检索结果-内蒙古大学图书馆

Image entropy equalization: A novel preprocessing technique for image recognition tasks

INFORMATION SCIENCES 2023年第1期647卷

作者： Hayashi, Toshitaka Cimr, Dalibor Fujita, Hamido Cimler, Richard Univ Hradec Kralove Fac Sci Hradec Kralove Czech Republic Univ Teknol Malaysia Japan Int Inst Technol MJIIT Kuala Lumpur 54100 Malaysia Univ Granada DaSCI Andalusian Inst Data Sci & Computat Intellig Granada Spain Iwate Prefectural Univ Reg Res Ctr Takizawa Japan Kotorizawa 2-27-5 Morioka Iwate 0200104 Japan

Image entropy is the metric used to represent a complexity of an image. This study considers the hypothesis that image entropy differences affect machine learning algorithms' performance. This paper proposes a novel preprocessing technique, image entropy equalization, to delete the image entropy differences. The goal is to transform all images into the same entropy. Such a process is implemented by editing all images into the same histogram. Image entropy equalization is evaluated by comparing the original and equalized images in various machine learning tasks. The main advantage of image entropy equalization is to improve the AUC score for one-class autoencoder (OCAE). This result gives a new hypothesis that using image entropy equalization could improve various studies using autoencoder (AE). In addition, the proposed method shows fair results for classification and regression tasks. On the other hand, the main challenges are that the equalization process depends on a reference histogram and is affected by diverse backgrounds.

关键词： Image entropy auto encoder Preprocessing One-class classification

来源：评论

学校读者我要写书评

暂无评论

ForestSubtype: a cancer subtype identifying approach based on high-dimensional genomic data and a parallel random forest

引用

BMC BIOINFORMATICS 2023年第1期24卷 289-289页

作者： Luo, Junwei Feng, Yading Wu, Xuyang Li, Ruimin Shi, Jiawei Chang, Wenjing Wang, Junfeng Henan Polytech Univ Sch Software Jiaozuo Peoples R China

Background: Cancer subtype classification is helpful for personalized cancer treat-ment. Although, some approaches have been developed to classifying caner subtype based on high dimensional gene expression data, it is difficult to obtain satisfactory classification results. Meanwhile, some cancers have been well studied and classified to some subtypes, which are adopt by most researchers. Hence, this priori knowledge is significant for further identifying new meaningful ***: In this paper, we present a combined parallel random forest and autoencoder approach for cancer subtype identification based on high dimensional gene expression data, ForestSubtype. ForestSubtype first adopts the parallel RF and the priori knowledge of cancer subtype to train a module and extract significant candidate features. Second, ForestSubtype uses a random forest as the base module and ten parallel random forests to compute each feature weight and rank them separately. Then, the intersection of the features with the larger weights output by the ten parallel random forests is taken as our subsequent candidate features. Third, ForestSubtype uses an autoencoder to condenses the selected features into a two-dimensional data. Fourth, ForestSubtype utilizes k-means++ to obtain new cancer subtype identifica-tion results. In this paper, the breast cancer gene expression data obtained from The Cancer Genome Atlas are used for training and validation, and an independent breast cancer dataset from the Molecular Taxonomy of Breast Cancer International Consor-tium is used for testing. Additionally, we use two other cancer datasets for validat-ing the generalizability of ForestSubtype. ForestSubtype outperforms the other two methods in terms of the distribution of clusters, internal and external metric results. The open-source code is available at https://***/lffyd/***: Our work shows that the combination of high-dimensional gene expression data and parallel ra

关键词： Cancer subtyping Random forest Gene expression data Machine learning auto encoder

来源：评论

学校读者我要写书评

暂无评论

A novel optimization-driven deep learning framework for the detection of DDoS attacks

引用

SCIENTIFIC REPORTS 2024年第1期14卷 1-16页

作者： Batchu, Raj Kumar Bikku, Thulasi Thota, Srinivasarao Seetha, Hari Ayoade, Abayomi Ayotunde Amrita Vishwa Vidyapeetham Amrita Sch Comp Dept Comp Sci & Engn Amaravati 522503 Andhra Prades India Amrita Vishwa Vidyapeetham Amrita Sch Engn Dept Math Amaravati 522503 Andhra Prades India VIT AP Univ Ctr Excellence AI & Robot Amaravati 522241 Andhra Prades India Univ Lagos Fac Sci Dept Math Lagos Nigeria

