检索结果-内蒙古大学图书馆

Temple Inscriptions Recognition and Transliteration in Devanagari Script 2

Temple Inscriptions Recognition and Transliteration in Devan...

2nd IEEE International Conference on Vision Towards Emerging Trends in Communication and Networking Technologies, ViTECoN 2023

作者： Babu, B. Sathish Shetty, Sannidhi Agarwal, Anushka Sreerama, Sai Lahari Bhustali, Vaishnavi K. Sanjana, Sanka RV College of Engineering Department of Artificial Intelligence and ML Bengaluru India RV College of Engineering Department of Computer Science and Engineering Bengaluru India

ISBN: (纸本)9798350347982

Ancient inscriptions, palm scripts, manuscripts, etc., have vital information about India's rich culture. Recognition and understanding of these inscriptions have been challenging for epigraphers and professionals. The goal of the proposed research is to advance optical character recognition methods for archival Vatteluttu script inscriptions, which date back to the 4th or 5th century AD. This paper discusses a deep learning model to transliterate the ancient Tamil inscriptions (Vatteluttu Script), which can be extended further to other languages. The proposed work is beneficial to epigraphists, archaeological researchers, and the general public who are interested in this topic. The developed deep learning model has achieved an accuracy of 84.12%. © 2023 IEEE.

关键词： Optical character recognition

来源：评论

学校读者我要写书评

暂无评论

Multi-orthogonal-oppositional enhanced African vultures optimization for combined heat and power economic dispatch under uncertainty

引用

Neural Computing and Applications 2025年第8期37卷 6097-6123页

作者： Rizk-Allah, Rizk M. Snášel, Václav Hassanien, Aboul Ella Basic Engineering Science Department Faculty of Engineering Menoufia University Shebin El-Kom32511 Egypt Faculty of Electrical Engineering and Computer Science VŠB-Technical University of Ostrava Poruba-Ostrava70800 Czech Republic Faculty of Computers and AI Cairo University Giza Egypt

This paper attempts to conceptualize a potent methodology by combining the African vultures optimization algorithm (AVOA) with a multi-orthogonal-oppositional strategy (M2OS), named AVO-M2OS, to address the nonconvexity and multidimensional nature of the combined heat and power economic dispatch (CHPED) problem under both crisp and uncertainty aspects. The AVO-M2OS uses the M2OS to simultaneously explore the search region, improving solutions’ diversity as well as solution quality. Therefore, AVO-M2OS can perform deeper exploration and exploitation features and thus mitigate the trapping at local optima, especially when tackling the more complicated nature of the CHPED problem. A three-stage analysis is conducted to assess the effectiveness of the proposed AVO-M2OS algorithm. During the first stage, the algorithm’s performance is evaluated on benchmark problems such as CEC 2005 and CEC 2019, employing statistical verifications and convergence characteristics. In the second stage, the significance of the results is evaluated using the nonparametric Friedman test to demonstrate that the results did not occur by chance. The results indicate that the AVO-M2OS algorithm outperforms the best existing algorithm (AVOA) by an average rank of the Friedman test exceeding 26% for the CEC 2005 suite while outperforming the gray wolf optimization (GWO) by 60% for the CEC 2019 suite. Moreover, the AVO-M2OS demonstrates exceptional performance compared to existing state-of-the-art algorithms, surpassing the best algorithm available by an average rank of the Friedman test that exceeds 41%. Finally, the AVO-M2OS’s applicability is achieved by minimizing the operational costs by finding the optimal power and heat generation scheduling for the CHPED problem. The recorded results realize that the AVO-M2OS algorithm offers accurate performance compared to competing optimizers, where it saves the operational cost of the 48-unit system by 24% on the original AVO variant. Furthermore, the u

关键词： Quality control

来源：评论

学校读者我要写书评

暂无评论

MedEx: Enhancing Medical Question-Answering with First-Order Logic based Reasoning and Knowledge Injection 31

MedEx: Enhancing Medical Question-Answering with First-Order...

