检索结果-内蒙古大学图书馆

arXiv 2024年

作者： Singh, Hrishikesh Sharma, Aarti Pant, Millie Department of Applied Mathematics & Scientific Computing Indian Institute of Technology - Roorkee India

In the era of evolving artificial intelligence, machines are increasingly emulating human-like capabilities, including visual perception and linguistic expression. Image captioning stands at the intersection of these domains, enabling machines to interpret visual content and generate descriptive text. This paper provides a thorough review of image captioning techniques, catering to individuals entering the field of machine learning who seek a comprehensive understanding of available options, from foundational methods to state-of-the-art approaches. Beginning with an exploration of primitive architectures, the review traces the evolution of image captioning models to the latest cutting-edge solutions. By dissecting the components of these architectures, readers gain insights into the underlying mechanisms and can select suitable approaches tailored to specific problem requirements without duplicating efforts. The paper also delves into the application of image captioning in the medical domain, illuminating its significance in various real-world scenarios. Furthermore, the review offers guidance on evaluating the performance of image captioning systems, highlighting key metrics for assessment. By synthesizing theoretical concepts with practical application, this paper equips readers with the knowledge needed to navigate the complex landscape of image captioning and harness its potential for diverse applications in machine learning and beyond. Copyright © 2024, The Authors. All rights reserved.

关键词： Deep learning

来源：评论

学校读者我要写书评

暂无评论

Classification Framework for Autoimmune Liver Disease using Machine Learning and Deep Learning Techniques

Classification Framework for Autoimmune Liver Disease using ...

引用

technology Innovation Management and Engineering Science International Conference (TIMES-iCON)

作者： E. Ben George G. Jeba Rosline Abraham Varghese Rajesh D. Gnana Department of Information Technology College of Computing and Information Sciences University of Technology and Applied Sciences Muscat Sultanate of Oman Department of Information Technology College of Computing and Information Sciences University of Technology and Applied Sciences Al Mussanah Sultanate of Oman

ISBN: (数字)9798350362602

ISBN: (纸本)9798350362619

Autoimmune diseases affect people of all ages, genders, and ethnic backgrounds, making it one of the most widespread health issues we face today. These disorders occur when the immune system mistakenly targets healthy cells within the body and causing a wide range of potentially life-threatening diseases. There are over 80 recognized autoimmune disorders, including type-1 diabetes, rheumatoid arthritis, Graves’ disease, alopecia, celiac disease, and autoimmune liver disease(AiLD). Autoimmune liver disease is one such chronic disorder characterized by liver inflammation and damage. If not detected on time and untreated within the given time, it might lead to liver cirrhosis or liver cancer. Identifying and treating autoimmune liver disease is a complex and time consuming process which involves symptom analysis, laboratory tests, and clinical assessments. Detecting the disease can be more challenging because many symptoms of autoimmune liver diseases overlap with those of other autoimmune conditions. As a result, there is a need for an optimized, accurate, efficient, and cost-effective methodology to identify the disease without extensive and invasive procedures such as organ transplants. This paper introduces an innovative approach that employs the power of supervised machine learning and deep learning techniques to predict the onset of autoimmune liver diseases in advance. The machine learning model is constructed using a comprehensive dataset comprising the complete medical history of patients, physical examination records, histopathology and genetic data. Along with the machine learning model, the deep learning model is also developed for liver MRI images of various patients to make accurate prediction. This system provides a second opinion for medical professionals that may avoid the need for extensive and invasive medical procedures, including biopsies, and ultimately improve patient care and outcomes.

关键词： Deep learning Accuracy Liver diseases Biological system modeling Predictive models Autoimmune diseases Task analysis

来源：评论

学校读者我要写书评

暂无评论

Skills Extraction from Entities in Arabic CVs 5

Skills Extraction from Entities in Arabic CVs

引用

5th IEEE International Conference on Cognitive Machine Intelligence, CogMI 2023

作者： Sadiq, Mohamed Abdul Karim Shanmugam, Thirumurugan Al-Fannah, Nasser Mohammed Kausar, Firdous College of Computing and Information Science University of Technology and Applied Sciences Suhar Oman Fisk University Department of Mathematics and Computer Science NashvilleTN United States

ISBN: (纸本)9798350323832

Named Entity Recognition in low resource languages such as Arabic is relatively less accurate. Despite this, many applications attempt query-document matching, which necessitates entity resolution. We explore certain possible solutions in this regard to enhance the matching process. A system is proposed to build a sub graph of terms in the Knowledge Graph. These terms specifically deal with skills in the Arabic documents related to the Human Resources domain. The accuracy measures of skills extraction using two different Language Models viz., SparkNLP and Hatmi are presented. The results are promising and have scope for improvement in entity resolution. © 2023 IEEE.

