soft voting
[/sɔft ˈvoʊtɪŋ/]
noun
votação suave
1. A machine learning ensemble method where classifiers assign probability scores or confidence values to predictions, which are then combined through averaging or weighted aggregation rather than using hard class labels
Soft voting improves model accuracy by leveraging the probabilistic outputs of multiple classifiers.
A votação suave melhora a precisão do modelo ao aproveitar as saídas probabilísticas de múltiplos classificadores.
2. In ensemble learning, a voting mechanism that uses confidence scores or probabilities from individual models to reach a final prediction
The ensemble used soft voting with a weighted average of the neural network and random forest predictions.
O conjunto usou votação suave com uma média ponderada das predições da rede neural e da floresta aleatória.
Soft voting is primarily used in technical and academic contexts within both Brazil and the USA. It is a term specific to machine learning and data science communities. While less commonly known among the general population, it has become standard terminology in Brazilian universities and tech companies engaged in AI development and research.
Look up more words on Fala2Me
The free English-Portuguese dictionary with real Brazilian accents, NYC slang, conjugator and more
Open Fala2Me →