检索结果-内蒙古大学图书馆

Deep learning compact binary codes for fingerprint indexing

Frontiers of Information Technology & Electronic Engineering 2018年第9期19卷 1112-1123页

作者： Chao-chao BAI Wei-qiang WANG Tong ZHAO Ru-xin WANG Ming-qiang LI School of Computer and Control University of Chinese Academy of Science School of Mathematical Sciences University of Chinese Academy of Science Key Laboratory of Big Data Mining and Knowledge Management Chinese Academy of Science

With the rapid growth in fingerprint databases, it has become necessary to develop excellent fingerprint indexing to achieve efficiency and accuracy. Fingerprint indexing has been widely studied with real-valued features,but few studies focus on binary feature representation, which is more suitable to identify fingerprints efficiently in large-scale fingerprint databases. In this study, we propose a deep compact binary minutia cylinder code（DCBMCC）as an effective and discriminative feature representation for fingerprint indexing. Specifically, the minutia cylinder code（MCC）, as the state-of-the-art fingerprint representation, is analyzed and its shortcomings are ***, we propose a novel fingerprint indexing method based on deep neural networks to learn *** novel network restricts the penultimate layer to directly output binary codes. Moreover, we incorporate independence, balance, quantization-loss-minimum, and similarity-preservation properties in this learning ***, a multi-index hashing（MIH） based fingerprint indexing scheme further speeds up the exact search in the Hamming space by building multiple hash tables on binary code substrings. Furthermore, numerous experiments on public databases show that the proposed approach is an outstanding fingerprint indexing method since it has an extremely small error rate with a very low penetration rate.

关键词： Fingerprint indexing Minutia cylinder code Deep neural network Multi-index hashing

来源：评论

学校读者我要写书评

暂无评论

Incomplete-view oriented kernel learning method with generalization error bound

引用

Information Sciences 2021年 581卷 951-977页

作者： Tian, Yingjie Fu, Saiji Tang, Jingjing School of Economics and Management University of Chinese Academy of Sciences Beijing100190 China Research Center on Fictitious Economy and Data Science Chinese Academy of Sciences Beijing100190 China Key Laboratory of Big Data Mining and Knowledge Management Chinese Academy of Sciences Beijing100190 China School of Business Administration Faculty of Business Administration Southwestern University of Finance and Economics Chengdu611130 China

Multi-view learning (MVL) is a promising direction and most MVL methods work under the assumption that data are complete in all views. However, this assumption is often violated in practice due to the difficulties of high cost, equipment failure, and so on. Besides, the amount of data with missing views is particularly huge. Therefore, it is challenging yet valuable to fully leverage the large-scale incomplete-view data. Enlightened by the reduced support vector machine (RSVM) and multi-view privileged support vector machine (PSVM-2V), this paper proposes an efficient kernel method called reduced PSVM-2V (RPSVM-2V). It can not only provide a novel solution to process incomplete-view data, but also can be adjusted to address large-scale complete-view learning problem efficiently. In addition, the idea of replacing the full kernel with a smaller rectangular reduced kernel is extended to develop another two MVL methods, i.e., reduced SVM-2K (RSVM-2K) and reduced multiple kernel learning method (RMKL). Furthermore, we analyze the generalization error bounds of RPSVM-2V and two extensions by using Rademacher complexity. The comprehensive experiments demonstrate that our proposed models can achieve comparable performance with less time and memory cost. The spectral analysis has further verified the effectiveness of the reduced kernel used in our models. © 2021 Elsevier Inc.

关键词： Support vector machines

来源：评论

学校读者我要写书评

暂无评论

An Efficient Algorithm for Judicious Partition of Hypergraphs 11th

An Efficient Algorithm for Judicious Partition of Hypergraph...

引用

11th Annual International Conference on Combinatorial Optimization and Applications (COCOA)

作者： Tan, Tunzi Gui, Jihong Wang, Sainan Gao, Suixiang Yang, Wenguo Univ Chinease Acad Sci Sch Math Sci Beijing Peoples R China Chinese Acad Sci Key Lab Big Data Min & Knowledge Management Beijing Peoples R China

ISBN: (纸本)9783319711478;9783319711461

Judicious partition of hypergraphs H=(V, H) is to optimize several quantities simultaneously, and the goal of this paper is to partition the vertex set V into K parts: {V1, V2,, VK} so as to minimize the max{L(Vi), L(V2),, L(VK)}, where L(V3) is the number of hyperedges incident to the part V3 (H). The bounds for the objective function are given and the relationship between the maximum hyperdegree and the objective value is analyzed. Before giving an efficient algorithm for the judicious partition of hypergraphs, a sub -problem is obtained, which is proved to be an unweighted set cover problem, apart from a tiny difference. A greedy algorithm is applied to solve the sub -problem. Last but not least, the judicious partition of hypergraphs is successfully divided into a series of sub -problems and an efficient algorithm is developed for the original problem.