关键词： Knowledge graph

来源：评论

学校读者我要写书评

暂无评论

A Comparative Study of Recent 2D Human Pose Estimation Methods 6

A Comparative Study of Recent 2D Human Pose Estimation Metho...

引用

6th International Conference on computing and Informatics, ICCI 2024

作者： Elshami, Nada E. Salah, Ahmad Mohsen, Heba Department of Computer Science Faculty of Computers and Information Technology Future University in Egypt New Cairo Egypt College of Computing and Information Sciences University of Technology and Applied Sciences Ibri Oman College of Computers and Informatics Zagazig University Zagazig Egypt

ISBN: (纸本)9798350373875

Human Pose Estimation is a computer vision technique utilized in various fields such as healthcare, security, and sports to detect the pose of single or multi-person utilizing various machine learning and deep learning methodologies. These methodologies can be applied in a variety of ways, including top-down and bottom-up approaches. Furthermore, this technique can be used with images or videos, as well as in 2D or 3D domains. This paper delves into recent advancements in deep learning methods used for estimating the pose, covering commonly employed datasets, evaluation metrics, and the challenges encountered in this field. It aims to assist new researchers in getting familiar with this technique based on work published between 2019 and 2023, with a spotlight on 2D and omitting 3D. © 2024 IEEE.

关键词： Deep learning

来源：评论

学校读者我要写书评

暂无评论

FusionMamba: dynamic feature enhancement for multimodal image fusion with Mamba

引用

Visual Intelligence 2024年第1期2卷 1-18页

作者： Xie, Xinyu Cui, Yawen Tan, Tao Zheng, Xubin Yu, Zitong School of Computing and Information Technology Great Bay University Guangdong Dongguan 523000 China Faculty of Applied Sciences Macao Polytechnic University Macao Department of Electrical and Electronic Engineering The Hong Kong Polytechnic University Hong Kong

Multimodal image fusion aims to integrate information from different imaging techniques to produce a comprehensive, detail-rich single image for downstream vision tasks. Existing methods based on local convolutional neural networks (CNNs) struggle to capture global features efficiently, while Transformer-based models are computationally expensive, although they excel at global modeling. Mamba addresses these limitations by leveraging selective structured state space models (S4) to effectively handle long-range dependencies while maintaining linear complexity. In this paper, we propose FusionMamba, a novel dynamic feature enhancement framework that aims to overcome the challenges faced by CNNs and Vision Transformers (ViTs) in computer vision tasks. The framework improves the visual state-space model Mamba by integrating dynamic convolution and channel attention mechanisms, which not only retains its powerful global feature modeling capability, but also greatly reduces redundancy and enhances the expressiveness of local features. In addition, we have developed a new module called the dynamic feature fusion module (DFFM). It combines the dynamic feature enhancement module (DFEM) for texture enhancement and disparity perception with the cross-modal fusion Mamba module (CMFM), which focuses on enhancing the inter-modal correlation while suppressing redundant information. Experiments show that FusionMamba achieves state-of-the-art performance in a variety of multimodal image fusion tasks as well as downstream experiments, demonstrating its broad applicability and superiority. © The Author(s) 2024.

关键词： Feature enhancement Image fusion Mamba Multimodal

来源：评论

学校读者我要写书评

暂无评论

Exploring Drug Recommendations Through Sentiment Analysis: A Machine Learning Perspective

Exploring Drug Recommendations Through Sentiment Analysis: A...

