DIGI405-26S1 — Texts, Discourses and Data

David Ewing  |  University of Canterbury  |  2026

Can a text classifier reliably reproduce human-assigned sentiment labels — and what does its performance reveal about label quality? Assignment 2 applies Logistic Regression and Decision Tree classifiers to the Cardiff NLP tweet_eval sentiment dataset (~45,000 tweets: negative, neutral, positive), evaluating six feature types across 3-class and binary task framings.