Abstract. Spam filtering is a text categorization task that has attracted significant attention due to the increasingly huge amounts of junk email on the Internet. While current best-practice systems use Naive Bayes filtering and other probabilistic methods, we propose using a statistical, but non-probabilistic classifier based on the Winnow algorithm. The feature space considered by most current methods is either limited in expressivity or imposes a large computational cost. We introduce orthogonal sparse bigrams (OSB) as a feature combination technique that overcomes both these weaknesses. By combining Winnow and OSB with refined preprocessing and tokenization techniques we are able to reach an accuracy of 99.68% on a difficult test corpus, compared to 98.88% previously reported by the CRM114 classifier on the same test corpus.
The visual ventral system needs to be better understood despite recent technological advancements. The primary visual cortex processes edge-based information and contrast patterns in the human brain, which are heavily studied. However, the nature of non-linearity involved in contrast processing still needs to be discovered. This study aims to determine the type of cells involved in contrast processing and the nature of non-linearity involved. Initially, the VOneNet computational model of V1, among other models of V1, was selected for its performance and interpretability. The model was used to verify the shift-invariance property of Complex Cells and classify a dataset of images labeled based on contrast using features similar to simple cells of V1. The results suggest simple cells with rectified linear activation are responsible for contrast processing. These findings have important implications for understanding the primate visual system and may inform the development of future models.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.