A Comparative Study on Question-Worthy Sentence Selection Strategies for Educational Question Generation

Chen, Guanliang; Yang, Jie; Gašević, Dragan

doi:10.1007/978-3-030-23204-7_6

Cited by 12 publications

(8 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This information is used during textual generation to construct the actual question. While recent works have shown that content selection for educationally valuable question generation is hard [12] and that many non-educational systems fail to ask about pedagogically valuable contents [13], our expert annotation study result suggests that the generated questions are linguistically-sound, central to the text and possess pedagogical value.…”

Section: Introductionmentioning

confidence: 74%

“…We call this process answer selection. Context selection was tackled through various text statistics [19], [20], through methods stemming from extractive summarization [12], [21] or through text classifiers [15], [22].…”

Section: A Content Selectionmentioning

confidence: 99%

Section: A Content Selectionmentioning

confidence: 99%

“…The transfer of extractive summarization methods to context selection has provided mixed results [12], [21]. In experiments with different data sets, no single best extractive summarization algorithm for content selection could be found [12]. Although LexRank [23] achieves the best scores in a comparison study of different algorithms on different datasets, the study results' are only a very rough estimate, as the study measures performance by generating questions via an upstream question generator.…”

Section: A Content Selectionmentioning

confidence: 99%

“…Furthermore, we phrase the AQG problem differently than related work. Whereas previous studies usually define question-worthy information implicitly through their algorithms [12], [15], [21], the proposed design is guided by the observation that definitions constitute pedagogically valuable information. Consequently, a direct comparison on the question level is difficult, even when conducting human evaluation due to the approaches' different focus.…”

Section: B Limitations Of the Evaluation Studymentioning

confidence: 99%

See 4 more Smart Citations

I Do Not Understand What I Cannot Define: Automatic Question Generation With Pedagogically-Driven Content Selection

Steuer,

Filighera,

Meuser

et al. 2021

Preprint

View full text Add to dashboard Cite

Most learners fail to develop deep text comprehension when reading textbooks passively. Posing questions about what learners have read is a well-established way of fostering their text comprehension. However, many textbooks lack self-assessment questions because authoring them is timeconsuming and expensive. Automatic question generators may alleviate this scarcity by generating sound pedagogical questions. However, generating questions automatically poses linguistic and pedagogical challenges. What should we ask? And, how do we phrase the question automatically? We address those challenges with an automatic question generator grounded in learning theory. The paper introduces a novel pedagogically meaningful content selection mechanism to find question-worthy sentences and answers in arbitrary textbook contents. We conducted an empirical evaluation study with educational experts, annotating 150 generated questions in six different domains. Results indicate a high linguistic quality of the generated questions. Furthermore, the evaluation results imply that the majority of the generated questions inquire central information related to the given text and may foster text comprehension in specific learning scenarios.

show abstract

Section: Introductionmentioning

confidence: 74%