Abstract
Despite the great success of graph-based transductive learning methods, most of them have serious problems in scalability and robustness. In this chapter, we propose an efficient and robust graph-based transductive classification method, called minimum tree cut (MTC), which is suitable for large scale data. Motivated from the sparse representation of graph, we approximate a graph by a spanning tree. Exploiting the simple structure, we develop a linear-time algorithm to label the tree such that the cut size of the tree is minimized. This significantly improves graph-based methods, which typically have a polynomial time complexity. Moreover, we theoretically and empirically show that the performance of MTC is robust to the graph construction, overcoming another big problem of traditional graph-based methods. Extensive experiments on public data sets and applications on text extraction fromimages demonstrate our method’s advantages in aspect of accuracy, speed, and robustness.
Original language | English |
---|---|
Title of host publication | Semi-Supervised Learning |
Subtitle of host publication | Background, Applications and Future Directions |
Publisher | Nova Science Publishers, Inc. |
Pages | 67-98 |
Number of pages | 32 |
ISBN (Electronic) | 9781536135572 |
ISBN (Print) | 9781536135565 |
Publication status | Published - 1 Jan 2018 |
Keywords
- Graph-based learning method
- Semi-supervised learning
- Transductive learning