引用

Electrical and Computer Engineering Researches (ICECER), International Conference on

作者： Imran Qureshi Zahra Nasser Hashil Alsiyabi Burhanuddin Mohammad S.G.M Shadab Mohammed Irfan University of Technology and Applied Science College of Computing and Information Sciences AlMusanna Oman

ISBN: (数字)9798331539733

ISBN: (纸本)9798331539740

Current drug recommendation systems often rely on clinical trials and general guidelines, which may not fully account for individual patient experiences. This study addresses the gap by leveraging sentiment analysis of patient reviews to enhance drug recommendations. The primary objective is to develop a machine learning-based drug recommendation system that incorporates sentiment from drug reviews, improving the personalization of drug recommendations. Using a dataset from UCI ML repository with over 215,000 instances, we apply sentiment analysis and various feature extraction techniques, including TF-IDF and Word2Vec, alongside machine learning classifiers to predict drug efficacy for conditions such as acne, hypertension, and depression. By integrating sentiment-based insights, this approach aims to provide more tailored recommendations for individual patients. Our results show that models like Linear SVC (TF-IDF) and Random Forest (manual features) outperform traditional methods, achieving up to 93% accuracy. These findings demonstrate the potential of sentiment analysis to enhance the relevance and accuracy of drug recommendations. This research contributes to the growing field of personalized healthcare by introducing a data-driven framework that captures patient sentiment, offering more accurate and individualized medication guidance.

关键词： Drugs Sentiment analysis Accuracy Reviews Static VAr compensators Medical services Manuals Ensemble learning Recommender systems Random forests

来源：评论

学校读者我要写书评

暂无评论

Optimizing ASReview Simulations:A generic Multiprocessing Solution for“Light-data'and“Heavy-data'Users

引用

Data Intelligence 2024年第2期6卷 320-343页

作者： Sergei Romanov Abel Soares Siqueira Jonathan de Bruin Jelle Teijema Laura Hofstee Rens van de Schoot Applied Data Science Department of Information and Computing ScienceFaculty of ScienceUtrecht University Netherlands eScienceCenter AmsterdamNL Department of Research and Data Management Services Information Technology ServicesUtrecht UniversityUtrechtthe Netherlands Department of Methodology and Statistics Faculty of Social and Behavioral SciencesUtrecht University

Active learning can be used for optimizing and speeding up the screening phase of systematic *** simulation studies mimicking the screening process can be used to test the performance of different machine-learning models or to study the impact of different training *** paper presents an architecture design withamultiprocessing computational strategyforrunningmanysuch simulation studiesinparallel,using the ASReview Makita workflow generator and Kubernetes software for deployment with cloud *** provide a technical explanation of the proposed cloud architecture and its *** addition to that,we conducted 1140 simulations investigating the computational time using various numbers of CPUs and RAM *** analysis demonstrates the degree to which simulations can be accelerated with multiprocessing computing *** parallel computation strategy and the architecture design that was developed in the present paper can contribute to future research with more optimal simulation time and,at the same time,ensure the safe completion of the needed processes.

关键词： Active learning Systematic review Simulation study Multiprocessing Cloud architecture

来源：评论

学校读者我要写书评

暂无评论

A recurrent neural network architecture for android mobile data analysis for detecting malware infected data

引用

Soft computing 2024年第21期28卷 12917-12928页

作者： Murugan, Prabhu Manimaran, A. Sundar, Ramesh Dakshinamoorthy, Prabakar Rajaram, Gnanajeyaraman Garg, Shruti Department of ECE Saveetha School of Engineering Simats Chennai602105 India Department of Artificial Intelligence and Machine Learning Saveetha School of Engineering Simats Chennai602105 India Department of Netwoking and Communication School of Computing SRM Institute of Science and Technology SRM Nagar Kattankulathur India Department of Data Science and Business System School of Computing SRM Institute of Science and Technology SRM Nagar Kattankulathur India Department of Applied Machine Learning Saveetha School of Engineering Saveetha Institute of Medical and Technological Sciences Tamilnadu Chennai India Birla Institute of Technology Mesra Ranchi India

One of the latest modern communication devices is a mobile device seriously affected by multiple malware. Malware is a virus software installed automatically by hackers on various computing devices. Malware corrupts the system software, kills *.exe files, and tries to access user-sensitive data from the device. Around the world, 80% of people use smartphones, and 70% use Android phones. All Android phones are installed with anti-virus software but not used properly. Thus, it creates a pathway for malware to enter Android mobile devices and deploy vulnerabilities. Several earlier methods have focused on using the malware detection method by analyzing static data, which cannot identify the dynamic behavior of the malware. Thus, this paper implemented one of the deep learning algorithms, RNN, for identifying and detecting malware activities in Android mobile devices statically and dynamically. An IoT network architecture is proposed for experimenting with the RNN algorithms verifying the performance. The results of the RNN method show that it is better than other models in malware detection in mobile devices. © The Author(s), under exclusive licence to Springer-Verlag GmbH Germany, part of Springer Nature 2024.

