CAJ | 학술논문

本文针对电子商务中的商品推荐问题，利用交易数据构建基于商品的复杂网络，提出一种基于复杂网络社团发现的商品推荐方法，并且应用数据集进行了实验分析。结果表明，本方法在某用户只有较少历史行为数据时，仍可对该用户进行推荐，而且推荐结果可以保证新颖度和覆盖度。另外，此方法可处理的数据规模适用于电子商务的商品推荐，并且计算复杂度较低。
본문침대전자상무중적상품추천문제，이용교역수거구건기우상품적복잡망락，제출일충기우복잡망락사단발현적상품추천방법，병차응용수거집진행료실험분석。결과표명，본방법재모용호지유교소역사행위수거시，잉가대해용호진행추천，이차추천결과가이보증신영도화복개도。령외，차방법가처리적수거규모괄용우전자상무적상품추천，병차계산복잡도교저。
A commodity recommendation methodis proposed based on community detection in complex networks which is built by the transaction data from E-commerce. Then a classical data set isanalyzed using the proposed method. Results show that this method can recommend commodities to the user even there is fewer behavior data about this user. And the recommendationsare novel and with high coverage. In addition, the commodities scale in E-commerce is suitable for this method to processwith low computational complexity.