Automated essay evaluation is a widely used practical solution for replacing time-consuming manual grading of student essays. Automated systems are used in combination with human graders in different high-stake assessments, where grading models are learned on essays datasets scored by different gra ders. Despite the definition of the standardized grading rules, human graders can unintentionally introduce subjective bias into scores. Consequently, a grading model has to learn from data that represents a noisy relationship between essay attributes and its grade. We propose an approach for partitioning a set of essays into subsets that represent similar graders, which uses an explanation methodology and clustering. The results confirm our assumption that learning from the ensemble of separated models can significantly improve the average prediction accuracy on artificial and real-world datasets.

Currently, there are many large, automatically constructed knowledge bases (KBs). One interesting task is learning from a knowledge base to generate new knowledge either in the form of inferred facts or rules that define regularities. One challenge for learning is that KBs are necessarily open world: we cannot assume anything about the truth values of tuples not included in the KB. When a KB only contains facts (i.e., true statements), which is typically the case, we lack negative examples, which are often needed by learning algorithms. To address this problem, we propose a novel score function for evaluating the quality of a first-order rule learned from a KB. Our metric attempts to include information about the tuples not in the KB when evaluating the quality of a potential rule. Empirically, we find that our metric results in more precise predictions than previous approaches.

Manual grading of students' essays is a time-consuming, labor-intensive and expensive activity for educational institutions. It is nevertheless necessary since essays are considered to be the most useful tool to assess learning outcomes. Automated essay evaluation represents a practical solution to this task, however, its main weakness is predominant focus on vocabulary and text syntax, and limited consideration of text semantics. In this work, we propose an extension to existing automated essay evaluation systems that incorporates additional semantic attributes. We design the novel attributes by transforming sequential parts of an essay into the semantic space and measuring changes between them to estimate coherence of the text. The resulting system (called SAGE - Semantic Automated Grader for Essays) achieves significantly higher grading accuracy compared with 8 other state-of-the-art automated essay evaluation systems.