Sentence alignment for English to Turkish translations by using deep learning system

Natural language processing has gained importance in recent years with the development of artificial intelligence and linguistics. In "Turkish to English - English to Turkish" one-to-one translations by translation companies or in website translations, the order of sentences is changed or randomized, requiring manpower to rearrange the sentences. In the field of natural language processing, there are various algorithms for sentence alignment. In this study, deep learning, a sub-branch of machine learning, is used to find effective matching methods for English-Turkish translated texts. With a deep learning method developed, datasets were collected from various sources from the internet and then sifted and organized through certain filters. After comparing the success of existing models, a more successful model was obtained by training the collected dataset and a comparison was made with cosine similarity.