In the paper we derive and discuss a wide class of algorithms for 3D Super-symmetric nonnegative Tensor Factorization (SNTF) or nonnegative symmetric PARAFAC, and as a special case...
Andrzej Cichocki, Marko Jankovic, Rafal Zdunek, Sh...
This paper proposes a novel hierarchical learning strategy to deal with the data sparseness problem in relation extraction by modeling the commonality among related classes. For e...
Increasingly scientific discoveries are driven by analyses of massively distributed bulk data. This has led to the proliferation of high-end mass storage systems, storage area clu...
Leximancer is a software system for performing conceptual analysis of text data in a largely language independent manner. The system is modelled on Content Analysis and provides u...
For the task of recognizing dialogue acts, we are applying the Transformation-Based Learning (TBL) machine learning algorithm. To circumvent a sparse data problem, we extract valu...