前言
最近整理了AAAI2020会议中关于推荐系统的论文,同时通过代码分析了下所接收论文的标题,发现了一些研究的热点以及趋势。
概述
通过对所接收的1590篇论文的标题进行分析,发现以下结论:
· 大部分的论文所用到的技术多为Neural Network(128)相关的;
· 大部分的文献聚焦在以下几个关键技术。比如Embedding(51),Attention(49),Adversarial(74),Reinforcement(49),Convolutional(42),Recurrent(16)等;
· 主要面向的研究任务有分类、回归、识别、追踪等,其中推荐的比重所占也不小。比如Classification(50),Regression(15),Prediction(39),Recognition(52),Tracking(20),Segmentation(28),Translation(32),Recommendation(21)等;
· 所研究的数据不仅关注准确性,关注指标更加多样化。比如Efficient(59),Robust(30),Dynamic(29),Adaptive(29),Hierarchical(26);
· 论文研究所用到的数据以图为主,视频、图像、文本比重相当。比如Graph(128),Video(35),Image(59),Heterogeneous(16),Text(35),Social(20)。
*其中括号里的数字表示出现次数。
推荐相关的文章
特此从1590篇论文中筛选出与推荐相关的27篇文章供大家提前阅读,提前领略牛人的最新想法。
· PEIA: Personality and Emotion Integrated Attentive Model for Music Recommendation on Social Media.
· Where to Go Next: Modeling Long-and Short‐Term User Preferences for Point-of‐Interest Recommendation.
· A Knowledge-Aware Attentional Reasoning Network for Recommendation.
· Enhancing Personalized Trip Recommendation with Attractive Routes.
· Efficient Heterogeneous Collaborative Filtering without Negative Sampling for Recommendation.
· An Attentional Recurrent Neural Network for Personalized Next Location Recommendation.
· Memory Augmented Graph Neural Networks for Sequential Recommendation.
· Leveraging Title-Abstract Attentive Semantics for Paper Recommendation.
· Diversified Interactive Recommendation with Implicit Feedback.
· Question-driven Purchasing Propensity Analysis for Recommendation.
· Sequential Recommendation with Relation-Aware Kernelized Self-Attention.
· Incremental Fairness in Two‐Sided Market Platforms: On Smoothly Updating Recommendations.
· Attention‐guide Walk Model in Heterogeneous Information Network for Multi-style Recommendation.
· Table2Analysis: Modeling and Recommendation of Common Analysis Patterns for Multi-Dimensional Data.
· Symmetric Metric Learning with Adaptive Margin for Recommendation.
· Multi-Feature Discrete Collaborative Filtering for Fast Cold-start Recommendation.
· Towards Comprehensive Recommender Systems: Time-Aware Unified Recommendations Based on Listwise Ranking of Implicit Cross-Network Data.
· Fast Adaptively Weighted Matrix Factorization for Recommendation with Implicit Feedback.
· Towards Hands‐free Visual Dialog Interactive Recommendation.
· Contextual-Bandit Based Personalized Recommendation with Time-Varying User Interests.
· Stochastically Robust Personalized Ranking for LSH Recommendation Retrieval.
· Revisiting Graph based Collaborative Filtering: A Linear Residual Graph Convolutional Network Approach.
· Multi-Component Graph Convolutional Collaborative Filtering.
· Deep Match to Rank Model for Personalized Click-Through Rate Prediction.
· Deep Time-Stream Framework for Click-Through Rate Prediction by Tracking Interest Evolution.
· Improved Algorithms for Conservative Exploration in Bandits.
· Linear Bandits with Feature Feedback.
总结
随着推荐系统的重要性越来越大,研究推荐的人逐渐在增多;随着工业界所产生的用户数据越来越多,工业界研究推荐的优势也越来越大。此次会议上出现了许多推荐的应用,比如音乐推荐、兴趣点推荐、旅游推荐、论文推荐等;同时也有相关的研究放到冷启动、推荐效率等问题上。
更多论文请前往Github项目RSPapers获取。













网友评论