关键词： Android malware

来源：评论

学校读者我要写书评

暂无评论

An Efficient Distributed Energy and Consumption Method for Ensuring Wireless Sensor Network (WSN) Coverage Using the Firefly Algorithm 4th

An Efficient Distributed Energy and Consumption Method for E...

引用

4th International Conference on Innovative Product Design and Intelligent Manufacturing System, IPDIMS 2022

作者： Uma, K. Ramesh Kumar, C. Shanmugam, Thirumurugan School of Information Technology Vellore Institute of Technology Tamil Nadu Vellore India School of Computing Science and Engineering Galgotias University UP Greater Noida India Department of Information Technology University of Technology and Applied Sciences Suhar Muscat Oman

ISBN: (纸本)9789819916641

In this article, we are talking about the wireless sensor network sub-field of a body area sensor network. The establishment of the wireless body area sensor network has taken place after the refinement of the wireless sensor network was underway. We have introduced an innovative energy-efficient routing system that incorporates a fuzzy logic also bio-inspired firefly algorithm, which considers these limitations. Together, our proposal of correlation region selection then bio-stimulated firefly algorithm centered NBSs use fuzzy logic-based correlations to minimize energy transmission, while our joint firefly NBS proposal makes a valuable contribution by cutting energy consumption. To identify and select enriched regions that contain high energy and detect information, our suggested fuzzy logic-based correlation region selection mechanism is configured to favor those regions. We, however, suggested a bio-stimulated firefly algorithm fitness function for the collection of NBSs. To prevent exhaustion of valuable transmission, the fitness function accounts for the diffusion antiquity and residual energy for NBSs. To confirm the in-depth influence of our suggested scheme, we deportment numerous simulations exhausting the Nano-SIM tool to measure the various aspects involved in conserving energy resources, sinking endwise deferment, and increasing packet delivery ratio. We found that by using an optimized selection for correlated regions besides NBSs in refining network lifespan and packet delivery ratio although dipping the typical endwise delay, our proposed system accomplishes the significant task of determining which regions and NBSs are more important to the network and allows for the lower end-to-end delay. © 2023, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

关键词： Fuzzy logic

来源：评论

学校读者我要写书评

暂无评论

InputJump: Augmented reality-facilitated cross-device input fusion based on spatial and semantic information

引用

虚拟现实与智能硬件（中英文） 2024年第6期6卷 502-526页

作者： Xin ZENG Xiaoyu WANG Tengxiang ZHANG Yukang YAN Yiqiang CHEN Institute of Computing Technology Chinese Academy of SciencesBeijing 100190China University of Chinese Academy of Sciences Beijing 101408China Goertek Inc. Bejing 261031China The Hong Kong University of Science and Technology Hong Kong 999077China Department of Computer Science Hajim School of Engineering and Applied SciencesUniversity of RochesterRochesterNY 14627US

The proliferation of computing devices requires seamless cross-device *** reality(AR)headsets can facilitate interactions with existing computers owing to their user-centered views and natural *** this study,we propose InputJump,a user-centered cross-device input fusion method that maps multi-modal cross-device inputs to interactive elements on graphical *** input jump calculates the spatial coordinates of the input target positions and the interactive elements within the coordinate system of the AR *** also extracts semantic descriptions of inputs and elements using large language models(LLMs).Two types of information from different inputs(e.g.,gaze,gesture,mouse,and keyboard)were fused to map onto an interactive *** proposed method is explained in detail and implemented on both an AR headset and a desktop *** then conducted a user study and extensive simulations to validate our proposed *** results showed that InputJump can accurately associate a fused input with the target interactive element,enabling a more natural and flexible interaction experience.

关键词： Mixed reality Spatial Semantic Large language models

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：