This paper analyses the Employment Scam Aegean Dataset and compares various machine learning algorithms including Logistic Regression, Decision Tree, Random Forest, XGBoost, K-Nearest Neighbor, Naïve Bayes and Support Vector Classifier on the task of fake job classification. The paper also proposes two self-attention enhanced Gated Recurrent Unit networks, one with vanilla RNN architecture and other with Bidirectional architecture, for classifying the fake job from real ones. The proposed framework uses Gated Recurrent Units with multi-head self-attention mechanism to enhance the long term retention within the network. In comparison to the other algorithms, the two GRU models proposed in this paper are able to obtain better result.

Keywords : Fake Job Classification; Text Classification; Gated Recurrent Unit; Recurrent Neural Networks.