The Cost of Dichotomizing Continuous Labels for Binary Classification Problems: Deriving a Bayesian-Optimal Classifier
Year: 2015
Type: article
Authors: Soroosh Mariooryad, Carlos Busso
Institutions: Google (United States), The University of Texas at Dallas
Language: English
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: closed
Awards: RI: Small: Collaborative Research: Exploring Audiovisual Emotion Perception using Data-Driven Computational Modeling, CAREER: Advanced Knowledge Extraction of Affective Behaviors During Natural Human Interaction, EAGER: Exploring the Use of Synthetic Speech as Reference Model to Detect Salient Emotional Segments in Speech
OpenAlex ID: W2527527138