引用

31st International Conference on Computational Linguistics, COLING 2025

作者： Zafar, aizan Mishra, Kshitij Ekbal, Asif Department of Computer Science and Engineering Indian Institute of Technology Patna India School of AI and Data Science Indian Institute of Technology Jodhpur India

ISBN: (纸本)9798891761964

In medical question-answering, traditional knowledge triples often fail due to superfluous data and their inability to capture complex relationships between symptoms and treatments across diseases. This limits models' ability to provide accurate, contextually relevant responses. To overcome this, we introduce MedEx, which employs First-Order Logic (FOL)-based reasoning to model intricate relationships between diseases and treatments. We construct FOL-based triplets that encode the interplay of symptoms, diseases, and treatments, capturing not only surface-level data but also the logical constraints of the medical domain. MedEx encodes the discourse (questions and context) using a transformer-based unit, enhancing context comprehension. These encodings are processed by a Knowledge Injection Cell that integrates knowledge graph triples via a Graph Attention Network. The Logic Fusion Cell then combines medical-specific logical rule triples (e.g., co-occurrence, causation, diagnosis) with knowledge triples and extracts answers through a feed-forward layer. Our analysis demonstrates MedEx's effectiveness and generalization across medical question-answering tasks. By merging logical reasoning with knowledge, MedEx provides precise medical answers and adapts its logical rules based on training data nuances. © 2025 Association for Computational Linguistics.

关键词： Question answering

来源：评论

学校读者我要写书评

暂无评论

Aligning Diffusion Behaviors with Q-functions for Efficient Continuous Control 38

Aligning Diffusion Behaviors with Q-functions for Efficient ...

引用

38th Conference on Neural Information Processing Systems, NeurIPS 2024

作者： Chen, Huayu Zheng, Kaiwen Su, Hang Zhu, Jun Department of Computer Science and Technology Tsinghua University China Institute for AI BNRist Center Tsinghua-Bosch Joint ML Center THBI Lab Tsinghua University China Guangzhou China

Drawing upon recent advances in language model alignment, we formulate offline Reinforcement Learning as a two-stage optimization problem: First pretraining expressive generative policies on reward-free behavior datasets, then fine-tuning these policies to align with task-specific annotations like Q-values. This strategy allows us to leverage abundant and diverse behavior data to enhance generalization and enable rapid adaptation to downstream tasks using minimal annotations. In particular, we introduce Efficient Diffusion Alignment (EDA) for solving continuous control problems. EDA utilizes diffusion models for behavior modeling. However, unlike previous approaches, we represent diffusion policies as the derivative of a scalar neural network with respect to action inputs. This representation is critical because it enables direct density calculation for diffusion models, making them compatible with existing LLM alignment theories. During policy fine-tuning, we extend preference-based alignment methods like Direct Preference Optimization (DPO) to align diffusion behaviors with continuous Q-functions. Our evaluation on the D4RL benchmark shows that EDA exceeds all baseline methods in overall performance. Notably, EDA maintains about 95% of performance and still outperforms several baselines given only 1% of Q-labelled data during fine-tuning. Code: https://***/thu-ml/Efficient-Diffusion-Alignment. © 2024 Neural information processing systems foundation. All rights reserved.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Deep learning based predictive analysis of energy consumption for smart homes

引用

Multimedia Tools and Applications 2025年第12期84卷 10665-10686页

作者： Malik, Sangeeta Malik, Sitender Singh, Ishmeet Gupta, Harsh Vardhan Prakash, Sidhant Jain, Rachna Acharya, Biswaranjanjan Hu, Yu-Chen Computer Science and Engineering Maharaja Surajmal Institute of Technology New Delhi India IT Department Maharaja Surajmal Institute of Technology New Delhi India IT Department Bhagwan Parshuram Institute of Technology New Delhi India Department of Computer Engineering-AI & BD Marwadi University Gujarat Rajkot India Department of Computer Science TungHai University No. 1727 Sec. 4 Boulevard Xitun District Taichung City407224 Taiwan

Predicting energy consumption has become crucial to creating a sustainable and intelligent environment. With the aid of forecasts of future demand, the distribution and production of energy can be optimized to meet the requirements of a vastly growing population. However, because of the varied types of energy consumption patterns, predicting the demand for any household can be difficult. It has recently gained popularity with social Internet of Things-based smart homes, smart grid planning, and artificial intelligence-based smart energy-saving solutions. Although there are methods for estimating energy consumption, most of these systems are based on one-step forecasting and have a limited forecasting period. Several prediction models were implemented in this paper to address the problem mentioned above and achieve high accuracy, including the baseline model, the Auto-Regressive Integrated Moving Average (ARIMA) model, the Seasonal Auto-Regressive Integrated Moving Average (SARIMAX with eXogenous factors) model, the Long Short-Term Memory (LSTM) Univariate model, and the LSTM Multivariate model. © The Author(s), under exclusive licence to Springer science+Business Media, LLC, part of Springer Nature 2024.

