Data science and machine learning should not only be research areas for scientists and researchers but should also be accessible and understandable to the general audience. Enabling students to understand the details behind the technology will support them to become aware consumers and encourage them to become active participants. In this paper, we present instructional materials developed for introducing middle school students to two key machine learning algorithms: decision trees and textit{k}-nearest neighbors. The materials were tested in an afterschool artificial intelligence program involving four students aged 12 to 13. A combination of hands-on activities, innovative technology, and intuitive examples facilitated student learning. With hand-drawn decision trees and penguin species classifications, students used the algorithms to solve problems and anticipate other possible applications. We present the technology used, curriculum materials developed, and classroom structure. Further, we discuss our findings from evaluating student work and interacting with students in the classroom. Following the guidelines from AI4K12 and introducing foundational machine learning algorithms, we hope to foster student interest in STEM fields.