On Term Weighting for Spam SMS Filtering

Due to rapid development of the technology, the usage of mobile telephones and short message services (SMS) have become widespread. Thus, the number of spam SMS messages has dramatically increased and the significance of identifying and filtering of suchlike messages raised. Moreover, since they have also risk to steal users’ personal information; the problem of identifying and filtering of Spam SMS messages stays popular in terms of also information and data security. In this study, the classification performances of five different term weighting methods on three different datasets containing SMS messages categorized as Spam and legitimate are compared by using two classifiers for corresponding problem. The results obtained showed that reasonable weighting of SMS contents plays an important role in identifying of spam SMS messages. On the other hand, it can be expressed that real classification potential of term weighting schemes reflected betterly the with feature vectors created by using fifty and higher number of terms on especially Turkish and English SMS message datasets. In addition, it has been observed that value ranges of the classification results of obtained from term weighting methods on Turkish SMS message dataset is wider for than ones obtained in English SMS message datasets.

İstenmeyen SMS Filtrelemede Terim Ağırlıklandırma

Teknolojideki hızlı gelişmeler, mobil telefonların sayısını arttırmış ve kısa mesaj hizmetlerinin (SMS) kullanımını yaygın hale getirmiştir. Bu durum, istenmeyen SMS sayılarını da dramatik bir biçimde arttırmış ve bu tip mesajların belirlenmesi veya filtrelenmesinin önemini arttırmıştır. Ayrıca, kullanıcıların kişisel bilgilerini çalma riski de taşıyabilecekleri için, istenmeyen SMS’lerin filtrelenmesi problemi günümüzde bilgi ve veri güvenliği açısından da popülerliğini korumaktadır. Bu çalışmada, bu probleme yönelik olarak, istenmeyen ve meşru olarak iki sınıfa kategorilendirilmiş SMS mesajlarını içeren üç farklı SMS mesaj veri seti üzerinde 5 farklı popüler terim ağırlıklandırma yönteminin sınıflandırma performansıları iki popüler sınıflandırıcı yardımıyla kıyaslanmıştır. Elde edilen sonuçlar; istenmeyen SMS belirleme performansında; SMS içeriklerinin makul bir biçimde ağırlıklandırılmasının önemli bir rol oynadığını göstermiştir. Diğer taraftan, özellikle Türkçe ve İngilizce SMS mesaj verisetleri üzerinde terim ağırlıklandırma şemalarının sahip oldukları potansiyel sınıflandırma performanslarınının elli ve üzeri terim kullanılarak yapılan deneylerde daha iyi yansıtılabildiği ifade edilebilir. Ayrıca Türkçe SMS mesaj veri seti üzerinde terim ağırlıklandırma yöntemlerinden elde edilen sınıflandırma sonuçlarının değer aralıklarının, İngilizce SMS mesaj verisetlerinde elde edilenlere nazaran daha geniş olduğu da gözlenmiştir.

PDF

___

[1] H. Faris, I. Aljarah, and B. Al-Shboul, "A hybrid approach based on particle swarm optimization and random forests for e-mail spam filtering," in International Conference on Computational Collective Intelligence, 2016: Springer, pp. 498-508.

[2] R. Varghese and K. Dhanya, "Efficient feature set for spam Email filtering," in 2017 IEEE 7th International Advance Computing Conference (IACC), 2017: IEEE, pp. 732-737.

[3] M. Diale, T. Celik, and C. Van Der Walt, "Unsupervised feature learning for spam email filtering," Computers & Electrical Engineering, vol. 74, pp. 89-104, 2019.

[4] M. A. Shafi’I et al., "A review on mobile SMS spam filtering techniques," IEEE Access, vol. 5, pp. 15650-15666, 2017.

[5] K. O. Kawade and K. S. Oza, "Content-based SMS spam filtering using machine learning technique," International Journal of Computer Engineering and Applications, vol. 7, p. 4, 2018.

[6] T. H. Apandi and C. A. Sugianto, "Analisis Komparasi Machine Learning Pada Data Spam Sms," Jurnal TEDC, vol. 12, no. 1, pp. 58-62, 2019.

[7] S. J. Delany, M. Buckley, and D. Greene, "SMS spam filtering: Methods and data," Expert Systems with Applications, vol. 39, no. 10, pp. 9899-9908, 2012.

