Research application of the spam filtering algorithm on social media
Вантажиться...
Дата
2021
Автори
Назва журналу
Номер ISSN
Назва тому
Видавець
Анотація
Today there are a lot of different social networks and messengers,
which in today's society, especially in times of pandemic corona-virus have become
an integral part of our daily lives, including work activities. At the same time there is
a lot of unnecessary information coming to users every second, so the problem of
dealing with spam messages in social networks and messengers is now very relevant.
By spam we mean any messages that a person, or an entire company, considers
unnecessary in a particular text stream.
The project is devoted to solving the scientific and applied problem of
detecting spam messages in the text context of any social network or messenger using
various spam detection algorithms. Three algorithms were implemented and
investigated: an algorithm using naive Bayesian classifier, Support-vector machine
and multilayer perceptron neural network.
The main idea is to develop a spam detection algorithm that is fast and easy to
integrate in a messenger (social network). It is proposed to use the obtained solutions
for IT companies. The developed algorithm should recognize spam based on the
context of a particular firm and quickly remove or mark it. Since the spam detection
task is essentially the task of sorting messages into A and B classes, the developed
algorithm can be used not only for spam filtering but also, for example, for
monitoring chat rooms for the messages that are important to a particular employee of
the company.
Опис
Oliinyk V. Research application of the spam filtering algorithm on social media / V. Oliinyk ; sci. advisors A. Podorozhniak, N. Liubchenko // Black Sea Science 2021. Information Technology, Automation and Robotics : рroc. of the Intern. Competition of Student Scientific Works / Odessa Nat. Acad. of Food Technologies ; eds. B. Yegorov, M. Mardar, S. Kotlyk [et al.]. – Odessa : ONAFT, 2021. – P. 264-274 : fig. – Ref.: 15 tit.
Ключові слова
spam, social network, naive Bayesian classifier, Support-vector machine, multilayer perceptron neural network