited by 514
Ferreira‐Mello, R., André, M., Pinheiro, A., Costa, E., & Romero, C. (2019) · WIREs Data Mining and Knowledge Discovery · doi:10.1002/widm.1332 · Cited by 193
“For open questions, the leading research line is to semantically evaluate student’s answers (Cutrone and Chang, 2010) and generate new questions (Flor and Riordan, 2018).”
BackgroundMulla, N., & Gharpure, P. (2023) · Progress in Artificial Intelligence · doi:10.1007/s13748-023-00295-9 · Cited by 132
“…constituent and dependency parsing [8], a representation using lexical functional grammar [9], semantic role labeling in a rule-based set-up [10], and neural network-based approaches including the generation of factoid questions using recurrent neural networks [11], where to focus on for…”
MethodA Review on Question Generation from Natural Language Text
Zhang, R., Guo, J., Chen, L., Fan, Y., & Cheng, X. (2021) · ACM Transactions on Information Systems · doi:10.1145/3468889 · Cited by 153
Syn-QG: Syntactic and Shallow Semantic Rules for Question Generation
Dhole, K., & Manning, C. D. (2020) · Proceedings of the conference - Association for Computational Linguistics. Meeting · doi:10.18653/v1/2020.acl-main.69 · Cited by 64
Metadata
- Authors
- Flor, M., & Riordan, B.
- Affiliations
- Educational Testing Service (US)
- Venue
- Proceedings of the Thirteenth Workshop on Innovative Use of NLP for Building Educational Applications
- Published
- 2018
- Pages
- 254-263
- OpenAlex
- W2807132464
- Cited by
- 39
- References
- 34
- Citations by year
- 2026: 2 · 2025: 5 · 2024: 1 · 2023: 7 · 2022: 2 · 2021: 10 · 2020: 4 · 2019: 7 · 2018: 1
- FWCI
- 2.25
- Type
- conference paper
- Language
- en
- Open access
- hybrid
- doi.org
- Licence
- cc-by
- Locations
- Workshop on Innovative Use of NLP for Building Educational Applications (publishedVersion, cc-by)
- Topics
- Physical Sciences › Computer Science › Artificial Intelligence › Natural Language Processing Techniques
- Physical Sciences › Computer Science › Artificial Intelligence › Topic Modeling
- Physical Sciences › Computer Science › Artificial Intelligence › Speech and dialogue systems
- Keywords
- automatic question generation, semantic role labeling, factual questions, wh-questions, yes-no questions
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
A language model drew the claims from the paper on 2026-10-11, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).
A second model labelled each claim on 2026-10-11 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.