Computational Linguistics | Ch 06: Machine Learning, Text Classification, and Evaluation | Video

Video corrections and clarifications: 03:20 — The graph’s bracket is the raw agreement difference (.80 − .515 = .285), not kappa. Kappa = .285/.485 ≈ .59; the slide’s formula and final result are correct (textbook pp.107–108). The original video is unchanged. These notes follow sampled visual checks against the textbook. Narration was not reviewed. Chapter 6: Machine Learning, Text Classification, and Evaluation Foundations of Computational Linguistics: For EFL University Students Video lesson • Printed textbook pages 101–117 A classifier learns to assign labels from examples, but a high score needs careful interpretation. This chapter connects features and annotation with Naive Bayes, evaluation measures, and the analysis of errors on new data. Central question: How can computers learn to sort texts from examples, and how can we judge whether they have learned a useful language pattern? Learning objectives 1. Explain how a supervised classifier learns from features and labels, and why training, development, and test data must be kept apart. 2. Apply Naive Bayes by hand to classify a short text and compare the result with a majority-class baseline. 3. Calculate Cohen’s kappa, precision, recall, and F-score from small count tables. 4. Analyze classification errors by linguistic category and evaluate whether a reported score supports a claim about new data. Five core topics 6.1 Features, Labels, and Training, Development, and Test Data (p. 102) Represent examples with features and labels while keeping training, development, and test sets separate. 6.2 Naive Bayes Classification and Sentiment Analysis (p. 104) Trace a Naive Bayes decision for a short text and compare it with a simple baseline. 6.3 Annotation, Gold Standards, and Inter-Annotator Agreement (p. 106) Examine how labels are created and use agreement measures to assess consistency among annotators. 6.4 Measuring Performance: Accuracy, Precision, Recall, and F-Score (p. 108) Calculate accuracy, precision, recall, and F-score from counts and select measures suited to the task. 6.5 Error Analysis, Overfitting, and Generalization (p. 110) Group errors by linguistic cause and assess whether performance extends beyond the training data. Study with the textbook Read pages 101–117, review this chapter companion, and attempt the activities before checking the explained answers. Refer to the book for exact examples, calculations, diagrams, notation, and qualifications. Free textbook: https://zenodo.org/records/23095075 Textbook DOI: https://doi.org/10.5281/zenodo.23095075 Chapter resources: https://sites.google.com/view/els4efl... ELS4EFL series website: https://sites.google.com/view/els4efl... YouTube channel:    / @elsforefl   About the textbook Series: English Language Studies for EFL University Students (ELS4EFL) Branch: Theoretical Linguistics Project Director: Prof. Munassir Alhamami First Edition | Version 1 | 2026 Citation Alhamami, M. (2026). Foundations of computational linguistics: For EFL university students (First Edition, Version 1). Zenodo. https://doi.org/10.5281/zenodo.23095075 This is an AI-generated study companion created with Gemini Notebook (NotebookLM) from the textbook. Consult the book for complete teaching, practice activities, and explained answers. Original textbook content is licensed under CC BY 4.0 except where otherwise stated. Material credited to other sources remains subject to its stated terms. Chapter and topic headings follow the textbook; the study explanations in this description are supplementary adaptations. https://creativecommons.org/licenses/... #ComputationalLinguistics #NLP #EFL