Support vector machines for spam categorization

By Harris Drucker, Donghui Wu, Vladimir Naumovich Vapnik. 1999

How researchers see this paper

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

In Natural Language Processing

Citations in the corpus

Towards the Science of Security and Privacy in Machine Learning (2016, background): “Classic examples of supervised learning tasks include: object recognition in images [4], machine translation [5], and spam ﬁltering [6].”
Security Evaluation of Pattern Classifiers under Attack (2014, background): “This kind of classifier has been considered by several authors [6], [28], [43], and it is included in several real spam filters.”
Security Evaluation of Support Vector Machines in Adversarial Environments (2014, neutral): “Support Vector Machines (SVMs) are among the most successful techniques that have been applied for this purpose [28, 54].”
SoK: Security and Privacy in Machine Learning (2018, background): “Classic examples of supervised learning tasks include: object recognition in images [6], machine translation [7], and spam filtering [8].”
Machine learning in automated text categorization (2001, neutral): “Similarly, an e-mail .lter might be trained to discard junk mail [Androutsopoulos et al. 2000; Drucker et al. 1999] and further classify nonjunk mail into topical categories of interest to the user.”
Comprehensible credit scoring models using rule extraction from support vector machines (2007, neutral): “Support Vector Machines are a state-of-the art data mining technique which have proven their performance in many applications [8], such as credit scoring [2], ﬁnancial time series prediction [13], spam categorization [9] and brain tumor classiﬁcation [17].”
Propensity score estimation: machine learning and classification methods as alternatives to logistic regression (2010, methodology): “SVMs have proven effective in practical classification tasks, such as spam detection [28] and the classification of cancers [29,30], but we are aware of only one instance in which an SVM has been used to estimate propensity scores [31].”
EVILCOHORT: Detecting Communities of Malicious Accounts on Online Services (2015, background): “Multiple papers dealt with detecting email content that is typical of spam by using machine learning [10, 25].”
The relationship between trust in AI and trustworthy machine learning technologies (2019, unknown): “translation [83], filtering spams [32], etc.”
Evaluation and Design of Robust Neural Network Defenses (2018, background): “[7, 94, 130, 74, 26] Most work in this space relies on one of two attacks: (a) inject good words [74] into spam to make it appear like legitimate email, or (b) modify the bad words [94] that make an email appear as spam so they are not recognized.”
A Survey on Machine Learning Techniques for Cyber Security in the Last Decade (2020, background): “Spams on mobile devices and mobile networks can be in the form of spam calls, text, and video messages [104]–[107].”
On Attacking Statistical Spam Filters (2004, methodology): “Other notable learning algorithms applied to spam (cid:12)ltering include boosting [2] and support vector machines [10, 3].”
Machine-Learning-Based Disease Diagnosis: A Comprehensive Review (2021, methodology): “SVM was introduced by Vapnik in the late twentieth century [17].”
Spam Detection Using Feature Selection and Parameters Optimization (2010, unknown): “B.3) in detail.”