A Personalized Word Learning System Based on Deep Neural Network 一種基於深度神經網絡的個性化單詞學習系統

The present invention provides a personalized word learning system based on deep neural network. The system comprises a memory and a processor that is operative to perform the following steps: constructing a task profile based on input amount, a task profile based on neighborhood, a learner profile based on input amount, and a learner profile based on neighborhood; obtaining a representation for the task profile based on input amount, a representation for the task profile based on neighborhood, a representation for the learner profile based on input amount, and a representation for the learner profile based on neighborhood; and conducting score prediction. The system according to the present invention enables higher accuracy and improved diversity for recommendation results.
本發明公開了一種基於深度神經網絡的個性化單詞學習系統。系統包括處理器和存儲器，處理器可執行：構建基於投入量的任務畫像、基於鄰域的任務畫像、基於投入量的學習者畫像、基於鄰域的學習者畫像；得到基於投入量的任務畫像的特徵表示、基於鄰域的任務畫像的特徵表示、基於投入量的學習者畫像的特徵表示、和基於鄰域的學習者畫像的特徵表示；以及進行評分預測。根據本發明的系統使得推薦結果具有更高的準確率和多樣性。