Distributed denial of service (DDoS) attack is one of the most hazardous assaults in cloud computing or networking. By depleting resources, this attack renders the services unavailable to end users and leads to significant financial and reputational damage. Hence, identifying such threats is crucial to minimize revenue loss, market share, and productivity loss and enhance the brand reputation. In this study, we implemented an effective intrusion detection system using deep learning approach. The suggested framework includes three phases: Data pre-processing, Data balancing, and Classification. First, we prepare the valid data, which is helpful for further processing. Then, we balance the given pre-processed data by Conditional generative adversarial network (CGAN), and as a result, we can minimize the bias towards the majority classes. Finally, we distinguish whether the traffic is attack or benign using a stacked sparse denoising autoencoder (SSDAE) with a firefly-black widow (FA-BW) hybrid optimization algorithm. All these experiments are validated through the CICDDoS2019 dataset and compared with well-received techniques. From these findings, we observed that the proposed strategy detects DDoS attacks significantly more accurately than other approaches. Based on our findings, this study highlights the crucial role played by advanced deep learning techniques and hybrid optimization algorithms in strengthening cybersecurity against DDoS attacks.

关键词： DDoS auto encoder CGAN Firefly Black widow Intrusion detection

来源：评论

学校读者我要写书评

暂无评论

A deep learning model for the accurate prediction of the microstructure performance of hot rolled steel

引用

INTERNATIONAL JOURNAL OF MANUFACTURING RESEARCH 2021年第3期16卷 262-279页

作者： Wang, Bin-bin Song, Yong Wang, Jing Univ Sci & Technol Beijing Engn Technol Res Inst Beijing 100083 Peoples R China

The prediction of microstructure performance can guide the adjustment of parameters during hot rolling. Scholars from all over the world has developed physical metallurgical models of rolling process based on the physical and thermodynamic characteristics of strip steel, but the prediction accuracy of the model is greatly affected by the complex production environment. In recent years, neural network method is used to build the prediction model of organisational performance. However, the prediction accuracy and robustness of the single hidden layer neural network model are poor. Deep learning method is introduced in this paper to establish the prediction model of hot rolling microstructure performance in this paper. The application results show that compared with the traditional model, the prediction accuracy of the hot rolled steels yield strength, tensile strength and elongation increased by 3.46%, 2.35%, and 5.11%, respectively. [Submitted 13 March 2019;Accepted 27 October 2019]

关键词： auto encoder deep learning hot rolled steel microstructure prediction steel properties

来源：评论

学校读者我要写书评

暂无评论

Management of sports injury treatment and radiological data analysis based on enhanced MRI image retrieval using autoencoder-based deep learning

引用

JOURNAL OF RADIATION RESEARCH AND APPLIED SCIENCES 2024年第3期17卷

作者： Zeng, Taisheng Lv, Hongsheng Ashraf, Muhammad Awais Ling, Mengde Chen, Yuping Liu, Yu Chen, Xinwen Li, Yunfeng Huang, Jianlong Quanzhou Normal Univ Fac Math & Comp Sci Quanzhou 362000 Peoples R China Fujian Prov Key Lab Data Intens Comp Quanzhou 362000 Peoples R China Fujian Prov Univ Key Lab Intelligent Comp & Informat Proc Quanzhou 362000 Peoples R China Quanzhou Med Coll Sch Clin Med Quanzhou 362021 Peoples R China Changan Univ Sch Informat Engn Xian 710064 Shaanxi Peoples R China Huaqiao Univ Coll Engn Quanzhou 362021 Peoples R China

