Isograph: Neighbourhood Graph Construction Based on Geodesic Distance for Semi-supervised Learning.
Marjan GhazvininejadMostafa MahdiehHamid R. RabieeParisa Khanipour RoshanMohammad Hossein RohbanPublished in: ICDM (2011)
Keyphrases
- geodesic distance
- semi supervised learning
- graph construction
- semi supervised
- manifold learning
- graph based semi supervised learning
- euclidean space
- shortest path
- euclidean distance
- neighborhood graph
- unlabeled data
- labeled data
- label propagation
- semi supervised classification
- distance transform
- unsupervised learning
- supervised learning
- training data
- machine learning
- active learning
- geometric structure
- distance metric
- text categorization
- gaussian mixture
- co training
- k nearest neighbor
- semi supervised learning methods
- labeled and unlabeled data
- data mining
- edge weights
- computer vision
- feature extraction
- shape analysis
- neural network
- object recognition
- image classification
- data analysis
- data points