[8] J. M. Gómez Hidalgo, G. C. Bringas, E. P. Sánz, and F. C. García, "Content based SMS spam filtering," in Proceedings of the 2006 ACM Symposium on Document Engineering, 2006, pp. 107-114.

[9] G. V. Cormack, J. M. G. Hidalgo, and E. P. Sánz, "Feature engineering for mobile (SMS) spam filtering," in Proceedings of the 30th annual international ACM SIGIR Conference on Research and Development in Information Retrieval, 2007, pp. 871-872.

[10] T. Almeida, J. M. G. Hidalgo, and T. P. Silva, "Towards sms spam filtering: Results under a new dataset," International Journal of Information Security Science, vol. 2, no. 1, pp. 1-18, 2013.

[11] M. T. Nuruzzaman, C. Lee, and D. Choi, "Independent and personal SMS spam filtering," in 2011 IEEE 11th International Conference on Computer and Information Technology, 2011: IEEE, pp. 429-435.

[12] M. B. Junaid and M. Farooq, "Using evolutionary learning classifiers to do MobileSpam (SMS) filtering," in Proceedings of the 13th annual conference on Genetic and evolutionary computation, 2011, pp. 1795-1802.

[13] A. K. Uysal, S. Gunal, S. Ergin, and E. S. Gunal, "A novel framework for SMS spam filtering," in 2012 International Symposium on Innovations in Intelligent Systems and Applications, 2012: IEEE, pp. 1-4.

[14] J. W. Yoon, H. Kim, and J. H. Huh, "Hybrid spam filtering for mobile communication," Computers & Security, vol. 29, no. 4, pp. 446-459, 2010.

[15] H. Najadat, N. Abdulla, R. Abooraig, and S. Nawasrah, "Mobile sms spam filtering based on mixing classifiers," International Journal of Advanced Computing Research, vol. 1, pp. 1-7, 2014.

[16] H.-Y. Lee and S.-S. Kang, "Word Embedding Method of SMS Messages for Spam Message Filtering," in 2019 IEEE International Conference on Big Data and Smart Computing (BigComp), 2019: IEEE, pp. 1-4.

[17] A. K. Jain, S. K. Yadav, and N. Choudhary, "A Novel Approach to Detect Spam and Smishing SMS using Machine Learning Techniques," International Journal of E-Services and Mobile Applications (IJESMA), vol. 12, no. 1, pp. 21-38, 2020.

[18] A. K. Uysal, S. Gunal, S. Ergin, and E. S. Gunal, "The impact of feature extraction and selection on SMS spam filtering," Elektronika ir Elektrotechnika, vol. 19, no. 5, pp. 67-73, 2013.

[19] Y.-T. Chen and M. C. Chen, "Using chi-square statistics to measure similarities for text categorization," Expert systems with applications, vol. 38, no. 4, pp. 3085-3090, 2011.

[20] K. Sparck Jones, "A Statistical Interpretation of Term Specificity and Its Application in Retrieval," Journal of Documentation, vol. 28, no. 1, pp. 11-21, 2004

[21] Y. Liu, H. T. Loh, and A. Sun, "Imbalanced text classification: A term weighting approach," Expert Systems with Applications, vol. 36, no. 1, pp. 690-701, 2009

[22] A. K. Uysal and S. Gunal, "A novel probabilistic feature selection method for text classification," Knowledge-Based Systems, vol. 36, pp. 226-235, 2012

[23] M. Lan, C. L. Tan, J. Su, and Y. Lu, "Supervised and traditional term weighting methods for automatic text categorization," IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 31, no. 4, pp. 721-735, 2009.

[24] K. Chen, Z. Zhang, J. Long, and H. Zhang, "Turning from TF-IDF to TF-IGM for term weighting in text classification," Expert Systems with Applications, vol. 66, pp. 245-260, 2016

[25] T. Dogan and A. K. Uysal, "Improved inverse gravity moment term weighting for text classification," Expert Systems with Applications, vol. 130, pp. 45-59, 2019.

[26] C.-C. Chang and C.-J. Lin, "LIBSVM: a library for support vector machines," ACM Transactions on Intelligent Systems and Technology (TIST), vol. 2, no. 3, p. 27, 2011.