关键词： Forecasting

来源：评论

学校读者我要写书评

暂无评论

ml-based Vehicle Accident Severity Prediction Framework for Traffic Management Systems 2

ML-based Vehicle Accident Severity Prediction Framework for ...

引用

2nd International Conference on computer, Communication and Control, IC4 2024

作者： Patel, Kavya Gondalia, Unnati Kakkar, Riya Gupta, Rajesh Tanwar, Sudeep Garg, Deepak Institute of Technology Nirma University Department of Computer Science and Engineering Gujarat 382481 India Sr University School of Computer Science and Ai Telangana Warangal India

ISBN: (纸本)9798350387933

The present research examines the current issues arises with the growing usage of personal vehicles in the transportation industry. This adoption is due to the technical advancements in the automobile sector. Although personal automobiles have many benefits, such as convenience and safety features, the ensuing increase in traffic congestion has become a serious problem and is contributing to an increase in traffic accidents. In today's dynamic transportation scene, striking a balance between the advantages and problems provided by this paradigm change is essential. We suggest an all-encompassing paradigm for accident severity prediction in response to these difficulties. The proposed framework makes use of cutting-edge sensor technologies and machine learning methods including KNearest Neighbours, Random Forest, XGBoost, and Bagging. This all-encompassing strategy seeks to improve overall transportation efficiency, optimise traffic management, and proactively predict the severity of accidents. The goal of this research is to improve road safety and accident management by providing useful information to emergency services, law enforcement, and decisionmakers. © 2024 IEEE.

关键词： Motor transportation

来源：评论

学校读者我要写书评

暂无评论

Intrusion Detection in NSL-KDD Dataset Using Hybrid Self-Organizing Map Model

引用

computer Modeling in engineering & sciences 2025年第4期143卷 639-671页

作者： Noveela Iftikhar Mujeeb Ur Rehman Mumtaz Ali Shah Mohammed J.F.Alenazi Jehad Ali Knowledge Unit of Systems and Technology University of Management and TechnologySialkot51310Pakistan Department of Computer Science University ofWahWah Cantt47010Pakistan Department of Computer Engineering College of Computer and Information Sciences(CCIS)King Saud UniversityRiyadh11451Saudi Arabia Department of AI Convergence Network Ajou UniversitySuwon16499Republic of Korea

Intrusion attempts against Internet of Things(IoT)devices have significantly increased in the last few *** devices are now easy targets for hackers because of their built-in security *** a Self-Organizing Map(SOM)hybrid anomaly detection system for dimensionality reduction with the inherited nature of clustering and Extreme Gradient Boosting(XGBoost)for multi-class classification can improve network traffic intrusion *** proposed model is evaluated on the NSL-KDD *** hybrid approach outperforms the baseline line models,Multilayer perceptron model,and SOM-KNN(k-nearest neighbors)model in precision,recall,and F1-score,highlighting the proposed approach’s scalability,potential,adaptability,and real-world ***,this paper proposes a highly efficient deployment strategy for resource-constrained network *** results reveal that Precision,Recall,and F1-scores rise 10%-30% for the benign,probing,and Denial of Service(DoS)*** particular,the DoS,probe,and benign classes improved their F1-scores by 7.91%,32.62%,and 12.45%,respectively.

关键词： Intrusion detection self-organizing map Internet of Things dimensionality reduction

来源：评论

学校读者我要写书评

暂无评论

ml-Orchestrated Malware Detection Framework for Windows APIs in Critical Infrastructure Systems 2

ML-Orchestrated Malware Detection Framework for Windows APIs...

引用

2nd International Conference on computer, Communication and Control, IC4 2024

作者： Vaghasiya, Shayalkumar Gupta, Rajesh Mohapatra, Santoshinee Tanwar, Sudeep Garg, Deepak Institute of Technology Nirma University Department of Computer Science and Engineering Gujarat 382481 India Sr University School of Computer Science and Ai Telangana Warangal India

ISBN: (纸本)9798350387933

The security of critical infrastructure systems is crucial as they provide basic amenities and services necessary for functioning of a community, also they represent important part of country's economy. An attack on this critical infrastructure systems can have significant consequences. In this paper we discuss the This paper addresses the increasing security challenges in critical infrastructure systems due to cyber threats, particularly malware and propose a framework based on machine learning for robust malware classification. Our proposed framework includes a dataset of Windows API functions used by the software, preprocessing techniques, and the XGBoost model which achieves an accuracy of 86.98%. Using this model, we can detect and categorize five different types of malwares to provide security against malware attacks. The results show model's ability to successfully classify different kinds of malware. Realtime monitoring is another feature of this framework which is important for actively protecting critical infrastructure systems. © 2024 IEEE.

