检索结果-内蒙古大学图书馆

13th IEEE Congress on Evolutionary Computation, CEC 2024

作者： Liu, Shengcai Chen, Caishun Qu, Xinghua Tang, Ke Ong, Yew-Soon Southern University of Science and Technology Centre for Frontier Ai Research A∗star Department of Computer Science and Engineering China Centre for Frontier Ai Research A∗star Singapore Tianqiao & Chrissy Chen Institute China Southern University of Science and Technology Guangdong Provincial Key Laboratory of Brain-Inspired Intelligent Computation Department of Computer Science and Engineering China Nanyang Technological University Centre for Frontier Ai Research A∗star College of Computing and Data Science Singapore

ISBN: (纸本)9798350308365

Evolutionary algorithms (EAs) have achieved remarkable success in tackling complex combinatorial optimization problems. However, EAs often demand carefully-designed operators with the aid of domain expertise to achieve satisfactory performance. In this work, we present the first study on large language models (LLMs) as evolutionary combinatorial optimizers. The main advantage is that it requires minimal domain knowledge and human efforts, as well as no additional training of the model. This approach is referred to as LLM-driven EA (LMEA). Specifically, in each generation of the evolutionary search, LMEA instructs the LLM to select parent solutions from current population, and perform crossover and mutation to generate offspring solutions. Then, LMEA evaluates these new solutions and include them into the population for the next generation. LMEA is equipped with a self-adaptation mechanism that controls the temperature of the LLM. This enables it to balance between exploration and exploitation and prevents the search from getting stuck in local optima. We investigate the power of LMEA on the classical traveling salesman problems (TSPs) widely used in combinatorial optimization research. Notably, the results show that LMEA performs competitively to traditional heuristics in finding high-quality solutions on TSP instances with up to 20 nodes. Additionally, we also study the effectiveness of LLM-driven crossover/mutation and the self- adaptation mechanism in evolutionary search. In summary, our results reveal the great potentials of LLMs as evolutionary optimizers for solving combinatorial problems. We hope our research shall inspire future explorations on LLM-driven EAs for complex optimization challenges. © 2024 IEEE.

关键词： Traveling salesman problem

来源：评论

学校读者我要写书评

暂无评论

Context is King: Rethinking Query Expansion in Modern Search Engines

Context is King: Rethinking Query Expansion in Modern Search...

引用

2025 International Conference on Artificial Intelligence and Data engineering, aiDE 2025

作者： Batri, Krishnan Lakshmi, S. Ramesh, S. Sangeetha, N. Annamalai, S. School of Computer Science and Engineering Bengaluru India Electronics and Communication Engineering Bengaluru India City Engineering College Department of AI&ML Bengaluru India

ISBN: (纸本)9798331527518

In the era of digital information overload, search engines have evolved beyond simple keyword matching to embrace context-aware methodologies. However, the task of accurately interpreting user intent remains a critical challenge, with traditional query expansion techniques often falling short of capturing semantic and contextual nuances. This paper critically examines the evolution of query expansion strategies, highlighting their limitations and exploring state-of-the-art techniques such as machine learning models, knowledge graphs, and transformer-based systems like BERT and GPT. Experimental results compare traditional methods like synonym expansion with modern approaches, demonstrating the superior relevance and diversity of results from context-aware systems. The study underscores the importance of integrating real-time contextual signals and dynamic user preferences to enhance search engine performance, paving the way for more personalized and accurate retrieval systems. © 2025 IEEE.

关键词： Search engines

来源：评论

学校读者我要写书评

暂无评论

Mental Health Predictive Analysis using Multiple Machine Learning Techniques 2

Mental Health Predictive Analysis using Multiple Machine Lea...

引用

2nd IEEE International Conference on Communication, Security and Artificial Intelligence, ICCSai 2023

作者： Tressa, Neethu Asha, V. Prasad, Arpana Prathith Pramodh Kumar, D. Saju, Binju New Horiznon College of Engineering Department of Master of Computer Applications Bengaluru India Adi Shankara Institute of Engineering & Technology Department of Computer Science - Ai Engineering Kerala Kalady India

ISBN: (纸本)9798350369960

The growth in mental health issues and the demand for high-quality medical care have prompted the requirement of machine learning techniques in the psychological research area. For the better prediction of mental health complications, this work presents an up-to-date thorough evaluation of machinelearning algorithms. This article reviews the difficulties, limitations and future directions of using machine learning approaches in psychological health. By searching reliable databases, relevant studies and publications on machine learning methods were collected for predicting mental health related issues. For the completion of this work, 40 research outcomes were analyzed using screening and classification procedures. The analyzed works were categorized according to mental health conditions such as schizophrenia, bipolar disorder, depression or anxiety, PTSD, and mental health conditions affecting children. While discussing the results, the difficulties and limitations that many researchers have encountered when studying mental health related problems were also considered. The paper also encompasses specific directions for future research work that are feasible and the need for utilizing varying approaches in machine learning for psychological health disorder predictions. © 2023 IEEE.

