检索结果-内蒙古大学图书馆

Female autism categorization using CNN based NeuroNet57 and ant colony optimization

computers in Biology and Medicine 2025年 189卷 109926-109926页

作者： Ashraf, Adnan Zhao, Qingjie Bangyal, Waqas Haider Raza, Mudassar Iqbal, Mudassar School of Computer Science and Technology Beijing Institute of Technology Beijing100081 China Dept. of Computer Science Kohsar University Punjab Murree40000 Pakistan Dept. of Computer Science Namal University Mianwali42250 Pakistan Renewable Energy Lab College of Engineering Prince Sultan University Riyadh11586 Saudi Arabia

Autism identification and classification using biomedical medical image analysis has advanced recently. Research shows autistic females have different phenotypic and age-related brain variations than males. Gender-specific hormones and genes affect autistic female brain circuitry, unfortunately, female phenotypic and genotypic data is quite deficient. Since physicians spend much time in assessing autistic females manually. Advanced large-scale deep learning algorithms are in dire need of accurate medical diagnosis. This research proposed a 57-layer CNN architecture called NeuroNet57 that can extract features from fMRI factually. After pre-training on the Brain Tumour dataset, the NeuroNet57 model extracts female phenotypic features from autism brain imagining data exchange (ABIDE)-I+II datasets using T1 modality fMRI scans, resulting in feature matrices of 14372 × 4096 for ABIDE_I and 16168 × 4096 for ABIDE_II. Our model uses ant colony optimization (ACO) to select feature subsets for dimensionality reduction. Further, nine machine learning classifiers are used to categorize females with autism spectrum disorder (ASD) from females with control behavior. The KNN-based fineKNN (FKNN) classifier had 92.21% accuracy on ABIDE-I and 93.49% on ABIDE-II. This proves the effectiveness of our proposed model. © 2025

关键词： Ant colony optimization

来源：评论

学校读者我要写书评

暂无评论

Comprehensive Assistive Language Model (CALM) for Neurodivergent Child Language Acquisition

Comprehensive Assistive Language Model (CALM) for Neurodiver...

引用

Data Science, Agents & Artificial Intelligence (ICDSAAI), International Conference on

作者： Shivranjani Balaji S. Vigneshwari Dept. of Computer Science and Engineering Sathyabama Institute of Science and Technology Chennai India

ISBN: (数字)9798331537555

ISBN: (纸本)9798331537562

The Comprehensive Assistance Language Model–CALM, is a Natural Language Processing framework designed to support consistent language acquisition in children and aid their phonetic, semantic, and syntactic understanding. The model is intentionally designed to engage the child through empathy, task simplification, age-appropriate adaptation and trait-based multimodality. Most neurodivergent children struggle with language acquisition through traditional learning methods and develop a language gap that continues to grow into their adolescence and adulthood. Timely intervention through immersive learning approaches can mitigate or prevent these challenges. CALM integrates two components-a prompt-tuned LLM and a trained transformer model. The inclusion of a user report prepared by the prompt-tuned LLM on traits, age, caretaker concerns and detected condition understanding, increases the response quality and the child’s learning engagement. Layering a transformer model for contextual refinement and grammar correction over the LLM-generated response ensures coherence and improves relevance in answering the child’s query. CALM uses emojis and graphics as part of the generated text to enhance visual learning and also has the option of speech input/output through the incorporated LLM. CALM has achieved meaningful results when measured using evaluation metrics such as accuracy and response time, as well as manually validated characteristics including empathy, bias, engagement, and relevance. CALM effectively simplifies tasks, visualises learning and aligns solutions to the child’s interests.

关键词： Adaptation models Visualization Accuracy Phonetics Transformers Natural language processing Time measurement Time factors Usability Tuning

来源：评论

学校读者我要写书评

暂无评论

FPGA-based Hardware Acceleration of Scalable Convolutional Layers for Deep Learning Inference

FPGA-based Hardware Acceleration of Scalable Convolutional L...

