检索结果-内蒙古大学图书馆

The Second International Conference of Young Computer Scientists, Engineers and Educators,ICYCSEE 2016

作者： Wenmin Huang Jiquan Ma Enbin Zhang College of Computer Science and Technology Heilongjiang University Key Laboratory of Database and Parallel Computing of Heilongjiang Province

A new Gaussian mixture model is used to improve the quality of propagation method for SFS in this paper. The improved algorithm can overcome most difficulties of propagation SFS method including slow convergence, inter-dependence of propagation nodes and error accumulation. To slow convergence and interdependence of propagation nodes, stable propagation source and inte-gration path are used to make sure that the reconstruction work of each pixel in the image is independent. A Gaussian mixture model based on prior conditions is proposed to fix the error of integration. Good result has been achieved in the experiment for Lambert composite image of the front illumination.

关键词： Shape from shading Propagation method Silhouette Gaussian mixture model Surface reconstruction

来源：评论

学校读者我要写书评

暂无评论

Sensitivity loss training based implicit feedback

Sensitivity loss training based implicit feedback

引用

International Conference on parallel and Distributed Systems (ICPADS)

作者： Kun Li Nan Wang Xinyu Liu Key Laboratory of Database and Parallel Computing of Heilongjiang Province School of Computer Science and Technology Heilongjiang University Harbin China

ISBN: (纸本)9781665408790

In recommender systems, due to the lack of explicit feedback features, datasets with implicit feedback are always accustomed to train all samples without separating them during model training, without considering the non-consistency of samples. This leads to a significant decrease in sample utilization and creates challenges for model training. Also, little work has been done to explore the intrinsic laws implied in the implicit feedback dataset and how to effectively train the implicit feedback data. In this paper, we first summarize the variation pattern of loss with model training for different rating samples in the explicit feedback dataset, and find that model training is highly sensitive to the ratings. Second, we design an adaptive hierarchical training function with dynamic thresholds that can effectively distinguish different rating samples in the dataset, thus optimizing the implicit feedback dataset into an explicit feedback dataset to some extent. Finally, to better learn samples with different ratings, we also propose an adaptive hierarchical training strategy to obtain better training results in the implicit feedback dataset. Extensive experiments on three datasets show that our approach achieves excellent performance and greatly improves the performance of the model.

关键词： Training Learning systems Adaptation models Sensitivity Conferences Recommender systems

来源：评论

学校读者我要写书评

暂无评论

Quick-merge sort algorithm based on Multi-core linux

Quick-merge sort algorithm based on Multi-core linux

引用

International Conference on Mechatronic Science, Electric Engineering and Computer

作者： Yong Liu Yan Yang School of Computer Science and Technology Key Laboratory of Database and Parallel Computing of Heilongjiang Province Harbin China

ISBN: (纸本)9781479925667

Sorting a list of input numbers is one of the most fundamental problems in the field of computer science. This paper presents an efficient implementation and detailed analysis of quick sort algorithm called QM-sort on Multi-core CPU architectures. Our QM-sort has two stages. The first stage is making blocks sorted, and the second stage is merging the sorted blocks. At the first stage, we propose a parallel quick sort algorithm named BPQsort. The run time of BPQsort achieves 40%-50% better than Qsort (STL), so we make blocks sorted more efficient. On the other hand, our algorithm can read file in parallel, thus reduce the time of reading file from disk. At the second stage, our parallel merge algorithm can operate the data on consecutive memory. Thus we avoid the expensive unaligned load/store operations, thus make our merge algorithm faster. At last, the run time of QM-sort is 10%-15% better than quick sort based on OpenMP.

关键词： Linux Instruction sets Hardware Algorithm design and analysis

来源：评论

学校读者我要写书评

暂无评论

Notice of Retraction: Collaborative filtering based on time division

Notice of Retraction: Collaborative filtering based on time ...