关键词： Decision Tree Classifier Flask Deployment K Nearest Neighbor Classifier Logistic Regression Algorithm Naïve Bayes Random Forest Stacking

来源：评论

学校读者我要写书评

暂无评论

Image Forgery Detection Using MD5 & Open CV 2

Image Forgery Detection Using MD5 & Open CV

引用

2nd IEEE International Conference on Emerging Research in Computational science, ICERCS 2024

作者： Shaikh, Mohammad Shahnawaz Biswal, Aparajita Pandwal, Akruti Khodifad, Nilesh Vaghela, Bhavesh Parul University Parul Institute of Engineering and Technology Department of AI and AIDS Vadodara India Parul University Computer Science & Engineering Parul Institute of Engineering and Technology Vadodara India

ISBN: (纸本)9798331534950

The more the advanced image making tools become accessible, the more image forgery is proliferated in hardware and software across all these domains - forensics, journalism, authentication, etc. In this research paper, we investigate methods on how to detect forged images by using Python programing, MD5 hashing algorithm and OpenCV library. The first part of the study utilizes MD5 to do an integrity verification by generating a unique hash value for each original image. The digital fingerprint you get from hashing each step of the image is this hash, and is a powerful way to compare against suspected forged images. MD5 is not perfect, but it is useful for performing an initial check of manipulated site;while not perfect, it does the trick. As powerful image processing software, OpenCV is used to enhance the detection process. This library provides detailed analysis of pixel patterns, possibly strange color discrepancies or just other things that would hint a manipulation. The proposed method performs advanced anomaly detection upon the extracted features from both original and suspect images with the aim of discovering inconsistencies meant to imply forgery. The research then proves through a series of experimental validations that the use of MD5 coupled with OpenCV significantly increases detection accuracy, while reducing false positives and detecting both overt and subtle manipulation. This not only underlines the need for creating a multilayered approach to counter the challenges posed by image forgery but also introduces a hybrid framework called UEFRG that combines three different methods done specifically to address image forgery. As digital content evolves, this study indicates the necessity of reliable detection methods to ensure integrity of visual media. The algorithm is further refined and machine learning techniques are explored for incorporation for future work to further improve the detection capabilities to keep pace with emerging threats in digital

关键词： Image matching

来源：评论

学校读者我要写书评

暂无评论

Extending LLMs to New Languages: A Case Study of Llama and Persian Adaptation 31

Extending LLMs to New Languages: A Case Study of Llama and P...

引用

31st International Conference on Computational Linguistics, COLING 2025

作者： Sani, Samin Mahdizadeh Sadeghi, Pouya Vu, Thuy-Trang Yaghoobzadeh, Yadollah Haffari, Gholamreza Department of Electrical and Computer Engineering University of Tehran Iran Tehran Institute for Advanced Studies Khatam University Iran Department of Data Science & AI Monash University Australia

ISBN: (纸本)9798891761964

Large language models (LLMs) have made great progress in classification and text generation tasks. However, they are mainly trained on English data and often struggle with low-resource languages. In this study, we explore adding a new language, i.e., Persian, to Llama (a model with a limited understanding of Persian) using parameter-efficient fine-tuning. We employ a multi-stage approach involving pretraining on monolingual Persian data, aligning representations through bilingual pretraining and instruction datasets, and instruction-tuning with task-specific datasets. We evaluate the model's performance at each stage on generation and classification tasks. Our findings suggest that incorporating the Persian language, through bilingual data alignment, can enhance classification accuracy for Persian tasks, with no adverse impact and sometimes even improvements on English tasks. Additionally, the results highlight the model's initial strength as a critical factor when working with limited training data, with cross-lingual alignment offering minimal benefits for the low-resource language. Knowledge transfer from English to Persian has a marginal effect, primarily benefiting simple classification tasks. © 2025 Association for Computational Linguistics.

关键词： Computational linguistics

来源：评论

学校读者我要写书评

暂无评论

Recognizing Human Emotions by using Lazy classifier with selected algorithm 9

Recognizing Human Emotions by using Lazy classifier with sel...

