检索结果-内蒙古大学图书馆

EPIMR:Prediction of Enhancer-Promoter Interactions by Multi-Scale ResNet on Image Representation

Big Data Mining and Analytics 2024年第3期7卷 668-681页

作者： Qiaozhen Meng Yinuo Lyu Xiaoqing Peng Junhai Xu Jijun Tang Fei Guo School of Computer Science and Technology College of Intelligence and ComputingTianjin UniversityTianjin 300350China Aeronautical Information Service Center of the Civil Aviation Administration of China(AISC.ATMB.CAAC) Beijing 100015China Center for Medical Genetics&Hunan Key Laboratory of Medical Genetics School of Life SciencesCentral South UniversityChangsha 410038China Shenzhen Institute of Advanced Technology Chinese Academy of SciencesShenzhen 518000China School of Computer Science and Engineering Central South UniversityChangsha 410083China

Prediction of enhancer-promoter interactions(EPIs)is key to regulating gene expression and diagnosing genetic *** to limited resolution,biological experiments perform not as well as expected while precisely identifying specific interactions,giving rise to computational biology *** EPI predictors have been developed,but their prediction accuracy still needs to be ***,we design a new model named EPIMR to identify enhancer-promoter ***,Hilbert Curve is utilized to represent sequences to images to preserve the position and spatial ***,a multi-scale residual neural network(ResNet)is used to learn the distinguishing features of different abstraction ***,matching heuristics are adopted to concatenate the learned features of enhancers and promoters,which pays attention to their potential interaction *** results on six cell lines indicate that EPIMR performs better than existing methods,with higher area under the precision-recall curve(AUPR)and area under the receiver operating characteristic(AUROC)results on benchmark and under-sampling ***,our model is pre-trained on all cell lines,which improves not only the transferability of cross-cell line prediction,but also cell line-specific prediction *** conclusion,our method serves as a valuable technical tool for predicting enhancer-promoter interactions,contributing to the understanding of gene transcription *** code and results are available at https://***/guofei-tju/EPIMR.

关键词： enhancer-promoter interactions Hilbert Curve multi-scale residual neural network(ResNet)

来源：评论

学校读者我要写书评

暂无评论

Lattice-based group encryptions with only one trapdoor

引用

science China(information sciences) 2022年第5期65卷 205-222页

作者： Jing PAN Jiang ZHANG Fangguo ZHANG Xiaofeng CHEN Willy SUSILO State Key Laboratory of Integrated Service Networks(ISN) Xidian University State Key Laboratory of Cryptology School of Computer Science and Engineering Sun Yat-sen University Guangdong Key Laboratory of Information Security Institute of Cybersecurity and Cryptology School of Computing and Information TechnologyUniversity of Wollongong

Group encryption(GE), the encryption analog of group signatures, is a fundamental primitive that offers a privacy-preserving service for a specific receiver concealed within a group of certified *** other cryptographic primitives, GE constructions are always considered relative to the potential danger of quantum computations. The only existing lattice-based variant appeared in the work of Libert et al.(Asiacrypt'16). Despite its non-trivial achievement, the construction suffers in terms of efficiency due to the extensive use of lattice trapdoors. In this paper, we develop an integrated zero-knowledge argument system that is friendly to both accumulated values and hidden matrices and supports efficient designs from *** on this system, we propose efficiency enhancing GE where only group users are required to possess the lattice trapdoors and the other parties are not. In particular, we utilize lattice-based cryptographic accumulators to confirm prospective group members and use the dual Regev encryption scheme to provide privacy for ciphertext recipients. These modifications significantly increase GE efficiency. In addition, under the intractability assumptions of the standard lattice problems, we prove the security of the proposed scheme in the standard model(assuming interaction during the proof phase), which retains the strongest level of security as the only currently available candidate.

关键词： Lattice cryptography lattice cryptography group encryptions lattice trapdoors accumulators zero-knowledge

来源：评论

学校读者我要写书评

暂无评论

A survey on deep learning for polyp segmentation: techniques, challenges and future trends

引用

Visual Intelligence 2025年第1期3卷 1-20页

作者： Mei, Jiaxin Zhou, Tao Huang, Kaiwen Zhang, Yizhe Zhou, Yi Wu, Ye Fu, Huazhu PCA Lab Key Laboratory of Intelligent Perception and Systems for High-Dimensional Information of Ministry of Education Nanjing China School of Computer Science and Engineering Nanjing University of Science and Technology Nanjing China School of Computer Science and Engineering Southeast University Nanjing China Institute of High Performance Computing A*STAR Singapore Singapore

Early detection and assessment of polyps play a crucial role in the prevention and treatment of colorectal cancer (CRC). Polyp segmentation provides an effective solution to assist clinicians in accurately locating and segmenting polyp regions. In the past, people often relied on manually extracted lower-level features such as color, texture, and shape, which often had problems capturing global context and lacked robustness to complex scenarios. With the advent of deep learning, more and more medical image segmentation algorithms based on deep learning networks have emerged, making significant progress in the field. This paper provides a comprehensive review of polyp segmentation algorithms. We first review some traditional algorithms based on manually extracted features and deep segmentation algorithms, and then describe benchmark datasets related to the topic. Specifically, we carry out a comprehensive evaluation of recent deep learning models and results based on polyp size, taking into account the focus of research topics and differences in network structures. Finally, we discuss the challenges of polyp segmentation and future trends in the field. © The Author(s) 2025.