引用

International Conference on Computer Science and Information Technology (CSIT)

作者： Yan Yang Long Yun Key Laboratory of Database and Parallel Computing of Heilongjiang Province Department of Computer Science and Technology Heilongjiang University Harbin China

This article has been retracted by the publisher.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Optimization and parallelization of MRF community detection algorithm for a specific network

引用

International Journal of Performability Engineering 2019年第8期15卷 2153-2164页

作者： Lu, Jun Zhang, Yuanzhong College of Computer Science and Technology Heilongjiang University Harbin150080 China Key Laboratory of Database and Parallel Computing of Heilongjiang Province Harbin150080 China

Research on the optimization and parallelization of the MRF network community detection algorithm for a specific network is carried out in this paper. Firstly, the principle of the existing algorithm is expounded, the algorithm is analyzed, and some problems are pointed out. Some optimization strategies and rules are proposed, including the extraction of variables and operations from inner loops to outer loops, the merging of related operations in loops, the removal of redundant loops, and the split of loops. In order to achieve better parallelism, OpenMP parallel computing of this method is realized by reversing the order of inner and outer loops. The influence of the density of network edges on the algorithm efficiency is also analyzed in this paper. The optimization and parallel algorithm can be applied to the module partition of Alzheimer's disease gene data, and the efficiency of the algorithm is greatly improved. The optimization strategies and rules proposed in this paper can be further extended to general situations. It is significant in practical applications. © 2019 Totem Publisher, Inc. All rights reserved.

关键词： Efficiency

来源：评论

学校读者我要写书评

暂无评论

Estimation of the number of distinct values over data stream based on compound sliding window

Journal of Software

引用

Journal of Software 2013年第1期8卷 19-24页

作者： Zhong, Yingli Zhu, Jinghua Ren, Meirui Yang, Yan School of Computer Science and Technology Heilongjiang University Harbin 150080 China Key Laboratory of Database and Parallel Computing Heilongjiang Province Harbin 150080 China

Estimating the number of distinct values in a data stream is a vital problem with many applications such as complex join query over multiple data streams. In this paper, we focus on the continuous and periodic distinct values estimation over sliding windows. We propose a compound sliding window model to compute the distinct values over basic sliding windows in an incremental way. LDV, HDV and AHDV are the three algorithms that are based on compound sliding windows. The basic idea behind the compound sliding windows is to organize the basic windows into a Hash table according to distinct values. Whenever a new data arrives at the data stream, it is inserted into a basic window. Once the basic window is full, a scan using distinct values is executed and the distinct values number is updated incrementally. Theoretical analysis and experiment results show that the distinct values estimation algorithms based on compound sliding windows have a great performance benefits. © 2013 ACADEMY PUBLISHER.

关键词： Decoding

来源：评论

学校读者我要写书评

暂无评论

Node classification problem based on potential information and representation learning between nodes

Node classification problem based on potential information a...

引用

International Conference on Information Science and Control Engineering (ICISCE)

作者： Mingming Li Yan Yang School of Computer Science Technology Heilongjiang University Hearbin Province China Key Laboratory of Database and Parallel Computing of Heilongjiang Province Harbin Province China

Representation learning of social network nodes refers to the feature expression of nodes in low-dimensional vector space. Most of the current methods are based on the network topology to achieve node feature learning, but the potential information of the network topology is not fully utilized. To solve this problem, we propose a graph embedding algorithm based on the similarity between nodes. This method obtains a new network topology through the second-order similarity between nodes. This network topology can enhance the correlation between nodes, thereby improving the effect of node classification in the network. We use Macro-F1 and Micro-F1 evaluation indicators to evaluate the benchmark algorithm and the SNE algorithm proposed in this paper on four real data sets. The experimental results show that the SNE algorithm can improve the effect of node classification.

关键词： Information science Electronic publishing Correlation Control engineering Network topology Social networking (online) Information services

来源：评论

学校读者我要写书评

暂无评论

Cross-level Relational Graph Contrastive Learning for Bundle Recommendation

Cross-level Relational Graph Contrastive Learning for Bundle...

