Narrative Modeling with StoryQ: Integrating Mathematics, Language Arts, and Computing to Create Pathways to Artificial Intelligence Careers

使用 StoryQ 进行叙事建模：整合数学、语言艺术和计算，打造人工智能职业之路

• 批准号: 1949110
• 负责人: Jie Chao
• 金额: $ 149.69万
• 依托单位: Concord Consortium
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2020
• 资助国家: 美国
• 起止时间: 2020-06-01 至 2024-05-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=1949110&HistoricalAwards=false
• 关键词: Narrative; Modeling; StoryQ; Integrating; Mathematics

The future workforce is being drastically reshaped by artificial intelligence (AI) technologies. The advancement in AI theories, algorithms and practices has not only created great demands for AI scientists, engineers, technicians and entrepreneurs, but also reformulated the nature of work in almost all industries. Importantly today's students must gain a fundamental understanding of AI in order to be prepared to enter the workforce of the future. This project will design a 12-lesson high school curriculum called StoryQ and associated teaching guides that will provide students with firsthand experience on how narrative modeling, one of the oldest fields in artificial intelligence, can be developed while working on their language arts writing projects. By integrating age-appropriate mathematics, language arts, and computing concepts, researchers will leverage advanced data exploration and text mining technologies, and employ research-based pedagogical approaches to help high school students learn basic concepts in machine learning and artificial intelligence. This project is funded by the Innovative Technology Experiences for Students and Teachers (ITEST) program, which supports projects that build understandings of practices, program elements, contexts and processes contributing to increasing students' knowledge and interest in science, technology, engineering, and mathematics (STEM) and information and communication technology (ICT) careers. Led by a multidisciplinary team of learning technology and data experts, machine learning researchers, and experts in the integration of narrative modeling and mathematics learning, the project will focus on helping students envision their future careers as powered by artificial intelligence. The researchers will create and test StoryQ, a web-based text mining and narrative modeling platform, and develop, implement, and test narrative modeling with StoryQ curriculum. Students will learn to design, build, test, and iteratively improve machine learning models of narratives sourced both from student and teacher selected literature and from students’ own writings. Beginning with core narrative concepts, students will engage in development cycles that lead them to explore their own writing, annotate models by hand, observe a trained text mining model at work, become familiar with error analysis, and ultimately build their own AI model and evaluation process. The project broadens participation among youth from underrepresented and underserved populations by recruiting participants from two Massachusetts school districts with ethnically and economically diverse populations. To create broadly inclusive learning experiences, students from diverse backgrounds will write narratives to express their cultures and personalities as part of the learning activities. Research questions include (1) How can learning environments be designed to help students understand core AI concepts including the structures in unstructured data and the roles of human insight in the development of AI technologies? and (2) How can learning environments be designed to help students develop awareness and interest in careers that are centered on text mining practices or broadly powered by AI technologies? The project will use a design-based research design. The research team will carry out class observations and in-depth analysis of observational data, and draw design principles for building learning environments that cultivate future STEM and ICT workforce. The project’s success will be evaluated by a group of external evaluators who are experts in diversity and inclusion in AI, computing education, mathematics education, and language arts and literacy education. The outcomes of the project include the resulting web-delivered classroom-ready AI curriculum modules, a teaching guide and teacher resources for high schools, which will be disseminated to teachers and professional development groups. The StoryQ technology will be freely distributed.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

