检索结果-内蒙古大学图书馆

2nd International Conference of Intelligent Methods, Systems and Applications, IMSA 2024

作者： Zgorni, Ghassan Qussay, Adham Elmasry, Salma Ayman, Renada Habib, Sama Amr Hossam, Magi School of Computing and Digital Technology ESLSCA University Cairo Egypt

ISBN: (纸本)9798350362633

In the realm of retail, particularly in supermarkets and hypermarkets, monitoring, product management, and stocking traditionally rely on human effort and manual labor. However, this approach often proves to be time-consuming, labor-intensive, and prone to errors. To address these challenges, this study proposes the implementation of computer vision and object detection technologies in the retail industry. By implementing computer vision models, the study aims to automate the process of detecting retail items on shelves, with a specific focus on a manually curated drinks dataset. To accomplish this objective, this paper performs a comparative analysis of two prominent object detection models, namely YOLO versions 8 and 9, and DETR, within the context of retail supermarket scenarios. A novel dataset comprising 2000 real-life images manually collected from retail shelves is created to assess the performance of these frameworks in accurately detecting and localizing products. The experimental findings indicate that YOLOv9 a chieved a mean average precision (mAP) score of 0.6, surpassing both DETR and YOLOv8 in terms of detection accuracy and localization capabilities. © 2024 IEEE.

关键词： Sales

来源：评论

学校读者我要写书评

暂无评论

Data Augmentation for Sentiment Analysis Enhancement: A Comparative Study 6

Data Augmentation for Sentiment Analysis Enhancement: A Comp...

引用

6th International Conference on computing and Informatics, ICCI 2024

作者： Yasser, Farida Hatem, Souzan Ayman, Laila Mohamed, Loujain Mohammed, Ammar School of Computing and Digital Technology ESLSCA University Egypt Cairo Egypt

ISBN: (纸本)9798350373875

For many years, researchers in the field of natural language processing have been exploring sentiment analysis, a method for understanding human feelings and thoughts expressed in text. Sentiment analysis works by first analyzing the sentiment of individual words or phrases, using methods like dictionaries, machine learning, or natural language processing. In machine learning, however, the reliance on training data quality and quantity poses challenges, including data scarcity and imbalanced label distribution. One possible way to increase the distribution of textual data is utilizing data augmentation methods in which the train data samples are artificially transformed to another data with similar context. One effective approach to generating a more extensive and varied textual samples involves utilizing the capabilities of large language ***, this paper introduces a comparative study on evaluating the impact of various augmentation methods namely, Random Deletion, Synonym Replacement, GPT3.5 generation, and Character Swapping. We compare the performance across six deep learning models, namely CNN-LSTM, BI-LSTM, BERT, TCN, Ensemble CNN Bidirectional GRU, and Deep neural network. The experimental results reveal that BERT exhibits significant accuracy improvements across different augmentation methods, showcasing gains of 14% in Random Deletion, 12.9% in Synonym Replacement, and 6.5% in Character Swapping. © 2024 IEEE.

关键词： Sentiment analysis

来源：评论

学校读者我要写书评

暂无评论

Extracting Health Evidence Information from Biomedical Literature using Large Language Models 11

Extracting Health Evidence Information from Biomedical Liter...

引用

11th IEEE/ACM International Conference on Big Data computing, Applications and Technologies, BDCAT 2024

作者： Shahid, Muhammad Ammar Afzal, Muhammad Seker, Huseyin Birmingham City University School of Computing and Digital Technology Birmingham United Kingdom