引用

2023 IEEE International Conference on Web Services, ICWS 2023

作者： Wang, Nan Sun, Jiabao Li, Jinbao College of Computer Science and Technology Heilongjiang University Key Laboratory of Database and Parallel Computing of Heilongjiang University Harbin China Qilu University of Technology Shandong Artificial Intelligence Institute School of Mathematics and Statistics Jinan China

ISBN: (纸本)9798350304855

Different from the traditional recommendation system which regards a single item as an atomic unit, the purpose of bundle recommendation is to recommend a set of items to users. In recent years, some work often uses the model based on GNN to solve the bundle recommendation problem. However, compared with the traditional recommendation models, the GNN-based bundle recommendation has a more serious data sparse problem, which significantly reduces the performance of the model. Meanwhile, due to the lack of modeling ability of complex interactive information and dependent information, it is still far from enough to learn the representation of users and bundles. In view of the advantages of contrast learning in mining its own supervisory signals, we propose a model named Cross-level relational graph Contrastive Learning for Bundle Recommendation (CL2BRec). Different from previous contrastive learning methods, our method learns representations of users and bundles from multiple relational graphs of different levels through a crossview contrastive learning paradigm. Specifically, we regard the user-item/bundle graph as a local interaction relation view, the bundle-item graph as a local dependency relation view, and the user-bundle-item tripartite graph as a global structure view to build a cross-level bundle recommendation model. Extensive experiments conducted on two benchmark datasets demonstrate that CL2BRec outperforms the state-of-the-art methods by over 12.56% and 8.6%, respectively. © 2023 IEEE.

关键词： Recommender systems

来源：评论

学校读者我要写书评

暂无评论

A Bidirectional Representation Recommendation Based On Relevance Information Enhancement To Solve User Interest Drift

A Bidirectional Representation Recommendation Based On Relev...

引用

2023 IEEE International Conference on Web Services, ICWS 2023

作者： Wang, Nan Li, Kun Jiang, Xuefeng Li, Jinbao College of Computer Science and Technology Heilongjiang University Key Laboratory of Database and Parallel Computing Heilongjiang University Harbin China Qilu University of Technology Shandong Artificial Intelligence Institute School of Mathematics and Statistics Jinan China

ISBN: (纸本)9798350304855

The phenomenon that user interests change dynamically over time is called user interest drift, which makes it difficult for recommendation systems to obtain users' real interests. In order to better provide users with personalized services, it is very important to study recommendation methods that can be applied to user interest drift. However, the new interests of users are often very different from the interests that users have interacted with historically. It makes the correlation between the original user's historical interaction and the user's new interest very weak. Meanwhile, the information contained in user's sessions is not enough to support the inference of the new interest, resulting in a serious decrease in model performance. To alleviate the above problem, we propose a bidirectional representation recommendation based on relevance information enhancement to solve user interest drift (RA-Bert). The model is based on Bert and uses bidirectional self-attention mechanism to represent user sessions. In this model, we design a relevance information enhancement module to adaptively capture the information associated with the user's new interest from neighbors' sessions. Then, we integrate the relevance information into user's history sessions, which can effectively enhance the correlation between user's history sessions and the new interest. Further, in order to effectively supplement the missing relevance information in users' sessions, we design a new attention unit, which explicitly models the difference information between users and neighbors. Extensive experiments have demonstrated the effectiveness of our approach. Our code is publicly available at the link: https://***/LucasZen/paper_code. © 2023 IEEE.

关键词： Recommender systems

来源：评论

学校读者我要写书评

暂无评论

Time constraint-based team formation in social networks

Time constraint-based team formation in social networks

引用

International Conference on Mechatronic Science, Electric Engineering and Computer

作者： Xue Han Yong Liu Xingkai Guo Xianyi Wu Xiaolong Song School of Computer Science and Technology Heilongjiang University Harbin China Key Laboratory of Database and Parallel Computing of Heilongjiang Province Heilongjiang University

ISBN: (纸本)9781479925667

We investigate the problem of discovering a team of experts from a social network. Given a task with the skill set and the time span length, our goal is to obtain a team that covers all the skills, and have a low communication cost. Furthermore, each member of the team must have common free time whose length is no less than the span length required by the task. Previous works only attempt to cover the required skills and minimize the communication cost without taking the common free time of the individuals into account. We are the first to consider time factor in team formation and thus propose a novel team formation formulation with time constraint. We show that the problem is NP-hard. We first extend the RarestFirst algorithm to deal with our new problem, then propose two novel pruning strategies, called time pruning and skill pruning respectively, and an efficient algorithm called TSP-TF, to efficiently obtain an effective team for a given task. Experimental results show that TSP-TF outperforms the generalized RarestFirst algorithm by more than an order of magnitude.

关键词： Algorithm design and analysis Social network services Time factors Collaboration Data mining Time complexity databases

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：