Spam detection by using word-vector learning algorithm in online social networks

Social media spam is one of the most important problems that professionals have to deal with in social networks on the internet. To solve this problem, the researchers have presented some solutions, mostly based on a number of different methods that take into account learning. The means and techniques used at the current time have achieved a good ratio of accuracy based on the so-called methods of blacklisting in order to determine undesirable activities in relation to sending and receiving an e-mail on social networks based on the conclusions obtained from previous experiments and studies. However, methods that rely on automated learning are not capable of detecting spam activities in proportion to real scenarios. We have seen that blacklist methods are not able to meet the disparities that we can see in activities related to the transmission of such messages, because manually checking Unique Resources Locators (URLs) is a time-consuming task. In this study, we present a deep learning method for spam detection in Twitter. For this purpose, the Word2Vec based on representation was first trained. We then used binary classification methods to distinguish between the spam and non-spam tweets. The empirical results conducted on tweets prove that the proposed methods outperform the classical approaches. Keywords: Spam Tweets; Spam Detection; classification; deep learning; Word2Vec algorithm; Social Network.