ISBN: (纸本)9798350367300

The current biomedical literature is huge, unstructured, and complex, posing a significant challenge to efficient information processing and consequently creating a substantial gap between medical research and clinical practice. This underscores the need for innovative approaches to convert unstructured information into a computable format readily available for clinical decision-making. The study aims to leverage large language models (LLMs) to accurately extract health evidence from unstructured biomedical text by utilizing the PICO (Patient, Intervention, Comparison, Outcome) framework. We implemented different variations of Generative Pre-trained Transformers (GPT), where GPT 4 and GPT 4 Turbo Preview consistently improve recall, precision, and accuracy compared to competitors like GPT 4o and GPT 3.5. We employed the document retrieval strategy, which involved the use of the "stuff method"instead of the "map-reduce"and "map-rerank"methods, aligning to achieve higher accuracy within a limited document set. Challenges include accurately synthesizing fragmented PICO elements and ensuring answer correctness due to limitations in context retrieval length. The performance of the proposed system is evaluated using the Retrieval Augmented Generation assessment (RAGAs) framework. The findings highlight LLMs' potential in medical research while emphasizing the need for enhancements in context retrieval and model correctness. The study's outcomes suggest that leveraging LLMs for biomedical information extraction can potentially improve the healthcare decision-making landscape. © 2024 IEEE.

关键词： Electronic health record

来源：评论

学校读者我要写书评

暂无评论

Comparative Design Analysis of a 1 MW Islanded and Non-Islanded Floating Solar Power System at Mangla Hydro Reservoir 11

Comparative Design Analysis of a 1 MW Islanded and Non-Islan...

引用

11th International Conference on Electrical and Electronics Engineering, ICEEE 2024

作者： Paramel, Arshida Aggarwal, Geetika Sabeel, Khursheed Dawood, Huda School of Computing Engineering and Digital Technology Teesside University Middlesbrough United Kingdom

ISBN: (纸本)9798350362541

Solar energy is increasingly recognised as an essential part of the global mission for the net zero goal, owing to its renewable nature and cost-effectiveness. In response to escalating power demands, private power operators and industrial investors seek ways to mitigate electricity costs. This study addresses a specific project: an industrial tender for a 1-megawatt hybrid floating solar power plant intended for a water purification industry in Pakistan. The sector has varying monthly load requirements based on product utilisation and aims to sell surplus energy back to the grid after meeting its operational demands. The choice of a floating solar power plant is driven by the prohibitively high land costs in the vicinity of the dam in Islamnagar, Azad Kashmir, coupled with the unfeasibility of meeting industrial water requirements at locations distant from the dam. This paper will conduct a comparative analysis of grid-connected and standalone options for this industry, focusing on performance metrics. The optimal system, characterised by high efficiency and minimal loss, will be selected for the solar plant design. This design will address aspects including load management, battery storage, and energy provision to the grid. © 2024 IEEE.

关键词： Costs

来源：评论

学校读者我要写书评

暂无评论

Using a NEAT approach with curriculums for dynamic content generation in video games

引用

Personal and Ubiquitous computing 2024年第3-4期28卷 629-641页

作者： Hind, Daniel Harvey, Carlo DMTLab School of Computing and Digital Technology Faculty of Computing Engineering and the Built Environment Birmingham City University Curzon Street West Midlands BirminghamB4 7XG United Kingdom

This paper presents a novel exploration of the use of an evolving neural network approach to generate dynamic content for video games, specifically for a tower defence game. The objective is to employ the NeuroEvolution of Augmenting Topologies (NEAT) technique to train a NEAT neural network as a wave manager to generate enemy waves that challenge the player’s defences. The approach is extended to incorporate NEAT-generated curriculums for tower deployments to gradually increase the difficulty for the generated enemy waves, allowing the neural network to learn incrementally. The approach dynamically adapts to changes in the player’s skill level, providing a more personalised and engaging gaming experience. The quality of the machine-generated waves is evaluated through a blind A/B test with the Games Experience Questionnaire (GEQ), and results are compared with manually designed human waves. The study finds no discernible difference in the reported player experience between AI and human-designed waves. The approach can significantly reduce the time and resources required to design game content while maintaining the quality of the player experience. The approach has the potential to be applied to a range of video game genres and within the design and development process, providing a more personalised and engaging gaming experience for players. © The Author(s) 2024.