关键词： Comprehensive evaluation Deep learning Medical imaging Polyp segmentation

来源：评论

学校读者我要写书评

暂无评论

CroPrompt: Cross-task Interactive Prompting for Zero-shot Spoken Language Understanding

CroPrompt: Cross-task Interactive Prompting for Zero-shot Sp...

引用

2025 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2025

作者： Qin, Libo Wei, Fuxuan Chen, Qiguang Zhou, Jingxuan Huang, Shijue Si, Jiasheng Lu, Wenpeng Che, Wanxiang School of Computer Science and Engineering Central South University China Key Laboratory of Data Intelligence and Advanced Computing in Provincial Universities Soochow University China Research Center for Social Computing and Information Retrieval Harbin Institute of Technology China Harbin Institute of Technology Shenzhen China Key Laboratory of Computing Power Network and Information Security Ministry of Education Qilu University of Technology Shandong Academy of Sciences China

ISBN: (纸本)9798350368741

Slot filling and intent detection are two highly correlated tasks in spoken language understanding (SLU). Recent SLU research attempts to explore zero-shot prompting techniques in large language models to alleviate the data scarcity problem. Nevertheless, the existing prompting work ignores the cross-task interaction information for SLU, which leads to sub-optimal performance. To solve this problem, we present the pioneering work of Cross-task Interactive Prompting (CroPrompt) for SLU, which enables the model to interactively leverage the information exchange across the correlated tasks in SLU. Additionally, we further introduce a multi-task self-consistency mechanism to mitigate the error propagation caused by the intent information injection. We conduct extensive experiments on the standard SLU benchmark and the results reveal that CroPrompt consistently outperforms the existing prompting approaches. In addition, the multi-task self-consistency mechanism can effectively ease the error propagation issue, thereby enhancing the performance. We hope this work can inspire more research on cross-task prompting for SLU. © 2025 IEEE.

关键词：

来源：评论

学校读者我要写书评

暂无评论

MACRO: Incentivizing Multi-Leader Game-Based Pareto-Efficient Crowdsourcing for Video Analytics 40

MACRO: Incentivizing Multi-Leader Game-Based Pareto-Efficien...

引用

40th IEEE International Conference on Data Engineering, ICDE 2024

作者： Chen, Yu Zhang, Sheng Zhou, Ziying Wang, Xiaokun Liang, Yu Chen, Ning Yan, Yuting Xiao, Mingjun Wu, Jie Qian, Zhuzhong Xu, Harry Nanjing University State Key Lab. for Novel Software Technology China School of Systems Science Beijing Normal University China School of Computer and Electronic Information School of Artificial Intelligence Nanjing Normal University China School of Computer Science and Technology University of Science and Technology of China China Temple University Center for Networked Computing United States Samueli School of Engineering University of California Computer Science Department Los Angeles United States

ISBN: (纸本)9798350317152

In recent years, many crowdsourcing platforms have emerged, using the resources of recruited workers to perform diverse outsourcing tasks, where the video analytics attracts much attention due to its practical implications. For maximum profits, platforms carefully choose the workers and determine the video analytics configurations to ensure accuracy;meanwhile, workers possess the flexibility to tailor the configurations for their indivi-dual gains, which makes it hard for platforms to optimize their profits considering the platform-worker conflicts. In this paper, we design an incentive mechanism for Multi-leader game-based video Analytics upon CROwdsourcing, named MACRO, to over-come the above situation. Under that mechanism, we first formu-late the utility optimization problems for platforms and workers, respectively. We then propose a dual ascent-based method to op-timally determine the video analytics configurations for a multi-platform game, ensuring Pareto efficiency. Moreover, in the context of a multi-leader game involving platform-worker conflicts, we design an incentive function with its incentive factor update strategy and propose an ADMM-based approach for maximizing incentives that motivate workers to contribute to the platforms' profits. Rigorous proofs demonstrate the linear convergence of the MACRO to the multi-leader Stackelberg equilibrium. Trace-driven experiments show that MACRO improves the Pareto efficiency by 26.3%, outperforming other approaches. © 2024 IEEE.

