A Neural Network Algorithm for Extracting Bilingual Lexica
Hong-Seok Kwon, Hyeong-Won Seo, Min-Ah Cheon, Jae‐Hoon Kim · Advanced science and technology letters · 2014
We propose a neural network algorithm for extracting bilingual lexica, which is based on the well-known Perceptron algorithm. In this paper, we modify the Perceptron algorithm in order to learn with unlabeled training data for bilingual lexicon extraction. To show the feasibility of the algorithm, we also present a novel approach on bilingual lexicon extraction with the proposed neural network algorithm. The empirical results have shown that our proposed method is significantly improving the performances of our model obtained with the modified Perceptron algorithm.