Research-to-Practice for Peer-to-Peer Learning in Engineering Education using Ensemble Methods to Deploy a Lifecycle Design Roadmap

The full paper to address the research to practice category includes the implementation and investigation of learners’ performances and peer-to-peer learning environment strategies to observe the relationships and roles associated with the engineering design lifecycle process. The ability to demonstrate and build on project-based learning introduced how potential decisions are readily accessible and analyze from the engineering design performances of learners’ interaction and decision approach. The interaction includes overall learners’ success rates and cost-effectiveness to model techniques used to improve prediction performance for evaluation. The study considers failure and success rates to understand how cost-effectiveness can be a benefit with the adoption of misclassification rates (e.g., success-prone decisions). In recent studies, the challenge to predict performances has been credited to the success of proposed methods and learning techniques to handle domain-specific areas based on developments and the lifecycle design process. The development of a peer-to-peer learning environment created a high-level overview to explore the engineering design process and the requirements for advancement of the learners’ experience. The learners’ decision in peer-to-peer team environment can be handled differently regarding the areas and types due to the framework proposed using ensemble models. In this evaluation, the questions using a pre- and post-survey assessment tool can be referenced according to the target groups’ decisions. These cues amongst the learner community in engineering education can be integrated using ensemble of attributes as the foundation in the design practices. This study aims to explore the homogeneous ensemble models using multiple method types for predictive analytics to understand the concepts for prediction as a proposed method and framework (i.e., for the assessment of misclassification modeling). As the ensemble classifiers are examined in the research-to-practice for engineering design, the experimental conditions were validated comparably to demonstrate ways to improve prediction and performance. The overall relationship to advance project-based learning in engineering education highlights the variables of how a deployed lifecycle can be used for research-to-practice for peer-to-peer environments using methods in data modeling.