3542 search results - page 326 / 709 » Phenomenal Data Mining: From Data to Phenomena |
150
click to vote
ACSC
2002
IEEE
15 years 11 months ago
2002
IEEE
We show how frequently occurring sequential patterns may be found from large datasets by first inducing a finite state automaton model describing the data, and then querying the m...
179
click to vote
SDM
2010
SIAM
15 years 8 months ago
2010
SIAM
We give a permutation approach to validation (estimation of out-sample error). One typical use of validation is model selection. We establish the legitimacy of the proposed permut...
185
click to vote
DMIN
2009
15 years 4 months ago
2009
Record linkage is the problem of identifying similar records across different data sources. The similarity between two records is defined based on domain-specific similarity functi...
185
click to vote
ICDM
2005
IEEE
16 years 10 days ago
2005
IEEE
In practice, learning from data is often hampered by the limited training examples. In this paper, as the size of training data varies, we empirically investigate several probabil...
205
click to vote
SDM
2011
SIAM
14 years 9 months ago
2011
SIAM
The biclustering, co-clustering, or subspace clustering problem involves simultaneously grouping the rows and columns of a data matrix to uncover biclusters or sub-matrices of the...