引用

9th IEEE International Conference on Smart Structures and Systems, ICSSS 2023

作者： Loganathan, V. Mohanaprakash, T.A. Bhavani, R. Savija, J. Saveetha Engineering College Department of Computer Science Engineering Chennai India Panimalar Engineering College Department of Computer Science and Engineering Chennai India St Joseph College of Engineering Department of It Chennai India Velammal Engineering College Department of Ai & Ds Chennai India

ISBN: (纸本)9798350384208

Music predates language, and humans are born with the ability to communicate themselves via music. Music has played a vital part in human history, and it is incorporated into many facets of human life Music is showing tremendously significant role in people's production and life. This research is focusing a n identification of human emotions like happiness, sad, relaxation, tension on Turkish music emotion dataset through inductive learning approach. This work is to focus on the various lazy learning approaches are implemented to recognize human emotions while listening music. The K=3 with Ball Tree NN search model acquires highest performance which as accuracy 85.75%, kappa 0.66, precision 0.86, F-Measure0.86, recall 0.86, Matthews Correlation Coefficient 0.65 which generated the best outcomes based on their efficacy in comparison to other models. The K=3 with Ball Tree model produces best result for recognizing human emotions compare with other models. © 2023 IEEE.

关键词： Trees (mathematics)

来源：评论

学校读者我要写书评

暂无评论

Robust Digital Watermarking based on Machine Learning

Robust Digital Watermarking based on Machine Learning

引用

Self Sustainable Artificial Intelligence Systems (ICSSAS), International Conference on

作者： Manish Rai Hemlata Department of Computer Science and Engineering (AI&ML) Manipal University Jaipur Jaipur India

In the generation of big data and systems administration, it is important to build up a safe and strong digital watermarking plan with high computational effectiveness to ensure copyrights of digital works. However, the greater part of the current strategies focuses on robustness and embedding limit, losing a security, or requiring huge computational assets in the encryption procedure. Digital watermarking has become an important method for ensuring the security of digital copyrights and ensuring the authenticity of data because of rapid advancement and broad usage of technologies related to multimedia and network connectivity. which has been used for practical application such as medical application, currency, digital product identification and many more. The success story of digital watermarking techniques deepened on the embedding of the watermark image and present image. In the current decade, the machine learning feature-based watermarking techniques create new milestone concerning security concern. The machine learning and hybrid digital watermarking performances an especially vital role in optimization and selection of features which improves the performance of watermarking process. In this paper, we have presented countless digital watermarking, methods which is based on various machine learning procedures along with their comparison, and limitation.

关键词：

来源：评论

学校读者我要写书评

暂无评论

NPBMT: A Novel and Proficient Buffer Management Technique for Internet of Vehicle-Based DTNs

引用

computers, Materials & Continua 2023年第10期77卷 1303-1323页

作者： Sikandar Khan Khalid Saeed Muhammad Faran Majeed Salman A.AlQahtani Khursheed Aurangzeb Muhammad Shahid Anwar Department of Computer Science Shaheed Benazir Bhutto University SheringalDir(U)Khyber Pakhtunkhwa18200Pakistan Department of Computer Science Kohsar University MurreePunjab47150Pakistan Department of Computer Engineering College of Computer and Information SciencesKing Saud UniversityP.O.Box 51178Riyadh11543Saudi Arabia Department of AI and Software Gachon UniversitySeongnam-si13120South Korea

Delay Tolerant Networks(DTNs)have the major problem of message delay in the network due to a lack of endto-end connectivity between the nodes,especially when the nodes are *** nodes in DTNs have limited buffer storage for storing delayed *** instantaneous sharing of data creates a low buffer/shortage ***,buffer congestion would occur and there would be no more space available in the buffer for the upcoming *** address this problem a buffer management policy is proposed named“A Novel and Proficient Buffer Management Technique(NPBMT)for the Internet of Vehicle-Based DTNs”.NPBMT combines appropriate-size messages with the lowest Time-to-Live(TTL)and then drops a combination of the appropriate messages to accommodate the newly arrived *** evaluate the performance of the proposed technique comparison is done with Drop Oldest(DOL),Size Aware Drop(SAD),and Drop Larges(DLA).The proposed technique is implemented in the Opportunistic Network Environment(ONE)*** shortest path mapbased movement model has been used as the movement path model for the nodes with the epidemic routing *** the simulation results,a significant change has been observed in the delivery probability as the proposed policy delivered 380 messages,DOL delivered 186 messages,SAD delivered 190 messages,and DLA delivered only 95 messages.A significant decrease has been observed in the overhead ratio,as the SAD overhead ratio is 324.37,DLA overhead ratio is 266.74,and DOL and NPBMT overhead ratios are 141.89 and 52.85,respectively,which reveals a significant reduction of overhead ratio in NPBMT as compared to existing *** network latency average of DOL is 7785.5,DLA is 5898.42,and SAD is 5789.43 whereas the NPBMT latency average is *** reveals that the proposed policy keeps the messages for a short time in the network,which reduces the overhead ratio.

