Single-participant structural connectivity matrices lead to greater accuracy in classification of participants than function in autism in MRI.
Matthew LemingSimon Baron-CohenJohn SucklingPublished in: CoRR (2020)
Keyphrases
- classification accuracy
- predictive accuracy
- accuracy rate
- feature extraction
- pattern recognition
- generalization ability
- roc curve
- high accuracy
- pattern classification
- machine learning
- classification rate
- magnetic resonance imaging
- feature space
- decision trees
- roc analysis
- classification algorithm
- correct classification
- training set
- text classification
- model selection
- multiple classifier systems
- class labels
- image classification
- supervised learning
- significantly higher
- fold cross validation
- random selection
- classification models
- support vector machine svm
- training samples
- least squares
- learning algorithm