Segmenting Hashtags using Automatically Created Training Data.
Arda ÇelebiArzucan ÖzgürPublished in: LREC (2016)
Keyphrases
- automatically created
- training data
- automatically generated
- wordnet
- shallow semantic
- data sets
- training set
- decision trees
- classification accuracy
- supervised learning
- test data
- topic hierarchy
- learning algorithm
- training dataset
- training samples
- training examples
- test set
- prior knowledge
- domain knowledge
- training process
- microblog posts
- labeled data
- class labels
- feature extraction
- search engine
- machine learning
- topic models
- natural language processing
- support vector machine
- generalization error
- keywords
- training instances
- topic specific
- extracting features
- multimedia
- learned from training data
- artificial intelligence