检索结果-内蒙古大学图书馆

Exploring deep gradient information for biometric image feature representation

Neurocomputing 2016年 213卷 162-171页

作者： Jianjun Qian Jian Yang Ying Tai Hao Zheng School of Computer Science and Engineering Nanjing University of Science and Technology China Key Laboratory of Trusted Cloud Computing and Big Data Analysis Nanjing Xiaozhuang University China

This paper presents a novel but simple biometric image feature representation method, called exploring deep gradient information (DGI). DGI first captures the local structure of an image by computing the histogram of gradient orientation of each macro-pixel (local patch around the reference pixel). Thus, one image can be decomposed into L sub-images (sub-orientation images) according to the gradient information of each macro-pixel since there are L bins in the local histogram. To enrich the gradient information, we also consider the gradient orientation and magnitude of original image as sub-images. For each sub-image, histogram of oriented gradient (HOG) is used to further explore the gradient orientation information. All HOG features are concatenated into one augmented super-vector. Finally, fisher linear discriminate analysis (FLDA) is applied to obtain the low-dimensional and discriminative feature vector. We evaluated the proposed method on the real-world face image datasets NUST-RWFR, Pubfig and LFW, the PolyU Finger-Knuckle-Print database and the PolyU Palmprint database. Experimental results clearly demonstrate the effectiveness of the proposed DGI compared with state-of-the-art algorithms, e.g., SIFT, HOG, LBP, POEM, LARK and IDLS.

关键词： Image representation Deep gradients Feature extraction Biometrics

来源：评论

学校读者我要写书评

暂无评论

Intelligent Question and Answering System Based on SVM and Cosine Similarity

Intelligent Question and Answering System Based on SVM and C...

引用

2017 International Conference on Mathematics, Modelling and Simulation Technologies and Applications（MMSTA 2017)

作者： Wan-li SONG Wei-wei CHEN Ming-zhu ZHANG Key Laboratory of Trusted Cloud Computing and Big Data Analysis School of Information EngineeringNanjing Xiao Zhuang University

It is a tough task for teachers to answer all questions from students effectively and *** this paper,we design and implements an intelligent question answering system using Natural Language Processing,template classification,support vector *** system also calculates the similarity between the question and answer pairs by cosine similarity algorithm,and returns the most similar *** the user is not satisfied with the answer,the system will write the question into the public section to fall back on other *** answer will be evaluated and added to the QA base if it is passed with the corresponding *** that the questions and answers in the QA base continue to *** use the QA base of a network forum as the basic library to carry out the *** implementation and experimental results indicate that the proposed approach is achievable.

关键词： Intelligent system Question classification Question similarity Cosine similarity

来源：评论

学校读者我要写书评

暂无评论

Online Test Paper Composition Based on Genetic Algorithm

Online Test Paper Composition Based on Genetic Algorithm

引用

2018 3rd International Conference on Modelling, Simulation and Applied Mathematics(MSAM 2018)

作者： Wanli Song School of Information Engineering Nanjing Xiaozhuang University Key Laboratory of Trusted Cloud Computing and Big Data Analysis Nanjing XiaoZhuang University

The application of online examination is used more and more widely. Test paper composition is the core function in this application. Online examination system requires that the test paper should be quick and flexible, and the composition of questions should be reasonable and random. Test paper should not only meet the specific needs of users, ensure fairness, impartiality, but also be accurate and efficient. The intelligent test paper composition algorithm based on genetic algorithm is studied in this paper. Then this method is used to organize the structure of the test paper automatically, composites the examination content. Proposals are put forward to examines the degree of students' mastery of knowledge based on this method. This algorithm can automatically composite test papers according to the conditions of difficulty degree, knowledge coverage, and the proportion of questions. It can automatically composite test papers for online assessment. The practice of online testing system based on this algorithm shows that the algorithm is scientific and effective. On the premise of guaranteeing the quality of the test paper, it greatly improves the efficiency and pertinence of online testing. It is beneficial to improve students' learning efficiency and intellectualization of education.

关键词： test paper composition education intellectualization genetic algorithm E-learning

来源：评论

学校读者我要写书评

暂无评论

An attribute reduction framework for inconsistent decision tables from view of the boundary region

An attribute reduction framework for inconsistent decision t...