Background: The retrieval of magnetic resonance imaging (MRI) data holds paramount importance in clinical settings and sports medicine due to the limitations of conventional methods, such as slow speed, low accuracy, and limited learning capabilities. Enhancing this retrieval process is critical for advancing sports injury diagnostics and treatment outcomes. Overcoming these challenges is vital for improving healthcare practices and sports medicine methodologies. Method: This study investigates the utilization of autoencoders in deep learning to efficiently retrieve MRI data from databases for sports injury diagnosis and treatment, with a focus on the model's ability to be trained with a small amount of labeled data. This research aims to enhance the MRI data retrieval process by leveraging autoencoders, showcasing the potential of deep learning technologies in sports injury diagnostics without the necessity of extensive labeled datasets for training. Results: Findings have showcased the remarkable benefits of this approach for MRI data retrieval tasks, achieving an average accuracy of 99.09%. This signifies the exceptional performance of the technique within this specific domain, demonstrating its effectiveness and reliability in extracting MRI data. Conclusions: This innovative methodology can enhance the management of archival data and diagnostic capabilities of medical images in sports injury contexts, offering an efficient and dependable solution for MRI data retrieval. It not only facilitates rapid clinical diagnosis and sports medicine research but also proposes a convenient approach for medical image file management.

关键词： Sports injury Radiological images auto encoder Attention networks Image retrieval Magnetic resonance imaging

来源：评论

学校读者我要写书评

暂无评论

Improving the Communication and Computation Efficiency of Split Learning for IoT Applications

Improving the Communication and Computation Efficiency of Sp...

引用

IEEE Global Communications Conference (GLOBECOM)

作者： Ayad, Ahmad Renner, Melvin Schmeink, Anke Rhein Westfal TH Aachen ISEK Teaching & Res Area D-52074 Aachen Germany

ISBN: (纸本)9781728181042

Distributed machine learning systems train neural network models by utilizing the devices and resources in the network. One such system that was recently introduced is split learning. It trains a deep neural network collaboratively between the server and the client without sharing the raw data, ensuring private and secure training. Implementing such systems on the edge devices adds computation and communication overhead, which might not suit many edge devices, especially in IoT systems, where resources are limited. In this paper, we introduce a modified split learning system that includes an auto encoder and an adaptive threshold mechanism. The modified system has less communication and computation overhead compared to the original split learning system. The modified system was deployed on an IoT system and the results proved the advantages of the proposed mechanism. The communication overhead and computation overhead were reduced with negligible performance loss.

关键词： Distributed machine learning edge computing IoT split learning auto encoder threshold mechanism

来源：评论

学校读者我要写书评

暂无评论

Deep Learning Analysis on Raw Image Data - Case Study on Holographic Cell Analysis

Deep Learning Analysis on Raw Image Data - Case Study on Hol...

引用

Conference on Medical Imaging - Computer-Aided Diagnosis

作者： Gozes, Gal Ben Baruch, Shani Rotman-Nativ, Noa Roitshtain, Darina Shaked, Natan T. Greenspan, Hayit Tel Aviv Univ Fac Engn Dept Biomed Engn Tel Aviv Israel

ISBN: (纸本)9781510640245

With the advances in deep learning analysis a question arises - if for a robust network learning cycle, the shift between raw data and image-based data, is of importance. In the current work, we start exploring this topic by focusing on the following case study: categorizing cancer cells from holographic images. The problem in the categorization process is the time consumption of the transformation from off-axis image holograms to OPD maps. While there have been attempts for fast transformation, they still take non-negligible time. We propose a novel approach for fast classification that skips the pre-processing of creating OPD maps and directly uses the raw data of holographic images. Our dataset contains two separate image acquisitions of primary cancer cells (SW480) and metastatic cancer cells (SW620) of colorectal adenocarcinoma imaged during flow. We extracted the OPD maps of those cells and used them to train and evaluate a ResNet model to create baseline results. Our convolutional neural network (CNN) model is based on Y-Net approach: during training synthetic OPD images are created from input holograms, using the real OPD maps while simultaneously classifying the cells. During inference time only the classification branch operates to further reduce running time. This approach saves computational time by over 90%.

关键词： Cancer Cell Classification CNN auto encoder Y-Net Cell imaging

来源：评论

学校读者我要写书评

暂无评论

Anomaly Detection on ADS-B Flight Data Using Machine Learning Techniques 13th

Anomaly Detection on ADS-B Flight Data Using Machine Learnin...