人工智能(AI)技术正在极大地重塑未来的劳动力。人工智能理论、算法和实践的进步不仅对人工智能科学家、工程师、技术人员和企业家产生了巨大的需求，而且重塑了几乎所有行业的工作性质。重要的是，今天的学生必须对人工智能有一个基本的了解，才能为进入未来的劳动力做好准备。这个项目将设计一个名为StoryQ的12节课的高中课程和相关的教学指南，为学生提供第一手经验，让他们在从事语言艺术写作项目的同时，如何开发叙事建模--人工智能中最古老的领域之一。通过整合适合年龄的数学、语言艺术和计算概念，研究人员将利用先进的数据探索和文本挖掘技术，并采用基于研究的教学方法来帮助高中生学习机器学习和人工智能的基本概念。该项目由学生和教师创新技术体验计划(ITEST)资助，该计划支持的项目旨在加深对实践、计划要素、背景和过程的理解，有助于提高学生对科学、技术、工程和数学(STEM)以及信息和通信技术(ICT)职业的知识和兴趣。该项目由学习技术和数据专家、机器学习研究人员以及叙事建模和数学学习相结合的专家组成的多学科团队领导，该项目将专注于帮助学生想象他们未来的职业生涯是由人工智能提供动力的。研究人员将创建和测试StoryQ，一个基于网络的文本挖掘和叙事建模平台，并开发、实施和测试StoryQ课程的叙事建模。学生将学习设计、构建、测试和迭代改进叙事的机器学习模型，这些模型来自学生和老师选择的文学作品以及学生自己的写作。从核心叙事概念开始，学生将参与开发周期，带领他们探索自己的写作，手工注释模型，观察工作中经过训练的文本挖掘模型，熟悉错误分析，最终构建自己的人工智能模型和评估流程。该项目通过招募来自两个族裔和经济人口不同的马萨诸塞州学区的参与者，扩大了代表人数不足和服务不足人口中的青年的参与。为了创造广泛包容的学习体验，来自不同背景的学生将撰写叙事来表达他们的文化和个性，作为学习活动的一部分。研究问题包括(1)如何设计学习环境来帮助学生理解核心人工智能概念，包括非结构化数据中的结构和人类洞察力在人工智能技术发展中的作用？以及(2)如何设计学习环境来帮助学生提高对以文本挖掘实践为中心或广泛由人工智能技术支持的职业的认识和兴趣？该项目将采用基于设计的研究设计。研究小组将对观察数据进行课堂观察和深入分析，并为构建培养未来STEM和ICT劳动力的学习环境得出设计原则。该项目的成功将由一组外部评估者进行评估，他们是人工智能、计算机教育、数学教育以及语言艺术和扫盲教育中的多样性和包容性方面的专家。该项目的成果包括由此产生的网络交付的准备好课堂的人工智能课程模块、教学指南和高中教师资源，这些资源将传播给教师和专业发展团体。StoryQ技术将免费分发。该奖项反映了NSF的法定使命，并通过使用基金会的智力优势和更广泛的影响审查标准进行评估，被认为值得支持。

项目成果

How learners produce data from text in classifying clickbait

学习者如何从文本中生成数据来对标题诱饵进行分类

• DOI: 10.1111/test.12339
• 发表时间: 2023
• 期刊: Teaching Statistics
• 影响因子: 0.8
• 作者: Horton, Nicholas J.;Chao, Jie;Palmer, Phebe;Finzer, William
• 通讯作者: Finzer, William

FanfictionNLP: A Text Processing Pipeline for Fanfiction

• DOI: 10.18653/v1/2021.nuse-1.2
• 发表时间: 2021-06
• 期刊: Proceedings of the Third Workshop on Narrative Understanding
• 作者: Michael Miller Yoder;Sopan Khosla;Qinlan Shen;Aakanksha Naik;Huiming Jin;Hariharan Muralidharan;C. Rosé

Spam Four Ways: Making Sense of Text Data

垃圾邮件四种方式：理解文本数据

• DOI: 10.1080/09332480.2022.2066414
• 发表时间: 2022
• 期刊: CHANCE
• 作者: Horton, Nicholas J.;Chao, Jie;Finzer, William;Palmer, Phebe
• 通讯作者: Palmer, Phebe

High school students’ data modeling practices and processes: From modeling unstructured data to evaluating automated decisions

高中生数据建模实践和流程：从非结构化数据建模到评估自动化决策

• DOI: 10.1080/17439884.2023.2189735
• 发表时间: 2023
• 期刊: Media and Technology
• 作者: Jiang, Shiyan;Tang, Hengtao;Tatar, Cansu;Rosé, Carolyn P.;Chao, Jie
• 通讯作者: Chao, Jie

Modeling Unstructured Data: Teachers as Learners and Designers of Technology-enhanced Artificial Intelligence Curriculum

非结构化数据建模：教师作为技术增强型人工智能课程的学习者和设计者

• 发表时间: 2021
• 期刊: International Society of the Learning Sciences Annual Meeting 2021
• 作者: Tatar, C.;Yoder, M. M.;Coven, M.;Wiedemann, K.;Chao, J.;Finzer, W.;Jiang, S.;Rosé, C. P.
• 通讯作者: Rosé, C. P.