A New Approach to Arabic Spam Tweet Detection in Twitter Using Machine Learning Algorithms
Küçük Resim Yok
Tarih
2022
Yazarlar
Dergi Başlığı
Dergi ISSN
Cilt Başlığı
Yayıncı
American Institute of Physics Inc.
Erişim Hakkı
info:eu-repo/semantics/openAccess
Özet
Artificial intelligence is involved in all aspects of life and with the development of social media, the methods used in sending spam have evolved and it becomes difficult to detect and control it manually. Therefore, due to the need to use artificial intelligence in language processing and the detection of propaganda messages by classifying posts and tweets that carry a disturbing message on Twitter in this study, We propose a new system to detect spam in Twitter for the Arabic language, based on the trend of hashtags, using artificial intelligence algorithms and extracting properties. It contains three data collection stages, image and text characteristics extraction, and classification stage based on combination of CNN algorithm and SVM algorithm. In addition to this search, annoying text and images, in addition to combining two classification algorithms. The classification accuracy is 98% and the results indicate the best compared to the rest of the models when tested after the testing process on four samples of the data collected from Twitter. © 2022 American Institute of Physics Inc.. All rights reserved.
Açıklama
1st International Conference on Advanced Research in Pure and Applied Science, ICARPAS 2021 -- 24 March 2021 through 25 March 2021 -- Al-Samawah -- 183732
Anahtar Kelimeler
Kaynak
AIP Conference Proceedings
WoS Q Değeri
Scopus Q Değeri
N/A
Cilt
2398
