Extending Explainable Ensemble Trees (E2Tree) to regression contexts

Ensemble methods such as random forests have transformed the landscape of supervised learning, offering highly accurate prediction through the aggregation of multiple weak learners. However, despite their effectiveness, these methods often lack transparency, impeding users' comprehension of how RF models arrive at their predictions. Explainable ensemble trees (E2Tree) is a novel methodology for explaining random forests, that provides a graphical representation of the relationship between response variables and predictors. A striking characteristic of E2Tree is that it not only accounts for the effects of predictor variables on the response but also accounts for associations between the predictor variables through the computation and use of dissimilarity measures. The E2Tree methodology was initially proposed for use in classification tasks. In this paper, we extend the methodology to encompass regression contexts. To demonstrate the explanatory power of the proposed algorithm, we illustrate its use on real-world datasets.

Decoding Knowledge Claims: The Evaluation of Scientific Publication Contributions through Semantic Analysis

The surge in scientific publications challenges the use of publication counts as a measure of scientific progress, requiring alternative metrics that emphasize the quality and novelty of scientific contributions rather than sheer quantity. This paper proposes the use of Relaxed Word Mover's Distance (RWMD), a semantic text similarity measure, to evaluate the novelty of scientific papers. We hypothesize that RWMD can more effectively gauge the growth of scientific knowledge. To test such an assumption, we apply RWMD to evaluate seminal papers, with Hirsch's H-Index paper as a primary case study. We compare RWMD results across three groups: 1) H-Index-related papers, 2) scientometric studies, and 3) unrelated papers, aiming to discern redundant literature and hype from genuine innovations. Findings suggest that emphasizing knowledge claims offers a deeper insight into scientific contributions, marking RWMD as a promising alternative method to traditional citation metrics, thus better tracking significant scientific breakthroughs.