Sparse distributed representations for words with thresholded independent component analysis

Reference:

Jaakko J. Väyrynen, Lasse Lindqvist, and Timo Honkela. Sparse distributed representations for words with thresholded independent component analysis. In Jennie Si and Ron Sun, editors, Proceedings of the International Joint Conference on Neural Networks (IJCNN 2007), pages 1031–1036. IEEE, August 2007.

Abstract:

We show that independent component analysis (ICA) can be used to find distributed representations for words that can be further processed by thresholding to produce sparse representations. The applicability of the thresholded ICA representation is compared to singular value decomposition (SVD) in a multiple choice vocabulary task with three data sets.

Suggested BibTeX entry:

@inproceedings{Vayrynen07IJCNN2007,
    author = {Jaakko J. V{\"{a}}yrynen and Lasse Lindqvist and Timo Honkela},
    booktitle = {Proceedings of the International Joint Conference on Neural Networks (IJCNN 2007)},
    editor = {Jennie Si and Ron Sun},
    month = {August},
    pages = {1031--1036},
    publisher = {IEEE},
    title = {Sparse distributed representations for words with thresholded independent component analysis},
    year = {2007},
}