A New English/Arabic Parallel Corpus for Phishing Emails

Phishing involves malicious activity whereby phishers, in the disguise of legitimate entities, obtain illegitimate access to victims’ personal and private information, usually through emails. Currently, phishing attacks threats are being handled effectively use latest email detection solutions. Most current systems assume be English, though other languages growing. In particular, Arabic is a widely used language therefore represents vulnerable target. However, there significant shortage corpora that can develop systems. This paper presents development new English-Arabic parallel corpus has been developed from anti-phishing share task text (IWSPA-AP 2018). The content was translated, had allotted 10 volunteers who university background were English experts. To evaluate effectiveness corpus, we models using Term Frequency–Inverse Document Frequency (TF-IDF) Multilayer Perceptron 1258 emails have equal ratios experimental findings show accuracy reaches 96.82% for dataset 94.63% providing some assurance potential value developed.