引用

13th International Conference on Computational Collective Intelligence (ICCCI)

作者： Tasdelen, Osman Carkacioglu, Levent Toreyin, Behcet Ugur Aselsan Inc TR-06200 Ankara Turkey Istanbul Tech Univ Informat Inst TR-34469 Istanbul Turkey

ISBN: (纸本)9783030880811;9783030880804

With the rapid increase in the number of flights all over the world, the management and control of flight operations has become difficult in recent years. Moreover, the expectations for the aviation sector indicate that this increase will continue in the upcoming years. Therefore, safer and systematic monitoring systems by eliminating the requirement of human-dependent tracking during the air travel of an aircraft and automating the detection of abnormal situations has become a major problem in aviation sector. With the recent advances in artificial intelligence, a safer and systematic tracking system for controlling the airspace by eliminating the need for human-dependent tracking during the flight of aircraft in the air has become possible. In this study, we aimed to create a system that detects and predicts movements to indicate abnormal, dangerous situations in the airspace by monitoring radar flight data using machine learning and deep learning techniques. We applied two different methods, i.e., Proximity Based kNN and auto encoder We used real-life historical radar flight data set which consists of Flight Radar 24 data were converted from ADS-B messages for learning. We created simulation data and used this data for testing and validation for our trained model. Within the scope of this project, we also developed a system to monitor air traffic through radar tracks with our model and present the abnormal situations to the user through a visual interface for decision support. In this visualization, we present the abnormal situations if one of the algorithms labeled as anomaly. Results for both methods have shown that our findings were similar to the real-life predictions.

关键词： Anomaly detection Deep learning Machine learning Proximity based kNN auto encoder Flight control ADS-B Air traffic management

来源：评论

学校读者我要写书评

暂无评论

引用

17th Annual IEEE International Conference on Distributed Computing in Sensor Systems (IEEE DCOSS)

作者： Johnson, Gordon Argyriou, Vasileios Politis, Christos Kingston Univ Comp Sci & Math London England

ISBN: (纸本)9781665439299

The objective is to determine how close an Artificial Intelligence agent is, in comparison to a human player, using only game play images. Identifying Artificial Intelligence agents during game play is typically done through the analysis and collection of bio-metric data, such as keyboard, mouse and other controller interfaces. This document presents a model of an auto encoder architecture, with Long Short Term Memory layers. Gradient and Non-Augmented Images have been evaluated and compared. Three distinct personalities of agents are evaluated, human players, a trained Artificial Intelligence agent, and a basic Artificial Intelligence agent. Through testing the Gradient Image augmentation sets present promising results, with the model successfully identifying the human as the closest similarity to the baseline, followed by the trained Artificial Intelligence agent.

关键词： artificial intelligence auto encoder motion history image similarity metric anomaly detection

来源：评论

学校读者我要写书评

暂无评论

A Practical Tutorial on Graph Neural Networks

引用

ACM COMPUTING SURVEYS 2022年第10S期54卷 205-205页

作者： Ward, Isaac Ronald Joyner, Jack Lickfold, Casey Guo, Yulan Bennamoun, Mohammed ISOLABS 311 Geneva StGlendale Glendale CA 91206 USA Univ Southern Calif Los Angeles CA 90007 USA ISOLABS 35 Stirling Hwy Crawley WA 6009 Australia ISOLABS 14 Yilgarn St Shenton Pk WA 6008 Australia Sun Yat Sen Univ Guangzhou Peoples R China Univ Western Australia 35 Stirling Hwy Crawley WA 6009 Australia Natl Univ Def Technol 137 Yanwachi Changsha 410073 Hunan Peoples R China

Graph neural networks (GNNs) have recently grown in popularity in the field of artificial intelligence (AI) due to their unique ability to ingest relatively unstructured data types as input data. Although some elements of the GNN architecture are conceptually similar in operation to traditional neural networks (and neural network variants), other elements represent a departure from traditional deep learning techniques. This tutorial exposes the power and novelty of GNNs to AI practitioners by collating and presenting details regarding the motivations, concepts, mathematics, and applications of the most common and performant variants of GNNs. Importantly, we present this tutorial concisely, alongside practical examples, thus providing a practical and accessible tutorial on the topic of GNNs.

关键词： Graph neural network tutorial artificial intelligence recurrent convolutional auto encoder decoder machine learning deep learning papers with code theory applications

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：