An LSTM-Based Plagiarism Detection via Attention Mechanism and a Population-Based Approach for Pre-training Parameters with Imbalanced Classes

Plagiarism is one of the leading problems in academic and industrial environments, which its goal to find similar items a typical document or source code. This paper proposes an architecture based on Long Short-Term Memory (LSTM) attention mechanism called LSTM-AM-ABC boosted by population-based approach for parameter initialization. Gradient-based optimization algorithms such as back-propagation (BP) are widely used literature learning process LSTM, mechanism, feed-forward neural network, while they suffer from some getting stuck local optima. To tackle this problem, metaheuristic (PBMH) can be used. end, employs PBMH algorithm, artificial bee colony (ABC), moderate problem. Our proposed algorithm initial values model all simultaneously. In other words, ABC finds promising point starting BP algorithm. For evaluation, we compare our with both conventional methods. The results clearly show that method provide competitive performance.