关键词： Hypergraph Judicious partition Approximation algorithm Minimum set cover

来源：评论

学校读者我要写书评

暂无评论

Research on Pricing of data Based on Bi-level Programming Model

引用

Annals of data Science 2024年第4期11卷 1391-1419页

作者： Ding, Yurong Tian, Yingjie School of Economics and Management University of Chinese Academy of Sciences Beijing100190 China Research Center on Fictitious Economy and Data Science Chinese Academy of Sciences Beijing100190 China Key Laboratory of Big Data Mining and Knowledge Management Chinese Academy of Sciences Beijing100190 China MOE Social Science Laboratory of Digital Economic Forecasts and Policy Simulation at UCAS Beijing100190 China

Effective value measurement and pricing methods can greatly promote the healthy development of data sharing, exchange and reuse. However, the uncertainty of data value and neglect of interactivity lead to information asymmetry in the transaction process. A perfect pricing system and well-designed data trading market (hereafter called data market) can widely promote data transactions. We take the three-agents data market as an example to construct a sound data trading process. The data owner who provides data records, the model buyer who is interested in buying machine learning (ML) model instances, and the data broker who interacts between the data owner and the model buyer. Based on the characteristics of data market, like truthfulness, revenue maximization, version control, fairness and non-arbitrage, we propose a data pricing methods based on different model versions. Firstly, we utilize market research and construct a revenue maximization (RM) problem to price the different versions of ML models and solve it with the RM-ILP process. However, the RM model based on market research has two major problems: one is that the model buyer has no incentive to tell the truth, that is, the model buyer will lie in the market research to obtain a lower model price;the other is that it asks the data broker to release version menu in advance, resulting in an inefficient operation of the data market. In view of the defects of the RM transaction model, we propose a model buyers behavior analysis, establish the revenue maximization function based on different data versions to establish a bi-level linear programming model. We further add the incentive compatibility constraint and the individual rationality constraint, taking the utility of the model buyer and the revenue of the data broker into account. This reflects the consumer driven model in the data transaction mode. Finally, the RM-BLP process is proposed to transform RM problem into an equivalent single-level integer program

关键词： Costs

来源：评论

学校读者我要写书评

暂无评论

CNN-MR for No Reference Video Quality Assessment 4

CNN-MR for No Reference Video Quality Assessment

引用

4th International Conference on Information Science and Control Engineering (ICISCE)

作者： Wang, Chunfeng Su, Li Huang, Qingming Univ Chinese Acad Sci Sch Comp & Control Beijing Peoples R China Chinese Acad Sci Key Lab Big Data Min & Knowledge Management Beijing Peoples R China

ISBN: (纸本)9781538630136

In this paper, we propose a no-reference video quality assessment (VQA) method based on Convolutional Neural Network (CNN) and Multi-Regression (CNN-MR). It is universal for non-specific types of distortion. First, we innovatively introduce the 2D convolutional neural network into VQA model to learn the spatial quality features at frame level. Second, the motion information is extracted as temporal quality features at sequence level. Finally, a multi-regression model is proposed to comprehensively measure video quality and the final quality is selected out from them according to human's psychological perception. The proposed CNN-MR method is tested on the famous LIVE database with numerous kinds of distortions. Compared with other state-of-the-art no reference VQA methods, the proposed method runs much faster while keeping the similar performance. As a no-reference VQA method, it is even comparable with most of the state-of-the-art full-reference VQA methods.

关键词： no-reference video quality assessment convolutional neural network multi-regression

来源：评论

学校读者我要写书评

暂无评论

Daily Ship Traffic Volume Statistics and Prediction Based on Automatic Identification System data 9

Daily Ship Traffic Volume Statistics and Prediction Based on...

