Abstract. Data stream values are often associated with multiple aspects. For example, each value observed at a given time-stamp from environmental sensors may have an associated ty...
Jimeng Sun, Charalampos E. Tsourakakis, Evan Hoke,...
Markov statistical methods may make it possible to develop an unsupervised learning process that can automatically identify genomic structure in prokaryotes in a comprehensive way...
There is a considerable body of work on sequence mining of Web Log Data We are using One Pass frequent Episode discovery (or FED) algorithm, takes a different approach than the tr...
In this paper, we propose Partition min-Hash (PmH), a novel hashing scheme for discovering partial duplicate images from a large database. Unlike the standard min-Hash algorithm th...
We report performance evaluation of our automatic feature discovery method on the publicly available Gisette dataset: a set of 29 features discovered by our method ranks 129 among...