Adaptive anti-spam filtering for agglutinative languages: a special case for Turkish

Özgür, Levent; Güngör, Tunga; Gürgen, Fikret

doi:10.1016/j.patrec.2004.07.004

Cited by 48 publications

(19 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Besides, the performance of ANN and DT classifiers clearly indicates that these methods are suitable for detecting spam e-mails. Moreover, the accuracy rates attained from the ANN method are superior to some other studies in the literature [3], [19]- [21]. Additionally, the extraction of other features, the evaluation of feature selection and classification methods on spam e-mail detection problem can be carried out as an interesting future work.…”

Section: Resultsmentioning

confidence: 93%

See 1 more Smart Citation

The investigation on the effect of feature vector dimension for spam email detection with a new framework

Ergin

Işık

2014

2014 9th Iberian Conference on Information Systems and Technologies (CISTI)

View full text Add to dashboard Cite

In this study, the effect of dimension for a feature vector on the classification of Turkish e-mails as spam or legitimate is investigated. Although hundreds of experimental studies are achieved especially for English, which is a nonagglutinative language, the number of efforts for Turkish, which is one of the most popular agglutinative languages in the world, is counted something on the fingers of one hand. Therefore, a solution is sought for Turkish spam e-mail problem taking the special characteristics of Turkish e-mails into consideration. The developed spam filtering framework has four components named as morphological decomposition, feature selection, training, and test phases. A fixed-prefix stemming approach is used to extract the features of an e-mail and then the Mutual Information (MI) method is carried out as the feature selection method. The Decision Tree (DT) and Artificial Neural Network (ANN) classifiers are employed and the recognition accuracies obtained from these methods are considerably satisfactory. The highest accuracy rates are 91.08% for ANN and 87.67% for DT methods when the dimensions of feature vectors are selected as 150x5) and (75x5), respectively.

show abstract

Section: Resultsmentioning

confidence: 93%

“…The existing techniques are classified into two categories: Static and dynamic methods [3]. Static methods are based on predefined addresses or header contents.…”

Section: Introductionmentioning

confidence: 99%

The investigation on the effect of feature vector dimension for spam email detection with a new framework

Ergin

Işık

2014

2014 9th Iberian Conference on Information Systems and Technologies (CISTI)

View full text Add to dashboard Cite

show abstract

“…While considering the reports like what is mentioned in [23] that indicates in 1998 about 10% of the incoming emails to a network has been spam, which according to [4] in 2003 has increased up to 24% and has been about 65% in July 2007 [5], it would be revealed that email, as one of the most popular internet services and a realization of the computer networks as a powerful communication medium [29] is confronted with the problem of spam, which causes money, time and bandwidth waste and is a true security issue. Symantec researches has turned up that in 2002, 63% of users have received more than 7 spam a day in contrast to the remaining 37% that have received up to 14.…”

Section: Introductionmentioning

confidence: 97%

email granulation based on distributed-interval type-2 fuzzy set methodologies

Tahayori

Visconti

Antoni

2007

2007 IEEE International Conference on Granular Computing (GRC 2007)

View full text Add to dashboard Cite

This paper proposes a dynamic model to classify incoming emails into five granules namely, spam, suspicious-spam, suspicious, suspicious-non-spam and non-spam, using distributed-interval type-2 fuzzy set methodologies. Toward this end we have used the concept of general intervals and applied the distributed intervals in interval type-2 fuzzy sets. The method confirmed that, the process of spam filtering is rather intellectual than statistical and moreover some different methods should be used complementarily to get to the higher precision.

show abstract

“…In this regard, classification of email is an important and growing task that concluded from the spam email's menace [1,3]. Several spam filtering techniques have been introduced by using machine learning approaches including SVM [4], GA [5], AIS [1,6,7], case based technique [8] and ANN [9] to combat spam messages.…”

Section: Introductionmentioning

confidence: 99%

Fuzzy Granular Classifier Approach for Spam Detection

Salehi

Selamat

Krejcar

et al. 2015

Computational Collective Intelligence

View full text Add to dashboard Cite

Abstract.Spam email problem is a major shortcoming of email technology for computer security. In this research, a granular classifier model is proposed to discover hyperboxes in the geometry of information granules for spam detection in three steps. In the first step, the k-means clustering algorithm is applied to find the seed_points to build the granular structure of the spam and non-spam patterns. Moreover, applying the interval analysis through the high homogeneity of the patterns captures the key part of the spam and non-spam classifiers' structure. In the second step, PSO algorithm is hybridized with the k-means to optimize the formalized information granules' performance. The proposed model is evaluated based on the accuracy, misclassification and coverage criteria. Experimental results reveal that the performance of our proposed model is increased through applying Particle Swarm Optimization and fuzzy set.

show abstract

Adaptive anti-spam filtering for agglutinative languages: a special case for Turkish

Cited by 48 publications

References 10 publications

The investigation on the effect of feature vector dimension for spam email detection with a new framework

The investigation on the effect of feature vector dimension for spam email detection with a new framework

email granulation based on distributed-interval type-2 fuzzy set methodologies

Fuzzy Granular Classifier Approach for Spam Detection

Contact Info

Product

Resources

About