Generating Quizzes to Support Training on Quality Management and Assurance in Space Science and Engineering Andres Garcia-Silva Cristian Berrio Jose Manuel Gomez-Perez

2025-05-06 0 0 118.48KB 3 页 10玖币

侵权投诉

Generating Quizzes to Support Training on Quality Management and

Assurance in Space Science and Engineering

Andres Garcia-Silva, Cristian Berrio, Jose Manuel Gomez-Perez

Expert.ai / Madrid, Spain

agarcia@expert.ai,cberrio@expert.ai,jmgomez@expert.ai

Jose Antonio Martinez-Heras, Patrick Fleith

Solenix / Darmstadt, Germany

jose.martinez@solenix.ch

Stefano Scaglioni

ESA / Darmstadt, Germany

stefano.scaglioni@esa.int

Abstract

Quality management and assurance is key for

space agencies to guarantee the success of

space missions, which are high-risk and ex-

tremely costly. In this paper, we present a sys-

tem to generate quizzes, a common resource

to evaluate the effectiveness of training ses-

sions, from documents about quality assurance

procedures in the Space domain. Our system

leverages state of the art auto-regressive mod-

els like T5 and BART to generate questions,

and a RoBERTa model to extract answers for

such questions, thus verifying their suitability.

1 Introduction

The complexity, cost, and risk of space missions

involving public or private investment and even

human lives make quality management a critical

requirement to guarantee their success. The Eu-

ropean Space Agency (ESA) makes a continuous

effort to train their staff in quality procedures and

standards. Trainees are evaluated to determine the

effectiveness of the training sessions, with quizzes

as one of the main tools used in such evaluations.

We present SpaceQQuiz (Space Quality Quiz),

a system designed to help trainers to generate

quizzes from documents describing quality proce-

dures. Such documents cover topics like Anomaly

and Problem Identiﬁcation, Reporting and Reso-

lution or Conﬁguration Management, and include

stakeholder responsibilities, activities, performance

indicators and outputs, among others.

To design SpaceQQuiz we use state-of-the-art

models based on transformers for Question Gener-

ation (QG) and Question Answering (QA). Since

we could not ﬁnd specialized models for the space

or quality management domains, we reuse models

already pre-trained on general-purpose document

corpora and ﬁne-tuned on SQuAD1.

The Stanford Question Answering Dataset

https://

rajpurkar.github.io/SQuAD-explorer/

Figure 1: SpaceQQuizz - Proposed architecture.

2 Quiz generation system

Figure 1shows the high-level architecture of Space-

QQuiz

. A question generation model is run on

each passage extracted from the document. The

generated questions and the corresponding pas-

sages are fed to a question answering model that

extracts the answer from the passage. Only ques-

tions with answers are included in the candidate

list that then is reﬁned by the trainer to generate the

quiz.

The process starts when the trainer uploads a

quality procedure document. The system extracts

the text from the PDF document using Apache

PDFBox

and uses regular expressions to identify

sections, subsections and paragraphs while remov-

ing non relevant text such as headers and footers.

The trainer is presented with a list of candidate sec-

tions so that she can choose the most interesting

ones for the quiz.

2.1 Question generation

To generate the questions we use a T5 model (Raf-

fel et al.,2020) and a BART model (Lewis et al.,

2020) ﬁne-tuned on question generation. We use

two models

in order to increase the number and

variety of questions for each text passage. Both T5

and BART have excelled in sequence generation

Demo:

https://esatde.expertcustomers.

ai/SpaceQQuiz/

user/pass demoINLG/demoINLG2022!

3Apache PDFBox https://pdfbox.apache.org

Models withdrawn from HuggingFace by their au-

thors. Description available at

https://github.com/

patil-suraj/question_generation

arXiv:2210.03427v2 [cs.CL] 4 Nov 2022

文档加载中……请稍候！
如果长时间未打开，您也可以点击刷新试试。

下载文档到电脑，查找使用更方便

10 玖币 0人已下载

立即下载

摘要：

GeneratingQuizzestoSupportTrainingonQualityManagementandAssuranceinSpaceScienceandEngineeringAndresGarcia-Silva,CristianBerrio,JoseManuelGomez-PerezExpert.ai/Madrid,Spainagarcia@expert.ai,cberrio@expert.ai,jmgomez@expert.aiJoseAntonioMartinez-Heras,PatrickFleithSolenix/Darmstadt,Germanyjose.martinez...

展开>> 收起<<

Generating Quizzes to Support Training on Quality Management and Assurance in Space Science and Engineering Andres Garcia-Silva Cristian Berrio Jose Manuel Gomez-Perez.pdf

共3页,预览1页

还剩页未读，继续阅读

声明：本站为文档C2C交易模式，即用户上传的文档直接被用户下载，本站只是中间服务平台，本站所有文档下载所得的收益归上传人(含作者)所有。玖贝云文库仅提供信息存储空间，仅对用户上传内容的表现方式做保护处理，对上载内容本身不做任何修改或编辑。若文档所含内容侵犯了您的版权或隐私，请立即通知玖贝云文库，我们立即给予删除！

Generating Quizzes to Support Training on Quality Management and Assurance in Space Science and Engineering Andres Garcia-Silva Cristian Berrio Jose Manuel Gomez-Perez

相关推荐

开通VIP享超值会员特权

作者详情

相关内容

热门标签

举报选择: