CAJ | 학술논문

目前常用的个性化推荐系统模型通常是基于协同过滤或者是基于内容的，也有部分基于关联规则的。这些算法没有考虑事务间的顺序，然而在很多应用中这样的顺序很重要。文章提出了一种简易的基于序列模式的推荐模型，并且考虑到大规模数据的处理，结合了 MapReduce 编程模型。这种简易的推荐模型可以用来辅助通常的个性化推荐系统。
목전상용적개성화추천계통모형통상시기우협동과려혹자시기우내용적，야유부분기우관련규칙적。저사산법몰유고필사무간적순서，연이재흔다응용중저양적순서흔중요。문장제출료일충간역적기우서렬모식적추천모형，병차고필도대규모수거적처리，결합료 MapReduce 편정모형。저충간역적추천모형가이용래보조통상적개성화추천계통。
The most commonly used model of personalized recommendation system is usually based on collaborative filtering or content-based , but also partly based on association rules . The algorithm does not consider the sequence between transactions , but the sequence is important in many applications . This paper presents a simple model based on sequential patterns for recommenda-tion system , and taking into account the large-scale data processing , combined with the MapReduce programming model . This sim-ple model of recommendation can be used to assist the usual personalized recommendation system .