关键词： Profitability

来源：评论

学校读者我要写书评

暂无评论

A Modified Feature Selection Algorithm for Intrusion Detection System Based on Student Psychology-Based Optimization with Explainable AI

A Modified Feature Selection Algorithm for Intrusion Detecti...

引用

2023 IEEE Asia-Pacific Conference on Computer science and Data Engineering, CSDE 2023

作者： Evan, Md. Raisul Islam Motiur Rahman, Mohammad Tanzir Mehedi, Sk. Nazib, Abdullah Islam, Rafiqul Rahman, Ziaur Mawlana Bhashani Science and Technology University Computer Science and Engineering Tangail1902 Bangladesh Mawlana Bhashani Science and Technology University Information and Communication Technology Tangail1902 Bangladesh School of Electrical Engineering and Robotics Queensland University of Technology BrisbaneQLD Australia School of Computing Mathematics and Engineering Charles Sturt University Bathurst Australia School of Computer Science Queensland University of Technology BrisbaneQLD Australia

ISBN: (纸本)9798350341072

Intrusion Detection Systems (IDS) play a vital role in safeguarding network security by monitoring and identifying both normal and intrusive activities. However, constructing an effective IDS from extensive datasets poses a significant challenge. The process of feature selection from these datasets is crucial, as it not only enhances accuracy but also reduces the time required to build the model. In this paper, we employ a modified version of the Student Psychology-Based Optimization (SPBO) algorithm to identify the most relevant features set for IDS. Following feature selection, various machine learning algorithms are employed for multi-level classification. To evaluate our approach, we utilize the NLS-KDD dataset. Our proposed algorithm surpasses several state-of-the-art feature selection methods in terms of accuracy, precision, recall, Fl-score, and finally faster convergence. Specif-ically, our proposed bio-inspired SPBO algorithm combined with the Random Forest (RF) algorithm achieves a comparable accuracy rate of 99.81 %. To ensure better comprehensibility, we employ two widely used Explainable Artificial Intelligence (XAI) techniques: SHapley Additive exPlanations (SHAP) and Local In-terpretable Model-Agnostic Explanations (LIME). These methods provide insightful explanations for our model's decisions, further enhancing its interpretability. © 2023 IEEE.

关键词： Feature Selection

来源：评论

学校读者我要写书评

暂无评论

Face Identification Method Using Scale Invariant Feature Transform Method for Criminal Management System 4

Face Identification Method Using Scale Invariant Feature Tra...

引用

4th International Conference on Smart Electronics and Communication, ICOSEC 2023

作者： Maragatharajan, M. Sudhan, M.B. Sathishkumar, L. Suprakash, S. Balakannan, S.P. School of Computing Science and Engineering Vit Bhopal University Madhya Pradesh Kothrikalan India Srm Institute of Science and Technology Chennai India Kalasalingam Academy of Research and Education Department of Information Technology Tamilnadu Krishnankoil India

ISBN: (纸本)9798350300888

Due to growing crime rates in India, big data for criminal records has increased more than ever. The increased database has resulted in slower detection and identification of pre-existing criminals for related crimes. In this new cyber world, India's security is more important than ever. Numerous organized criminal operations are anticipated, some of which might aim to undermine our law enforcement and defense organizations. technology becomes of utmost importance in such a situation. Video analytics is one use of technology used in law enforcement, and video analytics contains many different parts. Finding patterns in different videos is one of the elements. The task here is to analyze a photograph of a particular person and see its resemblance with respect to the related criminal. Now, if the aforementioned technology is applied to thousands of movies compiled from hundreds of intelligence sources, with cross-checks across all recordings, highly important information on trends in any kind of organized crime can be discovered. This study has used scale invariant feature transform (SITF) to compare the features of images. Therefore, a system where the details of the criminal can be observed on a descriptive basis by facial recognition will be of great use for the investigation. © 2023 IEEE.

关键词： Crime

来源：评论

学校读者我要写书评

暂无评论

Retrievable Image Encryption Based on Adaptive Block Compressed Sensing

Retrievable Image Encryption Based on Adaptive Block Compres...

引用

International Conference on Image, Vision and Intelligent Systems, ICIVIS 2023

作者： Fang, Jiuchuan Zhao, Hui Zheng, Mingwen Wang, Zijian Niu, Sijie Gao, Xizhan Shandong Provincial Key Laboratory of Network Based Intelligent Computing School of Information Science and Engineering University of Jinan Jinan250022 China School of Mathematics and Statistics Shandong University of Technology Zibo255000 China

ISBN: (纸本)9789819708543

The increasing popularity of the internet and mobile devices, cloud storage has emerged as a popular storage method. However, because data carries a lot of information, untrustworthy servers can access users’ private information. This paper introduces an approach based on adaptive block compressed sensing and hash retrieval. The objective of this paper is to reduce the storage space required by encryption algorithms while simultaneously improving interaction efficiency. Firstly, compress images using block compressed sensing, and use BCS-SPL algorithm to reconstruct. And curves are fitted based on the gradient features and root mean square values of the blocks. Then, the sub-rates are allocated based on the gradient characteristics of the image using a fitting curve, and then the image is compressed. Next, encrypt the image and save it. The features of the queried image are used for retrieval, the results are decrypted, and then reconstructed. Finally, through a series of security analyses, the method is effective. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2024.

