Abstract | ||
---|---|---|
We propose a kernel-based online semi-supervised algorithm that is applicable for large scale learning tasks. In particular, we use a multi-view learning framework and a co-agreement strategy to take into account unlabelled data and to improve classification performance of the algorithm. Unlike the standard online methods our algorithm is naturally applicable to many real-world situations where data is available in multiple representations. In addition our online algorithm allows learning non-linear relations in the data via kernel functions, that are efficiently embedded into the formulation of the algorithm. We test performance of the algorithm on several large-scale LIBSVM and UCI benchmark datasets and demonstrate improved performance in comparison to standard online learning methods. Last but not least, we make a Python implementation of our algorithm available for download (Available at https://github.com/laurensvdwiel/KeCo). |
Year | DOI | Venue |
---|---|---|
2015 | 10.1007/978-3-319-24282-8_26 | DISCOVERY SCIENCE, DS 2015 |
Keywords | Field | DocType |
Kernel, Non-linear, Online, Large-scale, Semi-supervised, Co-agreement, Multi-view, Classification | Online learning,Kernel (linear algebra),Data mining,Online machine learning,Online algorithm,Computer science,Algorithm,Artificial intelligence,Python (programming language),Machine learning,Kernel (statistics),Distributed computing | Conference |
Volume | ISSN | Citations |
9356 | 0302-9743 | 0 |
PageRank | References | Authors |
0.34 | 6 | 3 |
Name | Order | Citations | PageRank |
---|---|---|---|
Laurens Wiel | 1 | 0 | 0.34 |
Tom Heskes | 2 | 1519 | 198.44 |
Evgeni Levin | 3 | 0 | 0.34 |