Words versus Character n-Grams for Anti-Spam Filtering

By Ioannis Kanaris, K. Kanaris, Ioannis Houvardas, Efstathios Stamatatos. 2007

98 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Information Systems, Natural Language Processing

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

1. Yann LeCun
2. Junbo Jake Zhao
3. Yoshua Bengio
4. Quoc V. Le
5. Kyunghyun Cho
6. Walmir Matos Caminhas
7. Thiago S. Guzella
8. Luke S. Zettlemoyer

In Information Systems

1. Yann LeCun
2. Walmir Matos Caminhas
3. Thiago S. Guzella
4. Junbo Jake Zhao

In Natural Language Processing

1. Yann LeCun
2. Junbo Jake Zhao
3. Walmir Matos Caminhas
4. Thiago S. Guzella
5. Luke S. Zettlemoyer
6. Kyunghyun Cho
7. Hinrich Schütze
8. Quoc V. Le

Citations in the corpus

• Character-level Convolutional Networks for Text Classification (2015, methodology): “These include using character-level n-grams with linear classifiers [15], and incorporating character-level features to ConvNets [28] [29].”
• A review of machine learning approaches to Spam filtering (2009, methodology): “Kanaris et al. (2007) used character n-grams, when n is prespecified and a variable scheme, to train a linear SVM using a binary or frequency representation, employing the information gain for feature selection.”