关键词： Cloud storage

来源：评论

学校读者我要写书评

暂无评论

Air Quality Prediction Using Adaboost

Air Quality Prediction Using Adaboost

引用

2023 Intelligent computing and Control for Engineering and Business Systems, ICCEBS 2023

作者： Thamizhselvi, D. Kasi, Baskar Kamalakkannan, K. Bharath, S. Gowtham, S. Kishore, M.A. Sri Sairam Engineering College Department of Information Technology Chennai India School of Computing Science and Engineering Galgotias University Greater Noida India Srm Institute of Science and Technology Department of Automobile Engineering Chennai India

ISBN: (纸本)9798350394580

The survival of human beings is contingent upon the presence of air. The detrimental impact of contemporary culture on air quality is seen in nearly all regions. Pollution of the environment is caused by activities in the industrial, transit, and residential sectors. The monitoring and prediction of air quality have become imperative in contemporary times, particularly in developing nations such as India. The most efficient techniques for assessing contemporary threats are thought to be prediction technologies based on machine learning. This study looks at a six-year-old dataset with a particular emphasis on air pollution data gathered from 23 Indian cities. The purpose of this investigation is to gather data and create forecasting models for air pollution. The Gaussian Naive Bayes Algorithm was used to perform substantial pre-processing on the dataset, and correlation analysis was used to identify standout features. Exploratory data analysis is used to identify specific contaminants whose presence has an immediate effect on the air quality index and uncover hidden patterns in datasets. Resampling is used to remedy the imbalance in the data problem. The AdaBoost algorithm exhibits the highest possible degree of precision. © 2023 IEEE.

关键词： AdaBoost Data Imbalance Gaussian Naive Bayes Latent Patterns

来源：评论

学校读者我要写书评

暂无评论

A Web 3.0 Integrated Blockchain Enabled Access System Augmented by Meta-Heuristic Cognitive Learning Framework for Mitigating Threats in IoT Enabled Consumer Electronic Devices

引用

IEEE Transactions on Consumer Electronics 2025年第1期71卷 1201-1210页

作者： Aruna, Etikala Sahayadhas, Arun Kalpana, Ponugoti Khan, Surbhi B. Quasim, Tabrez Almusharrf, Ahlam VELS Institute of Science Technology and Advanced Studies Department of Computer Science and Engineering Pallavaram Chennai India AVN Institute of Engineering and Technology Department of Computer Science and Engineering Hyderabad501510 India University of Salford School of Science Engineering and Environment SalfordM5 4WT United Kingdom Chandigarh University University Centre for Research and Development Mohali India Chitkara University Centre for Research Impact and Outcome India University of Bisha College of Computing and Information Technology Department of Computer Science and Artificial Intelligence Bisha Saudi Arabia King Khalid Univeersity Saudi Arabia

Consumer Electronic Devices have become an open network model because of the infusion of the Internet of Things (IoT) and other communication technologies such as 5G/6G. Though these devices have provided the high-end sophistication even to common person, but it has proved its darker side by triggering more security breaches and privacy problems. Hence, securing and authenticating these Internet enabled consumer devices has become a probable issue to be solved for safer and secured communication. Therefore, this paper presents a novel fusion of Web 3.0-based Blockchain (WBC) and Deep learning (DL) technique for securing consumer electronic devices in an IoT ecosystem. The proposed framework k(MTD-BCAM) is devised into two components: Multiple-Threat Detection(MTD) and Access Management Mechanism(AMM). In the first component, a DL model is applied for threat detection, whereas WBC is meant for an efficient authentication process. Furthermore, a novel residual fast-gated recurrent neural network is proposed. To reduce the complexity, the komodo Mlipir optimization (KMO) approach is used to tune the hyper-parameters of the network. The comprehensive experimental outcome study of the proposed approach employs NSL-KDD datasets in which the distinct metrics of both DL and Blockchain (BC) are measured and analyzed. Results demonstrated the superior accuracy of the model by achieving 99.78% with less computational time and higher transaction speed. Additionally, the statistical validation and security strength of the model are also analyzed and examined with the varied state-of-art models. © 2025 IEEE. All rights reserved.

关键词： Recurrent neural networks

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：