引用

International Conference on Electrical, computer and Communication Engineering (ECCE)

作者： Nahin Ul Sadad Md. Nazrul Islam Mondal Dept. of Computer Science & Engineering Rajshahi University of Engineering & Technology Rajshahi Bangladesh

ISBN: (数字)9798350357509

ISBN: (纸本)9798350357516

Deep learning inference demands high computational efficiency and scalability, which traditional architectures struggle to provide. To address this, we propose an FPGA-accelerated scalable convolutional neural network (CNN) architecture optimized for inference. The design leverages FPGA’s parallelism and reconfigurability, integrating convolution, activation, and reshaping operations. It sequentially processes input data, synchronizing layers via control signals. The convolutional layer extracts spatial features, reshapes the feature map, and passes it to fully connected layers, with the output layer generating class probabilities. Our FPGA implementation achieves a remarkable 63000x speedup over CPU-based inference and a 6000x improvement compared to GPU-based inference with significantly reduced resource (except IO) requirements.

关键词： Deep learning Neural network hardware Convolution Scalability Process control computer architecture Parallel processing Feature extraction Convolutional neural networks Hardware acceleration

来源：评论

学校读者我要写书评

暂无评论

A Hybrid Deep Learning Architecture Using Dense Neural Network and Gated Recurrent Unit for Cloud Intrusion Detection

A Hybrid Deep Learning Architecture Using Dense Neural Netwo...

引用

International Conference on Electrical, computer and Communication Engineering (ECCE)

作者： Iffath Tanjim Moon Md. Nazrul Islam Mondal Dept. of Computer Science & Engineering Rajshahi University of Engineering & Technology Rajshahi Bangladesh

ISBN: (数字)9798350357509

ISBN: (纸本)9798350357516

In this paper, we have used a hybrid deep-learning architecture to detect cloud intrusion. The cloud plays an important role in modern data management, accessibility, and storage. It enables organizations to store vast amounts of data remotely. Due to frequent data sharing between platforms and remote access, the cloud environment is more vulnerable to cyber threats. An intruder can access confidential data due to vulnerabilities including inadequate encryption, absence of authentication, poor access control, and so on. We can efficiently use deep learning approaches to enhance cloud security. In this research, we have developed a hybrid dense neural network (DNN) and gated recurrent unit (GRU)-based architecture to detect cloud intrusion. We have also developed individual DNN and GRU-based architectures. In our proposed model, the DNN layers can capture the nonlinear complex pattern, while GRU can identify the temporal dependencies of network traffic data. In this research work, we have utilized the network portion of the ToN-IoT dataset. For multiclass classification, the hybrid DNN-GRU model has achieved the highest performance, obtaining an accuracy of 98.66%, precision of 98.70%, recall of 98.66%, f1 score of 98.67%, and AUC of 99.98%. Moreover, the performance for detecting nine different attacks has also been evaluated. This case shows the hybrid DNN-GRU model performs well, identifying backdoor and ransomware attacks with 100% accuracy. Thus, this hybrid architecture works more robustly and efficiently to detect cloud intrusion by combining the strengths of both DNN and GRU, thereby strengthening cloud security.

关键词： Deep learning Access control Accuracy Cloud computing security Neural networks Intrusion detection Telecommunication traffic Organizations Encryption Ransomware

来源：评论

学校读者我要写书评

暂无评论

Potato Leaf Disease Classification Using Lightweight CNN-SVM and Real-World Web-Based Tools

Potato Leaf Disease Classification Using Lightweight CNN-SVM...

引用

International Conference on Electrical, computer and Communication Engineering (ECCE)

作者： Hasibul Islam Peyal Md. Nazrul Islam Mondal Dept. of Computer Science & Engineering Rajshahi University of Engineering & Technology Rajshahi Bangladesh

ISBN: (数字)9798350357509

ISBN: (纸本)9798350357516

Potatoes play an important role as a major agricultural product since the economic system of Bangladesh strongly relies on farming. The presence of plant diseases particularly late blight poses substantial threats to economic stability together with food security and harvest yields. Lightweight Convolutional Neural Network (CNN) and Support Vector Machine (SVM) represent a new approach to automate the detection of diseases appearing on potato leaves. The validation accuracy obtained by this model reaches an average of 98.53%. The model sustains high performance by employing 0.45 million parameters which leads to minimized storage needs equivalent to 1.75 MB and reduced computational expenses. SHAP (SHapley Additive exPlanations) together with LIME (Local Interpretable Model-agnostic Explanations) serve as explainable AI techniques used to enhance model prediction transparency. The proposed model excels beyond VGG-16 and VGG-19 transfer learning standards in all evaluation metrics including precision, recall and F1-score even though its parameter count stands at 32.8× and 44.6× lower. Mobile deployment possibilities of the simplified modelarchitecture allow farmers to receive prompt disease diagnoses along with correct decision-making resources. A web-based platform enables users to analyze potato leaf diseases by uploading images in real-time. The study creates substantial progress in agricultural productivity alongside ensuring food security for Bangladesh.

