Analyzing the Effect of Document Representation on Machine Learning Approaches in Multi-Class e-Mail Filtering.
Helmut BergerMichael DittenbachDieter MerklPublished in: Web Intelligence (2006)
Keyphrases
- multi class
- machine learning approaches
- document representation
- bag of words
- document collections
- document clustering
- machine learning
- vector space model
- pairwise
- support vector machine
- vector space
- text mining
- machine learning methods
- language model
- machine learning algorithms
- web documents
- data fusion
- text documents
- feature selection
- spam filtering
- semantic information
- data mining methods
- image classification
- text data
- action recognition
- computer vision
- dimensionality reduction
- information retrieval
- clustering algorithm
- neural network
- image representation
- semantic relations
- information retrieval systems
- background knowledge
- learning algorithm
- loss function