Abstract: Various embodiments of the present disclosure disclose machine-learning based evaluation techniques for detecting feature bias. An evaluation framework is provided that utilizes new evaluation data structures for comprehensibly evaluating feature bias in machine learning models. The evaluation framework includes receiving evaluation dataset for a machine learning model that includes one or more different feature classes of an evaluation feature. The evaluation framework includes generating, using an evaluation function, at least two performance metrics for the machine learning model and generating a unitless dissimilarity metric for the evaluation feature based at least in part on the first performance metric, the second performance metric, and an average between the first and second performance metrics. In this way, the unitless dissimilarity metric is based on a variability associated with the evaluation feature.

Abstract: Various embodiments of the present disclosure disclose machine-learning based evaluation techniques for detecting feature bias. An evaluation framework is provided that utilizes new evaluation data structures for comprehensibly evaluating feature bias in machine learning models. The evaluation framework includes receiving evaluation dataset for a machine learning model that includes one or more different feature classes of an evaluation feature. The evaluation framework includes generating, using an evaluation function, at least two performance metrics for the machine learning model and generating a unitless dissimilarity metric for the evaluation feature based at least in part on the first performance metric, the second performance metric, and an average between the first and second performance metrics. In this way, the unitless dissimilarity metric is based on a variability associated with the evaluation feature.

Abstract: A method of computing relative rankings for multiple course trajectories may include accessing a data structure. The data structure may include a first node that may represent a first course and a plurality of nodes that may represent courses that are available after completing the first course. Each of the plurality of nodes may be associated with a weight, and the plurality of nodes may include a second node representing a second course. The method may also include receiving a first grade for a student for the first course and receiving a second grade for the student for the second course. The method may additionally include adjusting the weight associated with the second node using the first grade and the second grade. The method may further include providing a ranking of the plurality of courses using the weights of the plurality of courses.

Abstract: A method of computing relative rankings for multiple course trajectories may include accessing a data structure. The data structure may include a first node that may represent a first course and a plurality of nodes that may represent courses that are available after completing the first course. Each of the plurality of nodes may be associated with a weight, and the plurality of nodes may include a second node representing a second course. The method may also include receiving a first grade for a student for the first course and receiving a second grade for the student for the second course. The method may additionally include adjusting the weight associated with the second node using the first grade and the second grade. The method may further include providing a ranking of the plurality of courses using the weights of the plurality of courses.