NEYRON TARMOQLAR VA GENETIK XARITALAR ASOSIDA ELEKTRON POCHTA SPAM XABARLARINI ANIQLASH MODELI
Keywords:
spam, neyron tarmoq, elektron pochta, termin, genetik xarita, n-gramma, klassifikatsiyaAbstract
Ushbu maqolada spam elektron pochta xabarlarini aniqlash uchun genetik xaritalar va neyron tarmoqlar asosida model ishlab chiqish masalasi ko‘rib chiqilgan. Taklif etilgan yondashuv matnli xabarlarning raqamli kodlar ketma-ketligiga aylantirilishi, terminlar ajratilishi va keyinchalik ularning asosida spam yoki qonuniy xabar deb tasniflanishini nazarda tutadi. Enron dataset asosida o‘tkazilgan tajribalar modelning samaradorligini, ayniqsa qisqa n-grammalar (n = 1, 2) asosida yuqori aniqlikka erishishini ko‘rsatdi.
References
Metsis, A., Androutsopoulos, I., Paliouras, G. (2006). Spam Filtering with Naive Bayes – Which Naive Bayes? CEAS.
Sahami, M., Dumais, S., Heckerman, D., Horvitz, E. (1998). A Bayesian Approach to Filtering Junk E-Mail. AAAI Workshop.
Bruce Guenter Spam Dataset. http://untroubled.org/spam/
Georgios Paliouras et al. (2004). Machine Learning in Anti-Spam Filtering.
Xu, X., Wang, X., & Sun, J. (2019). Deep Learning Based Spam Filtering for Multilingual Emails. Springer.
Korpus: Enron Email Dataset, CMU (Carnegie Mellon University).
Ghosh, S., & Ghosh, S. (2016). "Spam detection using artificial neural network and support vector machine." International Journal of Computer Applications.
Huang, J., & Dun, J. (2008). "A distributed genetic algorithm for rule extraction from support vector machines." In 2008 International Conference on Computer Science and Software Engineering.
