Using jWebMiner 2.0 to Improve Music Classification Performance by Combining Different Types of Features Mined from the Web.
Gabriel VigliensoniCory McKayIchiro FujinagaPublished in: ISMIR (2010)
Keyphrases
- classification accuracy
- feature vectors
- feature set
- feature extraction
- classification method
- feature space
- classification process
- classification models
- specific features
- image features
- svm classifier
- extracted features
- pattern recognition
- website
- musical instrument
- image classification
- class labels
- co occurrence
- feature analysis
- search engine
- discriminative features
- extracting features
- genre classification
- machine learning
- benchmark datasets
- web applications
- support vector machine
- low level
- classification algorithm
- textual features
- feature subset
- multiple types
- association rules
- gender classification
- digital music
- data mining
- decision trees
- music retrieval
- textural features
- support vector
- music information retrieval
- end users
- feature representation
- high dimensionality
- classification rules
- false positives