引用

2016 International Conference on Machine Learning and Cybernetics, ICMLC 2016

作者： Qian, Jin Chang, Yu-Hui Lv, Ping Key Laboratory of Cloud Computing and Intelligent Information Processing of Changzhou City Jiangsu University of Technology Changzhou213015 China Jiangsu Key Laboratory of Big Data Analysis Technology/B-DAT Nanjing University of Information Science and Technology Nanjing China

ISBN: (纸本)9781509003891

Attribute reduction is one of the key issues for data preprocess in data mining. Many heuristic attribute reduction algorithms based on discernibility matrix have been proposed for inconsistent decision tables. However, these methods are usually computationally time-consuming. To address this issue, the derived consistent decision tables are defined for different definitions of relative reducts. The computations for different reducts of the original inconsistent decision tables are converted into the computations for their corresponding reducts of the derived consistent datasets. The relative discernibility object pair and the more optimal relative discernibility degree from view of the boundary region are designed to accelerate the attribute reduction process. An efficient attribute reduction framework using relative discernibility degree is proposed for large datasets. Experimental results show that our attribute reduction algorithms are effective and feasible for large inconsistent datasets. © 2016 IEEE.

关键词： Decision tables

来源：评论

学校读者我要写书评

暂无评论

News Recommendation System Based on Collaborative Filtering and SVM

News Recommendation System Based on Collaborative Filtering ...

引用

2018 3rd International Conference on Automation, Mechanical and Electrical Engineering (AMEE 2018)

作者： Wan-li SONG School of Information Engineering Nanjing Xiaozhuang University Key Laboratory of Trusted Cloud Computing and Big Data Analysis Nanjing Xiao Zhuang University

News system requires news classification and personalized recommendation to improve user's efficiency and interest, and to enhance user's experiences. This paper constructed a news automatic classification and recommendation system through natural language processing, text classification, collaborative filtering algorithm. The published news contents were word-segmented and model-trained automatically first to determine which category the news belonging to. Users can also manually modify the classification so that later classification can be updated and improved. After that, the similarity between users was calculated by collaborative filtering and the users having higher similarity with the recommended users were selected. The news seen by the certain users were recommended to the users that were divided into the same group. This paper takes the news corpus of Fudan University's text classification research center as experimental data. Text classification accuracy is tested by this corpus. The experimental results show that the system can serve the news users well. It achieves effective classification and recommendation of news personally.

关键词： Recommendation algorithm Automatic classification Similarity Collaborative filtering

来源：评论

学校读者我要写书评

暂无评论

Smoothed Multi-view Subspace Clustering 1

引用

2nd International Conference on Neural computing for Advanced Applications, NCAA 2021

作者： Chen, Peng Liu, Liang Ma, Zhengrui Kang, Zhao Jangsu Automation Research Institute Lianyungang Jiangsu China University of Electronic Science and Technology of China Chengdu Sichuan China Trusted Cloud Computing and Big Data Key Laboratory of Sichuan Province Chengdu Sichuan China

ISBN: (数字)9789811651885

ISBN: (纸本)9789811651878

In recent years, multi-view subspace clustering has achieved impressive performance due to the exploitation of complementary information across multiple views. However, multi-view data can be very complicated and are not easy to cluster in real-world applications. Most existing methods operate on raw data and may not obtain the optimal solution. In this work, we propose a novel multi-view clustering method named smoothed multi-view subspace clustering (SMVSC) by employing a novel technique, i.e., graph filtering, to obtain a smooth representation for each view, in which similar data points have similar feature values. Specifically, it retains the graph geometric features through applying a low-pass filter. Consequently, it produces a "clustering-friendly" representation and greatly facilitates the downstream clustering task. Extensive experiments on benchmark datasets validate the superiority of our approach. analysis shows that graph filtering increases the separability of classes. © 2021, Springer Nature Singapore Pte Ltd.

关键词： Low pass filters

来源：评论

学校读者我要写书评

暂无评论

Sentiment analysis of Tourism Micro-blog Comments

Sentiment Analysis of Tourism Micro-blog Comments

引用

2016 International Conference on Artificial Intelligence: Techniques and Applications(AITA 2016)

作者： Wan-li SONG Jun-hua WANG Key Laboratory of Trusted Cloud Computing and Big Data Analysis School of Information EngineeringNanjing XiaoZhuang University Nanjing Vocational Institute of Transport Technology College of Computer and Information Hohai University

Recent years,an amount of tourism micro-blog comments on the Internet have become an important source of information for potential customers and to improve the service quality. These micro-blog comments do help to research tourism resources or services before making decisions. Thus,sentiment analysis of tourism micro-blog comments has become a hot issue in the field of natural language processing and text mining. We designed a system called SASTMC by using web crawler,Chinese words segmentation,emotion words dictionary and an improved TF-IDF algorithm. It enhances expression ability of sentiment information of text words. Experiments on Sina micro-blog comments datasets demonstrate that our method can do the task well.

关键词： Micro-blog comment Web crawler Words segmentation Sentiment analysis TF-IDF

来源：评论

学校读者我要写书评

暂无评论

Time-based trajectory data partitioning for efficient range query 23rd

Time-based trajectory data partitioning for efficient range ...