关键词： Support vector machines Accuracy Explainable AI Computational modeling Transfer learning Food security Predictive models Real-time systems Convolutional neural networks Diseases

来源：评论

学校读者我要写书评

暂无评论

Deep Learning-Driven Cloud Intrusion Detection: Internet of Things and Cloud Networks

Deep Learning-Driven Cloud Intrusion Detection: Internet of ...

引用

Advances in computer Science, Electrical, Electronics, and Communication Technologies (CE2CT), International Conference on

作者： Iffath Tanjim Moon Md. Nazrul Islam Mondal Dept. of Computer Science & Engineering Rajshahi University of Engineering & Technology Rajshahi Bangladesh

ISBN: (数字)9798331518578

ISBN: (纸本)9798331518585

In this paper, we have utilized deep learning approaches to detect cloud intrusion for the Internet of Things (IoT). The emerging growth of IoT and cloud environments has revolutionized many industries by allowing real-time data sharing and storage. Cloud platforms offer the essential infrastructure for managing the massive volumes of data produced by IoT devices. However, due to several flaws, including insufficient encryption, insecure APIs, and weak authentication, intruders can exploit the vulnerabilities to interrupt services and obtain unauthorized access. Therefore, deep learning can be an effective solution to detect cloud intrusion in IoT. In this research, we have developed individual convolutional neural network (CNN) and long short-term memory (LSTM) models, along with a hybrid CNN-LSTM model for multiclass classification and binary classification to detect cloud intrusion. In addition, we have examined which models perform the best for cloud intrusion detection for IoT among the individual and hybrid models. The proposed models are effective in detecting cloud intrusion, as CNN can extract spatial features, and LSTM can capture temporal dependencies from network traffic data. Furthermore, we have utilized one of the most recent and comprehensive datasets for IoT, the ToN-IoT dataset, containing real-world data. In multiclass classification, the hybrid CNN-LSTM has given the best performance, achieving an accuracy of 97.65%, precision of 97.72%, recall of 97.65%, f1 score of 97.67%, and AUC of 99.93%. In binary classification, all CNN, LSTM, and hybrid CNN-LSTM have demonstrated high performance, where LSTM has achieved the highest accuracy of 99.97%. Moreover, we have evaluated the performance of multiclass classification for detecting nine distinct kinds of attacks. In this instance, CNN has shown better performance in detecting backdoor attacks. Thus, this research can make the IoT platform more secure by efficiently detecting cloud intrusion.

关键词： Deep learning Cloud computing Accuracy Intrusion detection Telecommunication traffic Feature extraction Real-time systems Internet of Things Convolutional neural networks Long short term memory

来源：评论

学校读者我要写书评

暂无评论

Identification of Influential Genes and Survival Probability of Colorectal Cancer Patients

Identification of Influential Genes and Survival Probability...

引用

International Conference on Electrical, computer and Communication Engineering (ECCE)

作者： Nitun Kumar Podder Salma Khatun MD Anamul Haque Dept. of Computer Science & Engineering Pabna University of Science and Technology Pabna Bangladesh

ISBN: (数字)9798350357509

ISBN: (纸本)9798350357516

Worldwide, colorectal cancer (CRC) ranks high among both men and women's cancers. When it comes to tumors of the large intestine, colorectal cancer (CRC) is by far the worst. Even while the vast majority of polyps are completely harmless, they do have a connection to cancer. Related to cancer is a small clump of cells called polyps; most of them are benign. Over time, some of those, like abnormal cell growth, growing through rapidly dividing cells, and abnormal growing of polyps inside the intestine, become malignant tumors called CRC. It is noticed that CRC-affected people have a great chance of comorbidities like attention deficit disorder, amyotrophic lateral sclerosis, cholesterol, heart failure, low density lipoprotein, systemic lupus erythematous, etc. An essential component of cancer diagnosis and therapy is the identification of influential genes (IFGs). Here, we use the cancer genome atlas (TCGA) dataset to find the relevant CRC IFGs for this purpose. Hence, we have used the Bonferroni correction technique and the Kruskal-Wallis test. Applying these techniques we got 24 major IFs among 16296 genes. Later, in applying those IFGs, we have performed on comorbidities, functional and ontological pathways, protein-protein interactions (PPIs), PCIs, and diseasome networks to predict CRC survival and also the Cox proportional hazard model and the product limit (PL) will be utilized to investigate the impact of clinical and genetic variables that significantly affect the survival of CRC patients. This paper provides a basis for detecting major cancer genes and highlight the colorectal cancer effect on associative diseases development.

