A Benchmark Collection for Mapping Program Educational Objectives to ABET Student Outcomes: Accreditation

This research aims to present a collection of dataset, which represents the mapping of program education objectives to the ABET student outcomes. The dataset has been collected by the authors from 32 self-study reports from Engineering programs accredited by ABET, which are available online. The paper presents the constraints under which, the dataset was produced, because its understanding plays a vital role in the usage of this collection in future researches. To illustrate the properties and usefulness of the collection, the dataset has been cleansed, preprocessed, some features have been selected, then it has been benchmarked using nine of the widely used supervised multiclass classification techniques (Binary Relevance, Label Powerset, Classifier Chains, Pruned Sets, Random k-label sets, Ensemble of Classifier Chains, Ensemble of Pruned Sets, Multi-Label k Nearest Neighbors and Back-Propagation Multi-Label Learning). The techniques have been compared to each other using five well-known measurements (Accuracy, Hamming Loss, Micro-F, Macro-F, and Macro-F). The Ensemble of Classifier Chains and Ensemble of Pruned Sets have achieved encouraging performance compared to the other experimented multi-label classification methods. The Classifier Chains method has shown the worst performance. In general, promising results have been achieved. New research directions and baseline experimental results for future studies in educational data mining in general and in accreditation in specific have been provided.