推荐系统中的冷启动问题研究综述

推荐系统能够快捷、准确地定位用户真正需要的信息,解决网络信息过载问题。其中协同过滤推荐技术是推荐系统应用最广泛和成功的技术,但该技术面临冷启动问题的挑战。本文分析冷启动问题的产生原因,阐述研究冷启动问题的意义,重点总结解决冷启动问题的算法现状,分析比较它们的性能差异和各自存在的优缺点,从而便于使用者在解决冷启动问题时对算法的选择和使用。     

Sub-ambient cold-start emissions reduction using SwRI® PO x catalyst technology

This paper describes the application of SwRI’s cold-start PO _x catalyst technology to reduce cold-start hydrocarbon emissions from a US Tier 2 vehicle at −7 °C. A reduction in −7 °C (20 °F) cold-start hydrocarbons will help US Tier 2 vehicles meet the proposed EPA NMOG standards. Improvements in cold temperature hydrocarbon emissions would also be beneficial in many parts of Europe during the winter months. In this work, a total hydrocarbon reduction of 19% was realized at 24 °C, in line with previous results, but only up to 3% at −7 °C. Insufficient oxygen in the engine-out exhaust gas at −7 °C was determined to be the reason why the PO _x catalyst failed to significantly reduce HC emissions. Addition of supplemental oxygen to the exhaust during the cold-start, to simulate an adjustment in the engine calibration to less rich operation, resulted in a total hydrocarbon reduction of 18% with the PO _x catalysts in place, but no benefit when the PO _x catalysts were removed. Hence, the PO _x catalyst approach can be used to good effect, even under sub-ambient cold-start conditions.     

Contrastive Meta-learning on Heterogeneous Information Networks for Cold-start Recommendation

In recommender systems, the cold-start issue is challenging due to the lack of interactions between users or items. Such an issue can be alleviated via data-level and model-level strategies. Traditional data-level methods employ auxiliary information like feature information to enhance the learning of user and item embeddings. Recently, Heterogeneous Information Networks (HINs) have been incorporated into the recommender system as they provide more fruitful auxiliary information and meaningful semantics. However, these models are unable to capture the structural and semantic information comprehensively and neglect the unlabeled information of HINs during training. Model-level methods propose to apply the meta-learning framework which naturally fits into the cold-start issue, as it learns the prior knowledge from similar tasks and adapts to new tasks quickly with few labeled samples. Therefore, we propose a contrastive meta-learning framework on HINs named CM-HIN, which addresses the cold-start issue at both data level and model level. In particular, we explore meta-path and network schema views to describe the higher-order and local structural information of HINs. Within meta-path and network schema views, contrastive learning is adopted to mine the unlabeled information of HINs and incorporate these two views. Extensive experiments on three benchmark datasets demonstrate that CM-HIN outperforms all state-of-the-art baselines in three cold-start scenarios.     

基于改进降噪自动编码器的点击率预测

点击率(CTR)预测是个性化广告和推荐系统中的一项基本任务. 针对提升点击率预测效果和处理冷启动问题, 本文中提出了一种基于改进降噪自动编码器的点击率预测模型ADVAE (ADditional Variational AutoEncoder),该模型在输入数据加入高斯随机噪声, 利用改进的降噪自动编码器生成新的嵌入特征, 然后分别进行低阶和高阶的特征交互来预测用户点击行为. 该方法可以在数据稀疏以及系统冷启动情况下, 更深层地学习特征嵌入与交叉之间的关系. 该模型关注特征域之间的交互, 动态修复低频数据的特征嵌入, 具有更强的鲁棒性. 此外, 该方法可以动态应用到其他深度学习模型, 具有更高的灵活性. 实验结果表明, 该方法在点击率预测和系统冷启动问题上的性能表现均优于现有方法.     