关键词： Proteins Intestines Malignant tumors Medical treatment Genomics Colorectal cancer Predictive models Hazards Cardiovascular diseases Bioinformatics

来源：评论

学校读者我要写书评

暂无评论

Efficient Host Threat Detection Using The Procsage Method

Efficient Host Threat Detection Using The Procsage Method

引用

Emerging Smart Computing and Informatics (ESCI), Conference on

作者： Shashikumar D R Madhura N S Lekha Reddy Navya M Purav S Shreyas H R Dept. Computer Science & Engineering Sai Vidya Institute of Technology Bangalore India

ISBN: (数字)9798331515683

ISBN: (纸本)9798331515690

Advanced Persistent Threats (APTs) pose a significant challenge to modern cybersecurity due to their stealth and complexity. This paper introduces a scalable host threat detection framework that utilizes provenance graphs and graph representation learning to overcome limitations of traditional defense mechanisms. By optimizing process- and thread-centric analyses within provenance graphs, the system minimizes computational overhead while maintaining robust behavioral insights. Leveraging semantic and topological feature extraction with GraphSAGE-based learning, the framework achieves precise anomaly detection via clustering techniques. Validation on benchmark datasets demonstrates substantial improvements in accuracy, efficiency, and scalability, showcasing the framework's potential for real-world applications.

关键词： Representation learning Computational modeling Scalability Instruction sets Semantics Benchmark testing Feature extraction Threat assessment computer security Anomaly detection

来源：评论

学校读者我要写书评

暂无评论

Reliable Heart Disease Models: Enhancing Data Integrity and Traceability

Reliable Heart Disease Models: Enhancing Data Integrity and ...

引用

International Conference on Trends in Electronics and Informatics (ICEI)

作者： Yaswanth Pothuraju Venkata Yajnesh Pisupati Ms. Velvizhi R Dept. of Computer Science and Engineering Sathyabama Institute of Science and Technology Chennai India

ISBN: (数字)9798331544607

ISBN: (纸本)9798331544614

Heart disease stands as a worldwide top killer so scientists seek to build dependable prognostic diagnostic and therapeutic models for cardiovascular diseases. This paper investigates how data integrity combined with traceability functions helps to improve the reliability of heart disease prediction models. This paper explores the present difficulties which stem from diverse data set types alongside machine learning method integration and diagnostic model validation methods for different ethnic groups. The article investigates how blockchain technology and space along with IoT systems along with advanced analytic techniques enhance the traceability and robustness of predictive models. This investigation demonstrates the creation of correct heart disease models while promoting interdisciplinary team effort and clear methods and firm data governance rules. The developed solutions will boost patient results through individualized care and supply essential information for public health tactics while handling healthcare and research ethical dilemmas.

关键词： Heart Databases Data integrity Space technology Scalability Machine learning Predictive models Data models Blockchains Cardiovascular diseases

来源：评论

学校读者我要写书评

暂无评论

Real-Time Gaze Tracking for Online Examination

Real-Time Gaze Tracking for Online Examination

引用

Pervasive Computational Technologies (ICPCT), International Conference on

作者： Preyanshu Bisht Suresh Kumar Dept. of Computer Science and Engineering Netaji Subhas University of Technology Delhi India

ISBN: (数字)9798331508685

ISBN: (纸本)9798331519476

The exponential growth in online education has increased the demand for automated systems to ensure academic integrity during online examinations. A real-time proctoring system addresses this need by monitoring a student's eye gaze and head movements during an exam. This system leverages Dlib's pre-trained frontal face detector and 68-point facial landmark predictor to detect facial features and track the direction of eye movement and head position. By analyzing these metrics, the system can flag behaviors associated with potential cheating, such as looking away from the screen for extended periods. This paper discusses the design and implementation of the proctoring system, details the detection algorithms, and evaluates its effectiveness for real-time monitoring.

关键词： Measurement Tracking Education Gaze tracking Detectors Real-time systems Detection algorithms Monitoring Facial features Faces

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：