引用

9th International Conference on Intelligent Human-Machine Systems and Cybernetics (IHMSC)

作者： Wang, Sainan Wang, Si Gao, Suixiang Yang, Wenguo Univ Chinese Acad Sci Sch Math Sci Beijing Peoples R China Chinese Acad Sci Key Lab Big Data Min & Knowledge Management Beijing Peoples R China

ISBN: (纸本)9781538630228

Daily ship traffic volume statistics and prediction are of great significance to shipping market. Reliable prediction of daily ship traffic volume can instruct shipping company to make sound judgment and decision for operational management. Because of the mobility of ships, it may be difficult to obtain the ship traffic volume automatically and frequently. This paper defines four types' daily ship traffic volume for a port and proposes autonomous statistic methods for counting daily ship traffic volume at port only based on Automatic Identification System (AIS) data. Take Shanghai port as instance, we count the daily ship traffic volume by using the proposed statistic methods for three common types of ship: cargo ship, passenger ship, and tanker ship. For ship traffic volume prediction, we apply a hybrid methodology that combines both Auto-Regressive and Moving Average (ARMA) and Artificial Neural Network (ANN) models to cargo ship and tanker ship, and show the comparison of ARMA, ANN and the hybrid methodology. Empirical results indicate that the hybrid methodology is efficient and superior for ship traffic volume prediction. But for passenger ship traffic volume, since the correlation coefficient plot shows feeble correlation in it, we just analyze some statistical characters for it.

关键词： Automatic Identification System (AIS) Daily ship traffic volume statistics and prediction Auto-Regressive and Moving Average (ARMA) Artificial Neural Network (ANN) Hybrid methodology

来源：评论

学校读者我要写书评

暂无评论

The performance of one belt and one road exchange rate: Based on improved singular spectrum analysis

引用

PHYSICA A-STATISTICAL MECHANICS AND ITS APPLICATIONS 2017年 483卷 299-308页

作者： Lai, Lin Guo, Kun China Univ Polit Sci & Law Business Sch Beijing Peoples R China Chinese Acad Sci Res Ctr Fictitious Econ & Data Sci Beijing Peoples R China Univ Chinese Acad Sci Sch Econ & Management Beijing Peoples R China Chinese Acad Sci Key Lab Big Data Min & Knowledge Management Beijing Peoples R China

"One Belt and One Road" strategy in China is on push of foreign trade openness at northwest, southwest and northeast, absorption of the excess capacity and new support for economic increase. However, the fluctuation in RMB exchange rate with the countries along the road is unstable so related Chinese enterprises will face high risk of exchange rate. Precise explanation or prediction for exchange rate has been the challengeable hop point in the international finance. This paper decomposed the One Belt One Road Exchange Rate Index (OBORR) and the RMB Effective Exchange Rate Index (CNYX) into trend term, market fluctuation term and noise term using improved singular spectrum analysis (SSA). It turns out that the increasing velocity of OBORR is greater than that of CNYX in the long term, and there is dynamic lead-lag structure in the medium term. In the short term, the fluctuation range and frequency of OBORR are greater than those of CNYX, which means there will be more exchange rate risks in One Belt and One Road countries. (C) 2017 Elsevier B.V. All rights reserved.

关键词： Singular spectrum analysis Increment for singular entropy One belt one road Exchange rate Thermal optimal path

来源：评论

学校读者我要写书评

暂无评论

AN EXTENDED MODEL FOR PROJECT PORTFOLIO SELECTION WITH PROJECT DIVISIBILITY AND INTERDEPENDENCY

引用

Journal of Systems Science and Systems Engineering 2016年第1期25卷 119-138页

作者： Xingmei Li Shu-Cherng Fang Xiaoling Guo Zhibin Deng Jianxun Qi School of Economics and Management North China Electric Power University Beijing 102206 China Industrial and Systems Engineering Department North Carolina State University Raleigh NC 27695 USA College of Engineering andlnformation Technology University of Chinese Academy of Sciences Beijing100049 China School of Management University of Chinese Academy of Sciences Beijing 100190 China Key Laboratory of Big Data Mining and Knowledge Management Chinese Academy of Sciences Beijing 100190 China

In this paper, we develop an extended model for the project portfolio selection problem over a planning horizon with multiple time periods. The model incorporates the factors of project divisibility and interdependency at the same time for real-life applications. The project divisibility is considered as a strategy, not an unfortunate event as in the literature, in choosing the best execution schedule for the projects, and the classical concept of＂project interdependencies＂ among fully executed projects is then extended to the portions of executed projects. Additional constraints of reinvestment consideration, setup cost, cardinality restriction, precedence relationship and scheduling are also included in the model. For efficient computations, an equivalent mixed integer linear programming representation of the proposed model is derived. Numerical examples under four scenarios are presented to highlight the characteristics of the proposed model. In particular, the positive effects of project divisibility are shown for the first time.

