Issue |
JNWPU
Volume 39, Number 5, October 2021
|
|
---|---|---|
Page(s) | 1070 - 1076 | |
DOI | https://doi.org/10.1051/jnwpu/20213951070 | |
Published online | 14 December 2021 |
Recommended method study based on incorporating complex network ripple net
基于联合复杂网络Cn-RippleNet模型的推荐方法
Key Laboratory of China's Ethnic Languages and Information Technology of Ministry of Education, Northwest Minzu University, Lanzhou 730030, China
Received:
12
April
2021
The RippleNet network models user preferences and is well applied in the recommended system. But Ripplenet didn't take into account the weight of entities in the knowledge graph, resulting in the inaccurate recommendation results. A RippleNet model incorporating the influence of the complex network nodes is proposed. After constructing the complex networks based on the knowledge maps, the maximum subnet model is extracted, the influence of the nodes in the map network is calculated, and the weight of the nodes is added to the RippleNet model as an entity. The experimental results showed that the present method increased the AUC and ACC values of RippleNet to 92.0% and 84.6%, made up for the problem that no entity influence was considered in the RippleNet network, and made the recommended results more in line with users' expectations.
摘要
RippleNet对用户偏好传播进行建模,并运用在推荐系统中,取得了良好的效果,但RippleNet没有考虑知识图谱中的实体权重,导致推荐的实体不够精确。提出了一种加入复杂网络节点影响力的Cn-RippleNet模型,在构建基于知识图谱的复杂网络之后,抽取其最大子网模型,计算图谱网络中节点影响力,并将其作为实体的权重添加至图谱实体中,最终计算出推荐结果。实验结果表明,该方法将RippleNet的AUC和ACC的值提高到了93.0%和85.6%,弥补了RippleNet没有考虑图谱实体影响力的问题,使推荐结果更符合用户预期。
Key words: knowledge graph / recommended system / complex network / node influence / rippleNet
关键字 : 知识图谱 / 推荐系统 / 复杂网络 / 节点影响力 / RippleNet
© 2021 Journal of Northwestern Polytechnical University. All rights reserved.
This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.
Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.
Initial download of the metrics may take a while.