SBIR Phase II: Bilingual Literacy Assessment and Skill Tracker

Award ID: 2112298

The broader impact/commercial potential of this Small Business Innovation Research (SBIR) Phase II project is related to the underserved, emergent bilingual population in US school systems. In spite of their growing size (approximately 20 percent of the total US population), emergent bilinguals still do not have adequate support to succeed in academic settings and are often victims of a subtractive bilingualism process that favors the acquisition of English at the expense of the home language. The lack of adequate bilingual resources has led to a disproportionate misplacement of emergent bilinguals in special education as well as an increased drop-out rates in bilingual populations, in general, and Latino communities in particular. To address these problems, a recommendation system powered by machine-learning algorithms that will generate a personalized bundle of fun, bilingual literacy activities linked to a fully-automated bilingual assessment is proposed. The unique characteristic of this solution is that the personalized activities are generated automatically based on assessment results and activity performance. The goal of the activities is to develop the skills identified by the assessment as needing improvement, both in English and in the home language of the child. This Small Business Innovation Research Phase II project addresses the educational needs of the growing, school-age, emergent bilingual population (i.e. children in the school system whose home language is not English or only English), as well as the loss of cultural diversity and linguistic acumen for society as a whole. The main research objectives involve: a) designing game-like bilingual activities that can get the young childs attention, b) developing age-appropriate and culturally-sensitive content for the bilingual activities, c) refining the speech recognizer to work well with children's speech in both Spanish and English, and d) implementing the latest trends in data-analysis and machine-learning technologies to develop a recommendation system that proposes personalized activities tailored to each childs needs based on assessment results and activity performance. The research and development plan aims at developing an innovative product that stands out from existing solutions. The key differentiator features are: 1) its focus on bilingual children and its adaptability in accepting bilingual answers for the activities, 2) its capacity to help children develop speaking skills due to the integration of automatic measures of speech performance, and 3) its recommendation engine that suggests personalized bilingual activities for each student tailored to their specific needs. 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.

NSF Program Director: Rajesh Mehta