SBIR Phase I: READ! Toolkit, A Knowledge-Based Expert System to Develop Beginning Literacy Curricula in Endangered or Underserved Languages

SBIR 第一阶段：阅读！

• 批准号: 1842833
• 负责人: Catherine Callow-Heusser
• 金额: $ 22.5万
• 依托单位: BETTERMENT LABS, L.L.C.
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2019
• 资助国家: 美国
• 起止时间: 2019-01-01 至 2019-09-30
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=1842833&HistoricalAwards=false
• 关键词: SBIR; Phase; READ; Toolkit; Knowledge

This SBIR Phase I project will develop and test a computer-based expert system that will guide development of beginning literacy curricula in any phonics-based language. The system will produce beginning literacy curricula mobile apps for phones and tablets and will include beginning reading, writing, vocabulary-building, and assessment activities. The expert system will (a) minimize cost, time and expertise required to develop early literacy curricula, (b) produce curricula that could increase early reading fluency and bilingualism, which is related to children's improved ability to manage their thoughts, behaviors and emotions, (c) provide effective curricula to learn languages other than English at home or school, (d) revitalize endangered or underserved languages while increasing connections to culture and heritage, which will improve long-term academic, health and economic outcomes for indigenous children or children who speak a language other than English at home, and increase their participation in STEM careers, and (e) contribute to research on bilingualism's impacts on academic, social, emotional, and cultural outcomes. The system will be licensed by underserved language communities, tribes, individuals, and organizations interested in improving literacy, with revenue produced by the mobile app products developed using the expert system shared between curriculum developers and the expert system developer.This SBIR Phase I project will develop and test a knowledge-based expert system that integrates early literacy research; language analysis to provide evidence-based recommendations for developing decodable text for phonics-based early literacy curricula; evidence-based teaching and formative assessment strategies; curriculum development knowledge; user interface/user experience best-practices; natural language processing for pronunciation feedback; and database and technology skills to guide users who may not have the knowledge and skills in developing a beginning literacy curricula in any language. The best practices and methods based on the research will be incorporated by analyzing text to create lists of sounds/letters and words, their frequencies, and sounds and words that come before and after each sound/word. From these lists, the system will generate language-specific recommendations based on the rules for optimal sequencing of sounds and words for learning to decode in another language. While some evidence-based guidelines needed for decoding in English, which is less phonetically regular than many underserved languages, may not apply in a language that is more phonetically regular, most are expected to. These rules will guide recommendations from which users of the knowledge-based expert system can select sequences sounds, words, and phrases to write decodable, connected text, culturally relevant stories for beginning literacy curricula.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.

该 SBIR 第一阶段项目将开发和测试基于计算机的专家系统，该系统将指导任何基于语音的语言的初级识字课程的开发。该系统将为手机和平板电脑制作入门识字课程移动应用程序，并将包括入门阅读、写作、词汇建设和评估活动。专家系统将(a)最大限度地减少开发早期识字课程所需的成本、时间和专业知识，(b)制定可以提高早期阅读流利性和双语能力的课程，这与提高儿童管理自己的思想、行为和情绪的能力有关， (c) 提供有效的课程，以便在家或学校学习英语以外的语言， (d) 振兴濒危或服务不足的语言，同时加强与文化和遗产的联系，这将改善土著人的长期学术、健康和经济成果儿童或在家说英语以外语言的儿童，并增加他们对 STEM 职业的参与，以及 (e) 为双语对学术、社会、情感和文化成果的影响的研究做出贡献。该系统将由服务不足的语言社区、部落、个人和有兴趣提高读写能力的组织获得许可，并通过使用课程开发人员和专家系统开发人员之间共享的专家系统开发的移动应用程序产品产生收入。该 SBIR 第一阶段项目将开发和测试集成早期读写能力研究的基于知识的专家系统；语言分析，为基于语音的早期识字课程开发可解码文本提供循证建议；循证教学和形成性评估策略；课程开发知识；用户界面/用户体验最佳实践；用于发音反馈的自然语言处理；数据库和技术技能，以指导可能不具备开发任何语言的入门识字课程知识和技能的用户。基于研究的最佳实践和方法将通过分析文本来合并，以创建声音/字母和单词、它们的频率以及每个声音/单词之前和之后的声音和单词列表。从这些列表中，系统将根据声音和单词的最佳排序规则生成特定于语言的建议，以便学习用另一种语言进行解码。虽然英语解码所需的一些基于证据的指南（英语比许多服务不足的语言在语音上不太规则）可能不适用于语音更规则的语言，但大多数指南预计会适用。这些规则将指导建议，基于知识的专家系统的用户可以从中选择序列声音、单词和短语来编写可解码的、相互关联的文本、文化相关的故事，以用于入门识字课程。该奖项反映了 NSF 的法定使命，并被认为是值得的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来获得支持。