Concept

Evaluation of binary classifiers

Related lectures (14)

Discusses the evaluation of binary classifiers, including recall, sensitivity, specificity, ROC curves, and performance measures.

Untitled

Generalized Linear Regression: Classification

Explores Generalized Linear Regression, Classification, confusion matrices, ROC curves, and noise in data.

Machine Learning in Philanthropy and Non-Profit

Delves into the application of machine learning in enhancing human rights documentation and advocacy for organizations like HURIDOCS.

Linear Models for Classification: Multi-Class Extensions

Covers linear models for multi-class classification, focusing on logistic regression and evaluation metrics.

Model Selection: ROC Curves, Regression Evaluation, Conclusion

Explores thresholding, ROC curves, regression evaluation, naive methods, and model selection in machine learning.

Linear Regression: Basics

Covers the basics of linear regression, binary and multi-class classification, and evaluation metrics.

Machine Learning Basics

Introduces machine learning basics, including data collection, model evaluation, and feature normalization.

Evaluation Protocols

Explores evaluation protocols in machine learning, including recall, precision, accuracy, and specificity, with real-world examples like COVID-19 testing.

Receiver-Operator Characteristics: ROC Curves

Explains ROC curves, Precision-Recall curve, RMSLE, and model validation.

Model Selection: Evaluation and Generalization

Explores model selection, evaluation, and generalization in machine learning, emphasizing unbiased performance estimation and the risks of over-learning.

Data Preprocessing: Handling Challenges

Delves into advanced data preprocessing techniques, covering categorical encoding, missing data handling, and unbalanced datasets, emphasizing performance metrics and classifier comparison.

Untitled

Support Vector Machines: SVMs

Explores Support Vector Machines, covering hard-margin, soft-margin, hinge loss, risks comparison, and the quadratic hinge loss.