Machine Learning Skills To K–12

Abstract

The promise of data-driven methodology in various computer disciplines has been shown by the many real-world implementations of methods based on Machine Learning (ML) over the last couple of decades. ML is finding its way into the computer curriculum in higher education, and an increasing number of organizations are introducing it into computer education in grades K–12. Researching how agency and intuition grow in these situations is critical as computational learning becomes increasingly common in K–12 computer instruction. However, knowing the difficulties associated with teaching algorithmic learning through grades K–12 presents an even more difficult barrier for computer education research, given the difficulties educators and schools now face in integrating traditional learning. This article describes the prospects in data mining schooling for grades K–12. These developments include adjustments to philosophy, technology, and practice. The research addresses several distinctions that K–12 computer educators should consider while addressing this problem and places the current results into the broader context of computing education. The research focuses on crucial elements of the fundamental change needed to properly incorporate ML into more comprehensive K–12 computer courses. Giving up on the idea that rule-based, "traditional" programming is necessary for next-generation computational thinking is a crucial first step.