Cold-Start Link Prediction via Weighted Symmetric Nonnegative Matrix Factorization with Graph Regularization

Link prediction has attracted wide attention among interdisciplinary researchers as an important issue in complex network. It aims to predict the missing links in current networks and new links that will appear in future networks. Despite the presence of missing links in the target network of link prediction studies, the network it processes remains macroscopically as a large connected graph. However, the complexity of the real world makes the complex networks abstracted from real systems often contain many isolated nodes. This phenomenon leads to existing link prediction methods not to efficiently implement the prediction of missing edges on isolated nodes. Therefore, the cold-start link prediction is favored as one of the most valuable subproblems of traditional link prediction. However, due to the loss of many links in the observation network, the topological information available for completing the link prediction task is extremely scarce. This presents a severe challenge for the study of cold-start link prediction. Therefore, how to mine and fuse more available non-topological information from observed network becomes the key point to solve the problem of cold-start link prediction. In this paper, we propose a framework for solving the cold-start link prediction problem, a joint-weighted symmetric nonnegative matrix factorization model fusing graph regularization information, based on low-rank approximation algorithms in the field of machine learning. First, the nonlinear features in high-dimensional space of node attributes are captured by the designed graph regularization term. Second, using a weighted matrix, we associate the attribute similarity and first order structure information of nodes and constrain each other. Finally, a unified framework for implementing cold-start link prediction is constructed by using a symmetric nonnegative matrix factorization model to integrate the multiple information extracted together. Extensive experimental validation on five real networks with attributes shows that the proposed model has very good predictive performance when predicting missing edges of isolated nodes.     

风电主齿箱系统低温待机方案的研究

以1.5 MW风力发电机组为例,对目前风电主齿箱系统低温环境待机困难的现状进行研究,通过该类主齿箱系统待机时加热与散热的计算,对循环周期内各时段润滑油的温度进行推算,根据推算结果,解释了主齿箱系统低温环境待机困难的主要原因,并提出了低温待机方案的改进策略.     

The Potential for Use of Gold in Automotive Pollution Control Technologies: A Short Review

Recent studies relevant to the future use of gold catalysts in the automotive industry are summarized. Gold catalysts have been examined for their potential in low temperature activity to combat cold-start emission problems and removal of NO _x from lean-burn gasoline and diesel engines. The justification for developing gold catalyst technologies is based both on their promising technical performance and the relatively stable price and greater availability of gold compared with the platinum group metals (PGMs). Use of gold catalysts under mild conditions could also produce lower proportions of CO in the hydrogen streams used for automotive fuel cells and they could be used for the selective oxidation of carbon monoxide in these hydrogen streams in order to increase the efficiency of these fuel cells. Technical barriers are indicated which still remain to be overcome before application of gold catalysts is successful in the automotive sphere.     

钻机用冷启动装置的研制与应用

石油钻机柴油机启动是整台钻机的关键之一，而提供压缩空气又是启动柴油机的必要条件。冷启动机组是为钻机供气的重要部件，相对于辅助电动压缩机而言，构成了柴油机启动双保险；文章叙述了冷启动装置在设计中关键设备的选型以及优点；本装置是由雅马哈汽油发动机和Z系列压缩机组成．具有启动使用可靠、体积小、安全环保等特点；用户使用反映良好，是现代钻机的重要辅助设备。     

协同过滤算法中冷启动问题研究

为了解决传统协同过滤算法的冷启动问题,提高算法的推荐质量,本文针对协同过滤算法中的冷启动问题进行研究,提出了两种改进的算法.新用户冷启动：融合用户信息模型的基于用户的协同过滤算法;新项目冷启动：采用层次聚类的基于项目的协同过滤算法.将新算法在网络开源数据集MovieLens上进行实验验证,比较改进算法和传统算法在查全率和查准率上的差异,结果表明改进算法能够有效地提高算法的推荐质量,缓解新用户和新项目的冷启动问题.     

隐式反馈场景中结合信任与相似度的排序推荐*

推荐系统作为解决信息过载问题的有效工具,能通过海量历史行为数据挖掘用户偏好,为其提供个性化推荐服务。针对如何利用隐式反馈数据实现个性化推荐进行研究,提出了一种结合信任与相似度的排序模型TSBPR。首先计算受信度与相似度的混合权重取代二值信任关系,初始化新用户信任列表实现将新用户连接进信任网络,其次利用邻居的特征及信任矩阵修正目标用户的特征矩阵解决信任传递问题,最后通过在贝叶斯排序模型(Bayesian Personalized Ranking,BPR)中引入重新构建的信任模型及用户特征得到优化的模型参数并生成最终的项目排序列表。通过实验仿真,证明了TSBPR模型可以提高推荐性能和有效解决冷启动问题。