Exploring Data Augmentation and Resampling Strategies for Transformer-Based Models to Address Class Imbalance in AI Scoring of Scientific Explanations in NGSS Classroom

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Djagba, Prudence, Haudek, Kevin, Franovic, Clare G. C., Kaldaras, Leonora
Format: Preprint
Veröffentlicht: 2026
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866917427082690560
author Djagba, Prudence
Haudek, Kevin
Franovic, Clare G. C.
Kaldaras, Leonora
author_facet Djagba, Prudence
Haudek, Kevin
Franovic, Clare G. C.
Kaldaras, Leonora
contents Automated scoring of students' scientific explanations offers the potential for immediate, accurate feedback, yet class imbalance in rubric categories particularly those capturing advanced reasoning remains a challenge. This study investigates augmentation strategies to improve transformer-based text classification of student responses to a physical science assessment based on an NGSS-aligned learning progression. The dataset consists of 1,466 high school responses scored on 11 binary-coded analytic categories. This rubric identifies six important components including scientific ideas needed for a complete explanation along with five common incomplete or inaccurate ideas. Using SciBERT as a baseline, we applied fine-tuning and test these augmentation strategies: (1) GPT-4--generated synthetic responses, (2) EASE, a word-level extraction and filtering approach, and (3) ALP (Augmentation using Lexicalized Probabilistic context-free grammar) phrase-level extraction. While fine-tuning SciBERT improved recall over baseline, augmentation substantially enhanced performance, with GPT data boosting both precision and recall, and ALP achieving perfect precision, recall, and F1 scores across most severe imbalanced categories (5,6,7 and 9). Across all rubric categories EASE augmentation substantially increased alignment with human scoring for both scientific ideas (Categories 1--6) and inaccurate ideas (Categories 7--11). We compared different augmentation strategies to a traditional oversampling method (SMOTE) in an effort to avoid overfitting and retain novice-level data critical for learning progression alignment. Findings demonstrate that targeted augmentation can address severe imbalance while preserving conceptual coverage, offering a scalable solution for automated learning progression-aligned scoring in science education.
format Preprint
id arxiv_https___arxiv_org_abs_2604_19754
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Exploring Data Augmentation and Resampling Strategies for Transformer-Based Models to Address Class Imbalance in AI Scoring of Scientific Explanations in NGSS Classroom
Djagba, Prudence
Haudek, Kevin
Franovic, Clare G. C.
Kaldaras, Leonora
Artificial Intelligence
Machine Learning
LLM STEM education, Analytic rubric, imbalance
Automated scoring of students' scientific explanations offers the potential for immediate, accurate feedback, yet class imbalance in rubric categories particularly those capturing advanced reasoning remains a challenge. This study investigates augmentation strategies to improve transformer-based text classification of student responses to a physical science assessment based on an NGSS-aligned learning progression. The dataset consists of 1,466 high school responses scored on 11 binary-coded analytic categories. This rubric identifies six important components including scientific ideas needed for a complete explanation along with five common incomplete or inaccurate ideas. Using SciBERT as a baseline, we applied fine-tuning and test these augmentation strategies: (1) GPT-4--generated synthetic responses, (2) EASE, a word-level extraction and filtering approach, and (3) ALP (Augmentation using Lexicalized Probabilistic context-free grammar) phrase-level extraction. While fine-tuning SciBERT improved recall over baseline, augmentation substantially enhanced performance, with GPT data boosting both precision and recall, and ALP achieving perfect precision, recall, and F1 scores across most severe imbalanced categories (5,6,7 and 9). Across all rubric categories EASE augmentation substantially increased alignment with human scoring for both scientific ideas (Categories 1--6) and inaccurate ideas (Categories 7--11). We compared different augmentation strategies to a traditional oversampling method (SMOTE) in an effort to avoid overfitting and retain novice-level data critical for learning progression alignment. Findings demonstrate that targeted augmentation can address severe imbalance while preserving conceptual coverage, offering a scalable solution for automated learning progression-aligned scoring in science education.
title Exploring Data Augmentation and Resampling Strategies for Transformer-Based Models to Address Class Imbalance in AI Scoring of Scientific Explanations in NGSS Classroom
topic Artificial Intelligence
Machine Learning
LLM STEM education, Analytic rubric, imbalance
url https://arxiv.org/abs/2604.19754