SUMMARY AND RECOMMENDATION SYSTEM USING DEEP LEARNING

摘要

Disclosed are a paper summary and recommendation system using deep learning and a method thereof. The method according to the present invention comprises the steps of: classifying papers and deriving topics; extracting topics and papers containing words similar to search words; and training representative papers by category by using a sequence-to-sequence (Seq2Seq) model and generating a summary for a given paper. Therefore, paper summary and recommendation are possible through continuous data training by applying deep learning techniques, thereby improving the quality of a paper summary and recommendation service.