Improving Occupational Data Quality: Automated Coding and Corpus Review Using Pretrained Models

• In Event: AERA Roundtable Session Thursday 1:45 pm Four Seasons Ballroom 4
  In Roundtable Session: Machine Learning Techniques II: Advancing Coding/Essay Scoring and Estimation of Treatment Effects (Table 2)

Abstract

This research investigates the utilization of pre-trained models for automated occupational coding and corpus review to advance the precision of occupational data in educational research. Conventional manual coding is both resource-intensive and susceptible to errors. By harnessing machine learning, specifically pre-trained models such as ERNIE 3.0, this study tackles issues like data sparsity and coding inaccuracies, thereby enhancing efficiency and accuracy in occupational data analysis. The paper showcases the superior performance of the pre-trained model compared to manual and traditional machine learning approaches, presenting a promising solution for resource-constrained environments.

Authors

• Wen Li, Beijing Normal University

• Fei Yang, Beijing Normal University

• Sheng Zhang, Beijing Normal University