关键词： Delay tolerant networks buffer management message drop policy ONE simulator NPBMT

来源：评论

学校读者我要写书评

暂无评论

A Novel Approach to Avoid Road Traffic Accidents and Develop Safety Rules for Traffic Using Crash Prediction Model Technique 6th

A Novel Approach to Avoid Road Traffic Accidents and Develop...

引用

6th International Conference on Microelectronics and Telecommunication engineering, ICMETE 2022

作者： Ahammad, Sk Hasane Sukesh, M. Narender, Mekala Ettyem, Sajjad Ali Al-Majdi, Kadhum Saikumar, K. Department of Electronics and Communication Engineering Koneru Lakshmaiah Education Foundation Guntur India Department of Computer Science and Engineering Vaagdevi College of Engineering Telangana Warangal India Department of Computer Science and Engineering Jayamukhi Institute of Technological Sciences Narsampet Warangal India National University of Science and Technology Muscat Iraq Ashur University College Baghdad Iraq School of Engineering Department of AI-ML Malla Reddy University Telangana Hyderabad500043 India

ISBN: (纸本)9789811995118

The expansion of nations and communities has resulted in a variety of externalities, such as an increase in traffic accidents. Many attempts have been undertaken to minimize the injuries and fatalities and their intensity. Traffic safety modeling is a most significant technique to motivate harmless mobility because it is capable of the creation of Crash Prediction Models (CPMs) as well as the investigation of the fundamentals that contribute to the incidence of crashes. Statistical modeling has been utilized in this process in the past, regardless of the fact that they are aware of the limits of this sort of strategy which allows you to experiment with other options, such as using machine learning approaches. Machine learning approaches applied to collision datasets can assist researchers in better knowing the features of motorist behavior, highway surroundings, and meteorological circumstances that are linked to varying mortality risk levels. If we build a reliable predictive model capable of automatically classifying the degree of injury in diverse traffic accidents, we may be able to discover patterns involved in severe wrecks. These patterns of behavior and road accidents can be used to design traffic safety rules. © 2023, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

关键词： Machine learning

来源：评论

学校读者我要写书评

暂无评论

Cluster Based Ensemble Learning for Classification of Breast Cancer Data

Cluster Based Ensemble Learning for Classification of Breast...

引用

Technology & engineering Management Conference - Asia Pacific (TEMSCON-ASPAC), IEEE

作者： Bharath Kumar Nangunuri Gandhe Sripriya Sirnam Keerthi S V Vaishnavi M.A. Jabbar Computer Science Engineering(AI&ML) Vardhaman College Of Engineering Shamshabad Hyderabad

ISBN: (数字)9798350384659

ISBN: (纸本)9798350384666

The most common cause of death is breast cancer (BC). The precise classification of breast cancer data is essential for the diagnosis of cancer, and patients can save money by avoiding unnecessary surgeries by understanding the difference between benign and malignant tumors. The appropriate method of action can be determined using the classification of breast cancer. A well-known area of medical research is the classification of patient populations with benign and malignant diseases. Because machine learning can extract significant characteristics from a collection of medical data, it is commonly used in the prediction of breast cancer. More emphasis has been paid recently to the use of machine learning techniques in medical diagnosis. This research aims to use ensemble learning to solve the classification problem of breast cancer data. Techniques for ensemble learning are applied to enhance a classifier’s performance. It covers RBF, SVM, and Hierarchical Clustering. The widely-studied open-access data set “Wisconsin Breast Cancer Data Set (WBCD)” was the subject of extensive experiments in this study. Training and testing portions of the data set are separated. The model’s effectiveness was evaluated using several parameters: accuracy, sensitivity, specificity, positive predictive value, negative predictive value, error rate, and false-positive rate. The proposed strategy surpassed the existing methodologies, according to experimental findings, and recorded a remarkable accuracy of 97.66% when classifying data related to breast cancer. The suggested ensemble learning would be effective in assisting oncologists in precisely identifying cancer tumors and assisting patients in receiving the best possible care.

关键词： Training Support vector machines Sensitivity Sociology Surgery Predictive models Breast cancer

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：