Controllable Multi-Interest Framework for Recommendation

最新推荐文章于 2022-11-28 10:56:12 发布

Adam婷

最新推荐文章于 2022-11-28 10:56:12 发布

阅读量3.4k

点赞数

分类专栏：算法机器学习论文研读

本文链接：https://blog.csdn.net/weixin_41697507/article/details/107547801

版权

机器学习同时被 3 个专栏收录

161 篇文章 8 订阅 ¥19.90 ¥99.00

订阅专栏

超级会员免费看

算法

161 篇文章 4 订阅

订阅专栏

论文研读

38 篇文章 1 订阅

订阅专栏

本文提出了一种新颖的可控多兴趣推荐框架ComiRec，用于解决用户在一段时间内的多兴趣捕获和推荐问题。该框架通过多兴趣模块从用户行为序列中捕获多个兴趣，并使用聚合模块结合不同兴趣的项目进行整体推荐。实验表明，ComiRec在亚马逊和淘宝两个真实世界数据集上显著优于现有模型，并已在阿里巴巴分布式云平台成功部署。

摘要由CSDN通过智能技术生成

Controllable Multi-Interest Framework for Recommendation

在这里插入图片描述

ABSTRACT

Recently, neural networks have been widely used in e-commerce recommender systems, owing to the rapid development of deep learning. We formalize the recommender system as a sequential recommendation problem, intending to predict the next items that the
user might be interacted with. Recent works usually give an overall embedding from a user’s behavior sequence. However, a unified user embedding cannot reflect the user’s multiple interests during a period. In this paper, we propose

了解本专栏