关键词： Project management project portfolio selection problem project divisibility project interdependency

来源：评论

学校读者我要写书评

暂无评论

Robustness and orthogonality: Time series forecasting via wavelets

引用

Information Sciences 2025年 717卷

作者： Yingjie Tian Siyu Zhao Xingyu Zhang School of Economics and Management University of Chinese Academy of Sciences Beijing 100190 China Research Center on Fictitious Economy and Data Science Chinese Academy of Sciences Beijing 100190 China Key Laboratory of Big Data Mining and Knowledge Management Chinese Academy of Sciences Beijing 100190 China School of Mathematical Sciences University of Chinese Academy of Sciences Beijing 100049 China School of Computer Science and Technology University of Chinese Academy of Sciences Beijing 101408 China National Key Laboratory of Space Integrated Information System Institute of Software Chinese Academy of Sciences Beijing 100190 China

Time series forecasting (TSF) is an important issue that spans various fields in daily life. With advancements in deep learning, numerous TSF methods have been developed leveraging both time and frequency domain characteristics of time series data. However, existing TSF methods primarily concentrate on data characteristics, yet which key property truly enhances model performance has not been thoroughly explored. Especially, for frequency domain-based methods, the reliance on stringent assumptions like the Linear Time-Invariant (LTI) system limits their applicability to non-stationary time series data. To address these issues, this paper introduces a novel TSF method called Robust Orthogonal decomposition via Wavelets (ROW). ROW employs wavelet transform to remove reliance on LTI assumptions and can capture local characteristics. Furthermore, this paper provides robustness analysis of ROW, i.e., the impact of orthogonal wavelet basis functions on model robustness. Through two examples, we illustrate that the robustness constant of ROW has a tighter upper bound under orthogonal wavelet basis, which indicates that ROW is more robust under perturbations. Experimental results also validate the effectiveness and robustness of ROW compared to existing methods.

关键词： Wavelet decomposition

来源：评论

学校读者我要写书评

暂无评论

Image recognition and empirical application of desert plant species based on convolutional neural network

引用

Journal of Arid Land 2022年第12期14卷 1440-1455页

作者： LI Jicai SUN Shiding JIANG Haoran TIAN Yingjie XU Xiaoliang School of Economics and Management University of Chinese Academy of SciencesBeijing 100190China School of Mathematical Sciences University of Chinese Academy of SciencesBeijing 100049China Key Laboratory of Big Data Mining and Knowledge Management Chinese Academy of SciencesBeijing 100190China Research Center on Fictitious Economy and Data Science Chinese Academy of SciencesBeijing 100190China Xinjiang Institute of Ecology and Geography Chinese Academy of SciencesUrumqi 830011China

In recent years,deep convolution neural network has exhibited excellent performance in computer vision and has a far-reaching *** plant taxonomic identification requires high expertise,which is *** nature reserves have problems such as incomplete species surveys,inaccurate taxonomic identification,and untimely updating of status *** and accurate recognition of plant images can be achieved by applying convolutional neural network technology to explore the best network *** 24 typical desert plant species that are widely distributed in the nature reserves in Xinjiang Uygur Autonomous Region of China as the research objects,this study established an image database and select the optimal network model for the image recognition of desert plant species to provide decision support for fine management in the nature reserves in Xinjiang,such as species investigation and monitoring,by using deep *** desert plant species were not included in the public dataset,the images used in this study were mainly obtained through field shooting and downloaded from the Plant Photo Bank of China(PPBC).After the sorting process and statistical analysis,a total of 2331 plant images were finally collected(2071 images from field collection and 260 images from the PPBC),including 24 plant species belonging to 14 families and 22 genera.A large number of numerical experiments were also carried out to compare a series of 37 convolutional neural network models with good performance,from different perspectives,to find the optimal network model that is most suitable for the image recognition of desert plant species in *** results revealed 24 models with a recognition Accuracy,of greater than 70.000%.Among which,Residual Network X_8GF(RegNetX_8GF)performs the best,with Accuracy,Precision,Recall,and F1(which refers to the harmonic mean of the Precision and Recall values)values of 78.33%,77.65%,69.55%,and 71.26%,*** t

关键词： desert plants image recognition deep learning convolutional neural network Residual Network X_8GF(RegNetX_8GF) Mobile Network V2(MobileNetV2) nature reserves

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：