引用

23rd International Conference on database Systems for Advanced Applications, DASFAA 2018

作者： Yue, Zhongwei Zhang, Jingwei Zhang, Huibing Yang, Qing Guangxi Key Laboratory of Trusted Software Guilin University of Electronic Technology Guilin541004 China Guangxi Cooperative Innovation Center of Cloud Computing and Big Data Guilin University of Electronic Technology Guilin541004 China Guangxi Key Laboratory of Automatic Measurement Technology and Instrument Guilin University of Electronic Technology Guilin541004 China

ISBN: (纸本)9783319914541

The popularity of mobile terminals has given rise to an extremely large number of trajectories of moving objects. As a result, it is critical to provide effective and efficient query operations on large-scale trajectory data for further retrieval and analysis. Considering data partition has a great influence on processing large-scale data, we present a time-based partitioning technique on trajectory data. This partitioning technique can be applied on the distributed framework to improve the performance of range queries on massive trajectory data. Furthermore, the proposed method adopts time-based hash strategy to ensure both the partition balancing and less partitioning time. Especially, existing trajectory data are not required to be repartitioned when new data arrive. Extensive experiments on three real data sets demonstrated that the performance of the proposed technique outperformed other partitioning techniques. © Springer International Publishing AG, part of Springer Nature 2018.

关键词： Trajectories

来源：评论

学校读者我要写书评

暂无评论

Government intervention model based on behavioral heterogeneity for China’s stock market

引用

Financial Innovation 2022年第1期8卷 2569-2587页

作者： Zhong-Qiang Zhou Jie Li Wei Zhang Xiong Xiong School of Big Data Applications and Economics Guizhou University of Finance and EconomicsGuiyang 550025China Guizhou Key Laboratory of Big Data Statistical Analysis Guizhou University of Finance and EconomicsGuiyang 550025China School of Finance Nanjing Audit UniversityNanjing 211815JiangsuChina College of Management and Economics Tianjin UniversityTianjin 300072China China Center for Social Computing and Analytics Tianjin UniversityTianjin 300072China

Active government intervention is a striking characteristic of the Chinese stock *** study develops a behavioral heterogeneous agent model(HAM)comprising fundamentalists,chartists,and stabilizers to investigate investors’dynamic switching mechanisms under government *** model introduces a new player,the stabilizer,into the HAM as a proxy for the *** use the model to examine government programs during the 2015 China stock market crash and find that it can replicate the dynamics of investor sentiment and asset *** addition,our analysis of two simulations,specifically the data-generating processes and shock response analysis,further corroborates the key conclusion that our intervention model not only maintains market stability but also promotes the return of risk asset prices to their fun-damental *** study concludes that government interventions guided by the new HAM can alleviate the dilemma between reducing price volatility and improving price efficiency in future intervention programs.

关键词： Government intervention Excess volatility Behavioral heterogeneity Heterogeneous agent model

来源：评论

学校读者我要写书评

暂无评论

Intersection-union dual-stream cross-attention Lova-SwinUnet for skin cancer hair segmentation and image repair

引用

Computers in Biology and Medicine 2024年 180卷 108931页

作者： Qin, Juanjuan Pei, Dong Guo, Qian Cai, Xingjuan Xie, Liping Zhang, Wensheng Shanxi Key Laboratory of Big Data Analysis and Parallel Computing Taiyuan University of Science and Technology Taiyuan China State Key Laboratory for Novel Software Technology at Nanjing University Nanjing University Nanjing China Beijing China

Skin cancer images have hair occlusion problems, which greatly affects the accuracy of diagnosis and classification. Current dermoscopic hair removal methods use segmentation networks to locate hairs, and then uses repair networks to perform image repair. However, it is difficult to segment hair and capture the overall structure between hairs because of the hair being thin, unclear, and similar in color to the entire image. When conducting image restoration tasks, the only available images are those obstructed by hair, and there is no corresponding ground truth (supervised data) of the same scene without hair obstruction. In addition, the texture information and structural information used in existing repair methods are often insufficient, which leads to poor results in skin cancer image repair. To address these challenges, we propose the intersection-union dual-stream cross-attention Lova-SwinUnet (IUDC-LS). Firstly, we propose the Lova-SwinUnet module, which embeds Lovasz loss function into Swin-Unet, enabling the network to better capture features of various scales, thus obtaining better hair mask segmentation results. Secondly, we design the intersection-union (IU) module, which takes the mask results obtained in the previous step for pairwise intersection or union, and then overlays the results on the skin cancer image without hair to generate the labeled training data. This turns the unsupervised image repair task into the supervised one. Finally, we propose the dual-stream cross-attention (DC) module, which makes texture information and structure information interact with each other, and then uses cross-attention to make the network pay attention to the more important texture information and structure information in the fusion process of texture information and structure information, so as to improve the effect of image repair. The experimental results show that the PSNR index and SSIM index of the proposed method are increased by 5.4875 and 0.0401 compared w

关键词： Image segmentation

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：