TY - GEN
T1 - Unsupervised cross-domain learning by interaction information co-clustering
AU - Ando, Shin
AU - Suzuki, Einoshin
PY - 2008/12/1
Y1 - 2008/12/1
N2 - In real-world data mining applications, one often has access to multiple datasets that are relevant to the task at hand. However, learning from such datasets can be difficult as they are often drawn from different domains, i.e., not identically distributed or differ in class or feature sets. In this paper, we consider the problem of learning the class structures of related domains in an unsupervised manner. Its setting generalizes that of information filtering and novelty detection applications which addresses both known and unknown classes. We propose a co-clustering framework for estimating and adapting the class structures of two related domains, enabling the analyses of shared and unique classes. We define an objective function using interaction information to take account of the divergence between the corresponding clusters of respective domains. We present an iterative algorithm which alternates object and feature clustering and converges to a local minimum of the objective function. We present empirical results using text benchmarks, comparing the proposed algorithm and combinations of conventional approaches in problems of partitioning documents and detecting unknown topics.
AB - In real-world data mining applications, one often has access to multiple datasets that are relevant to the task at hand. However, learning from such datasets can be difficult as they are often drawn from different domains, i.e., not identically distributed or differ in class or feature sets. In this paper, we consider the problem of learning the class structures of related domains in an unsupervised manner. Its setting generalizes that of information filtering and novelty detection applications which addresses both known and unknown classes. We propose a co-clustering framework for estimating and adapting the class structures of two related domains, enabling the analyses of shared and unique classes. We define an objective function using interaction information to take account of the divergence between the corresponding clusters of respective domains. We present an iterative algorithm which alternates object and feature clustering and converges to a local minimum of the objective function. We present empirical results using text benchmarks, comparing the proposed algorithm and combinations of conventional approaches in problems of partitioning documents and detecting unknown topics.
UR - http://www.scopus.com/inward/record.url?scp=67049100148&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=67049100148&partnerID=8YFLogxK
U2 - 10.1109/ICDM.2008.92
DO - 10.1109/ICDM.2008.92
M3 - Conference contribution
AN - SCOPUS:67049100148
SN - 9780769535029
T3 - Proceedings - IEEE International Conference on Data Mining, ICDM
SP - 13
EP - 22
BT - Proceedings - 8th IEEE International Conference on Data Mining, ICDM 2008
T2 - 8th IEEE International Conference on Data Mining, ICDM 2008
Y2 - 15 December 2008 through 19 December 2008
ER -