关键词： Machine learning

来源：评论

学校读者我要写书评

暂无评论

A Curious Case of Searching for the Correlation between Training Data and Adversarial Robustness of Transformer Textual Models 62

A Curious Case of Searching for the Correlation between Trai...

引用

Findings of the 62nd Annual Meeting of the Association for Computational Linguistics, ACL 2024

作者： Dang, Cuong Le, Dung D. Le, Thai FPT Software AI Center Viet Nam College of Engineering and Computer Science VinUniversity Viet Nam Department of Computer Science Indiana University United States

ISBN: (纸本)9798891760998

Existing works have shown that fine-tuned textual transformer models achieve state-of-the-art prediction performances but are also vulnerable to adversarial text perturbations. Traditional adversarial evaluation is often done only after fine-tuning the models and ignoring the training data. In this paper, we want to prove that there is also a strong correlation between training data and model robustness. To this end, we extract 13 different features representing a wide range of input fine-tuning corpora properties and use them to predict the adversarial robustness of the fine-tuned models. Focusing mostly on encoder-only transformer models BERT and RoBERTa with additional results for BART, ELECTRA, and GPT2, we provide diverse evidence to support our argument. First, empirical analyses show that (a) extracted features can be used with a lightweight classifier such as Random Forest to predict the attack success rate effectively, and (b) features with the most influence on the model robustness have a clear correlation with the robustness. Second, our framework can be used as a fast and effective additional tool for robustness evaluation since it (a) saves 30x-193x runtime compared to the traditional technique, (b) is transferable across models, (c) can be used under adversarial training, and (d) robust to statistical randomness. Our code is publicly available at https://***/CaptainCuong/RobustText_ACL2024. © 2024 Association for Computational Linguistics.

关键词： Adversarial machine learning

来源：评论

学校读者我要写书评

暂无评论

Chaotic Water Wave Optimization and Support Vector Machine of Role-Based Access Control Policies in Enterprise IAM

Chaotic Water Wave Optimization and Support Vector Machine o...

引用

2024 International Conference on Distributed Systems, computer Networks and Cybersecurity, ICDSCNC 2024

作者： Priyanka, K. Harshini, Chikka Aluvala, Srinivas Abbas, Haider Mohammed Ram Kumar, Ramaswamy Padmanathan Nitte Meenakshi Institute of Technology Department of Ise Bengaluru India Department of Cse Hyderabad India Sr University Department of Computer Science and Engineering Warangal India The Islamic University College of Technical Engineering Department of Computers Techniques Engineering Najaf Iraq Gokaraju Rangaraju Institute of Engineering and Technology Department of Ai&ml Bachupally Hyderabad India

ISBN: (纸本)9798350375442

The access control state of a system is complex, and Machine Learning (ml) based access control decision engines have demonstrated advantages of policies. However, struggle to effectively implement Role-Based Access Control (RBAC) policies because require large enterprises Identity and Access Management (IAM) system with a high volume of users and access logs. This paper, proposed Chaotic Water Wave Optimization (CWWO) algorithm combined with a Support Vector Machine (SVM) to classify users based on selected features, leading to more accurate and secure role to assignment. The dataset used is obtained from amazon and pre-processing is done using Min-max normalization, which scales the feature to a specific range, ensuring that different features are on same scale. Feature selection is performed using the CWWO algorithm, which helps in selecting relevant feature and reduce dimensionality. The classification using SVM for binary classification, which assigns relevant feature and handle large amounts of data, achieving better accuracy. The proposed ensemble model technique for ship motion detection is achieving better outcomes such as high accuracy of 0.98, precision of 0.97, F1-score of 0.96 and recall of 0.98 and respectively. The existing technique such Random Forest (RF) and Decision Tree (DT) are evaluated of proposed method. © 2024 IEEE.

关键词： Random forests

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：