The impact of different training sets on medical documents classification.
Roberto GattaMauro VallatiBerardino De BariMahmut OzsahinPublished in: AI-AM/NetMed@ECAI (2014)
Keyphrases
- training set
- classification accuracy
- document classification
- supervised learning
- classification algorithm
- information retrieval
- svm classifier
- feature space
- pattern recognition
- support vector
- automatic classification
- training samples
- text documents
- feature vectors
- random selection
- support vector machine
- image classification
- classification method
- medical records
- automatic categorization
- free text
- data sets
- pre classified
- active learning
- feature selection
- decision trees
- training data
- keywords
- text mining
- support vector machine svm
- document collections
- class labels
- class distribution
- xml documents
- weak classifiers
- text classifiers
- retrieval systems
- naive bayes
- user queries
- benchmark datasets