APPARATUS AND METHOD FOR RECOMMENDING COLLABORATIVE FILTERING BASED ON LEARNABLE-TIME ORDINARY DIFFERENTIAL EQUATION

To provide an apparatus and a method for recommending collaborative filtering based on learnable-time ordinary differential equation.SOLUTION: An apparatus for recommending collaborative filtering based on learnable-time ordinary differential equation according to the present invention includes: an initial-embedding generation module that generates first and second initial embeddings; a dual co-evolutionary ordinary differential equation module that includes first and second ordinary differential equation processing members which repeat, for a specific time, a process of receiving first and second previous embeddings, respectively, and outputting first and second subsequent embeddings, respectively, through mutual influence as a learning time elapses, the first and second previous embeddings initially corresponding to the first and second initial embeddings, respectively; and a final-embedding generation module that generates a first final embedding by cumulatively summing the first previous and subsequent embeddings, and generates a second final embedding by cumulatively summing the second previous and subsequent embeddings.SELECTED DRAWING: Figure 3 【課題】学習時間常微分方程式基盤の協業フィルタリング推薦装置及び方法を提供する。【解決手段】本発明の学習時間常微分方程式基盤の協業フィルタリング推薦装置は、第１及び第２の初期の埋め込みを生成する初期の埋め込み生成モジュールと、学習時間が経過するにつれて、初期に前記第１及び第２の初期の埋め込みに各々該当する第１及び第２の前の埋め込みの各々の入力を受けて相互影響を通じて第１及び第２の次の埋め込みを各々出力する過程を特定時間の間繰り返す第１及び第２の常微分方程式処理部材を含む二重共振常微分方程式モジュールと、第１の前の埋め込み及び第１の前の次の埋め込みを累積合算して第１の最終の埋め込みを生成し、前記第２の前の埋め込み及び前記第２の次の埋め込みを累積合算して第２の最終の埋め込みを生成する最終の埋め込み生成モジュールと、を備える。【選択図】図３