Abstract | ||
---|---|---|
It is common in classification methods to first place data in a vector space and then learn decision boundaries. We propose reversing that process: for fixed decision boundaries, we "learn" the location of the data. This way we (i) do not need a metric (or even stronger structure) - pairwise dissimilarities suffice; and additionally (ii) produce low-dimensional embeddings that can be analyzed visually. We achieve this by combining an entropy-based embedding method with an entropy-based version of semi-supervised logistic regression. We present results for clustering and semi-supervised classification. |
Year | DOI | Venue |
---|---|---|
2005 | 10.1145/1102351.1102485 | ICML |
Keywords | Field | DocType |
semi-supervised logistic regression,large margin non-linear embedding,decision boundary,semi-supervised classification,fixed decision boundary,entropy-based version,classification method,place data,entropy-based embedding method,low-dimensional embeddings,pairwise dissimilarities suffice,vector space | Pairwise comparison,Vector space,Embedding,Nonlinear system,Pattern recognition,Computer science,Reversing,Artificial intelligence,Cluster analysis,Logistic regression,Machine learning | Conference |
ISBN | Citations | PageRank |
1-59593-180-5 | 5 | 0.88 |
References | Authors | |
7 | 2 |
Name | Order | Citations | PageRank |
---|---|---|---|
Alexander Zien | 1 | 1255 | 146.93 |
j quinonero candela | 2 | 345 | 32.89 |