Implementation in C+CUDA of Multi-Label Text Categorizers
Lucas De Paula Veronese, Alberto Ferreira De Souza, Claudine Santos Badue, Elias Silva de Oliveira, Patrick Marques Ciarelli, Fabiano Alan Serafim Ferrari · 2008
In automated multi-label text categorization problems with large numbers of labels, the training databases are large, which may render the categorization time prohibitive for online systems. In this work, we evaluate the parallel implementation in C+CUDA of two multi-label text categorizers: the first is based on the k-Nearest Neighbors (k-NN) algorithm [1] and the second is based on Probabilistic Neural Networks (PNN) [2]. We implemented these algorithms in three different ways: sequential in C, parallel in C+CUDA, and parallel using the C+CUBLAS library. where N k is the number of neurons of the pattern layer associated to ck . The categories c k ranked above a threshold are predicted to the input document d x .