On recommendation problems beyond points of interest

Recommendation systems aim to recommend items or packages of items that are likely to be of interest to users. Previous work on recommendation systems has mostly focused on recommending points of interest (POI), to identify and suggest top-k items or packages that meet selection criteria and satisfy compatibility constraints on items in a package, where the (packages of) items are ranked by their usefulness to the users. As opposed to prior work, this paper investigates two issues beyond POI recommendation that are also important to recommendation systems. When there exist no sufficiently many POI that can be recommended, we propose (1) query relaxation recommendation to help users revise their selection criteria, or (2) adjustment recommendation to guide recommendation systems to modify their item collections, such that the users׳ requirements can be satisfied.We study two related problems, to decide (1) whether the query expressing the selection criteria can be relaxed to a limited extent, and (2) whether we can update a bounded number of items, such that the users can get desired recommendations. We establish the upper and lower bounds of these problems, all matching, for both combined and data complexity, when selection criteria and compatibility constraints are expressed in a variety of query languages, for both item recommendation and package recommendation. To understand where the complexity comes from, we also study the impact of variable sizes of packages, compatibility constraints and selection criteria on the analyses of these problems. Our results indicate that in most cases the complexity bounds of query relaxation and adjustment recommendation are comparable to their counterparts of the basic recommendation problem for testing whether a given set of (resp. packages of) items makes top-k items (resp. packages). In other words, extending recommendation systems with the query relaxation and adjustment recommendation functionalities typically does not incur extra overhead.     

基于改进用户兴趣模型的个性化图书推荐算法

为了有效提高推荐算法的精确度,提出了一种适用于个性化图书推荐的改进隐含狄利克雷分配(Latent Dirichlet Allocation,LDA)用户兴趣模型。首先在借阅者-借阅者评分矩阵的基础上,通过增加借阅者特征信息相似度计算和借阅者-图书属性相似度计算,对图书内容相似度计算方法进行了改进。然后采用LDA主题挖掘模型来实现个性化图书推荐,并给出了相应的参数估计过程。实验结果显示:相比传统算法,提出的算法具有较高的准确度,能有效对图书进行挖掘,为借阅者推荐个性化和潜在感兴的书籍。     

基于深度神经网络和加权隐反馈的个性化推荐

改进的矩阵分解(SVD++)将用户和物品特征向量的内积作为用户对物品的评分,而内积无法捕捉用户与物品之间复杂的高阶非线性关系.此外,SVD++在融入用户隐式反馈时,未区分不同交互物品对于用户特征表达的贡献.针对上述问题,文中提出基于深度神经网络和加权隐反馈的推荐算法(DeepNASVD++),采用深度神经网络建模用户与物品之间的关系,使用注意力机制计算历史交互物品在建模用户隐式反馈时的权重.在公开数据集上的实验验证文中算法的有效性.     

基于经验分布和KL散度的协同过滤推荐质量评价研究

提出了一种基于经验分布和KL散度的协同过滤推荐质量评价方法RQE-EDKL（recommendation quality evaluation based on empirical distribution and KL divergence）。RQE-EDKL首先利用历史用户—商品数据生成不同商品数量下的商品历史使用概率分布；然后，利用该分布与各个协同过滤推荐方法得到的用户商品使用概率进行比较，计算其KL散度；最后，将KL散度最小的推荐结果视为最佳推荐结果并推送给用户。在TalkingData数据集上的实验结果表明，RQE-EDKL评价方法能够有效地在不同的推荐结果中选择更为切合用户真实需求的推荐结果，从而提高了协同过滤推荐的质量。     

基于热扩散影响力传播的社交网络个性化推荐算法

有针对性地为用户提供推荐,提高互联网信息利用率是个性化推荐系统的主要目标.文中基于热扩散传播概率模型,结合用户在社交网络中隐含的跟随关系,提出基于热扩散影响力传播的社交网络个性化推荐算法.首先,算法将现实生活中人与人的朋友关系转化为购物网络中用户与用户的跟随关系,构建异构信息网络图,计算用户之间的复合相似度.然后,利用基于热扩散概率模型模拟社会网络中影响力的传播过程,计算社交网络中用户的跟随概率分数并精确排序,筛选与目标用户相似的邻近用户.最后,根据目标邻近用户对各个产品的评分,将评分较高、具有潜在兴趣的产品推荐给目标用户,实现个性化的用户推荐.在公开数据集上与现有的个性化推荐算法进行对比,实验表明,文中算法具有较好的精确度和多样化的推荐效果.     

Android fine-grained permission control system with real-time expert recommendations

In current Android architecture design, users have to decide whether an app is safe to use or not. Expert users can make savvy decisions to prevent unnecessary privacy breach. However, inexperienced users may not be able to decide correctly. To assist inexperienced users to make a right permission granting decisions, we propose RecDroid. RecDroid is a crowdsourcing recommendation framework that facilitates a user-help-user environment regarding smartphone permission control. In this framework, the responses from expert users are aggregated and recommended to other users. We implement our prototype on Android platform and evaluated the system through simulation and real user study.     

对市政公用工程CECS标准编制工作及若干问题的思考

该文首先诠释了CECS标准(推荐性标准)的基本内涵,结合作者的工作实际,论述了近年来市政公用工程的工作实践和主要成果,对CECS标准编制工作提出了若干思考,并指出了存在的问题和相关建议,以期推动我国工程建设标准化工作在祖国现代化建设中发挥更大的作用.     

Comprehension and acceptability of on-board traffic information: Beliefs and driving behaviour

Co-Drive on-board traffic information system is a complementary tool providing a dynamic management of transportation infrastructure and traffic as well as the diffusion of accurate real-time information about the road environment and motorists’ driving behaviour. The aim of this study was to examine drivers’ acceptability of Co-Drive by investigating the impact of traffic information provided via on-board display devices on motorists’ beliefs and behaviour.