关键词： Human computer interaction

来源：评论

学校读者我要写书评

暂无评论

Hyperspectral anomaly detection via memory‐augmented autoencoders

引用

CAAI Transactions on Intelligence technology 2023年第4期8卷 1274-1287页

作者： Zhe Zhao Bangyong Sun Faculty of Printing Packaging Engineering and Digital Media TechnologyXi'an University of TechnologyXi'anChina Norwegian Colour and Visual Computing Laboratory Norwegian University of Science and TechnologyGjovikNorway

Recently,the autoencoder(AE)based method plays a critical role in the hyperspectral anomaly detection ***,due to the strong generalised capacity of AE,the abnormal samples are usually reconstructed well along with the normal background ***,in order to separate anomalies from the background by calculating reconstruction errors,it can be greatly beneficial to reduce the AE capability for abnormal sample reconstruction while maintaining the background reconstruction performance.A memory‐augmented autoencoder for hyperspectral anomaly detection(MAENet)is proposed to address this challenging ***,the proposed MAENet mainly consists of an encoder,a memory module,and a ***,the encoder transforms the original hyperspectral data into the low‐dimensional latent ***,the latent representation is utilised to retrieve the most relevant matrix items in the memory matrix,and the retrieved matrix items will be used to replace the latent representation from the ***,the decoder is used to reconstruct the input hyperspectral data using the retrieved memory *** this strategy,the background can still be reconstructed well while the abnormal samples *** conducted on five real hyperspectral anomaly data sets demonstrate the superiority of the proposed method.

关键词： anomaly detection hyperspectral images memory autoencoder

来源：评论

学校读者我要写书评

暂无评论

Development of a new gas condensate viscosity model using artificial intelligence

引用

Journal of King Saud University - Engineering Sciences 2022年第7期34卷 376-383页

作者： Faraji, F. Ugwu, J.O. Chong, P.L. School of Computing Engineering and Digital Technology Teesside University Middlesbrough United Kingdom

Accurate estimation of gas condensate fluid properties is a challenging task due to the evolving condensate liquid from the gas phase below the saturation pressure. Among the fluid properties, viscosity of condensate liquid has the largest prediction uncertainty. The existing literature methods cannot cope with the nonlinearity and physics of gas condensate mixture (transition from a single phase to two phases) below the saturation pressure. Hence, in this study based on the experimental condensate viscosity data, a simple linear equation as a function of pressure, temperature and solution gas to oil ratio was developed. For this purpose, a comprehensive data source of 1368 experimental data points acquired from open literature has been used. For developing the new condensate viscosity correlation, an artificial intelligence (AI) method known as the Takagi–Sugeno–Kang (TSK) fuzzy algorithm was utilized. The accuracy of the developed correlation was compared with five previously published literature models. The superiority of the new correlation over the existing literature models is confirmed by statistical parameters of a least root mean square error of 0.0194, a mean average error of 0.0163 and an average absolute relative deviation percentage of 7.123. The proposed condensate viscosity correlation is valid in a pressure range of 0.25–75.84 MPa), a temperature range of 303–443.15°K and Rs of 41.96–3496 scf/STB. The proposed correlation can be used as an alternative approach to the existing models for accurately estimating the gas condensate viscosity and for conducting reliable reservoir simulation studies. © 2021 The Authors

关键词： Viscosity

来源：评论

学校读者我要写书评

暂无评论

Real-Time Sign Language to Speech Recognition System for BSL

Real-Time Sign Language to Speech Recognition System for BSL

引用

2024 Global Congress on Emerging Technologies, GCET 2024

作者： Abid, Hishaam Aldmour, Rakan Rehman, Ateeq Ur Asmat, Hamid University of Staffordshire School of Digital Technology Innovation and Business Dept. Computing United Kingdom

ISBN: (纸本)9798331542603

This paper presents a groundbreaking real-time system that transforms British Sign Language (BSL) gestures into audible speech using long short-term memory (LSTM), employing innovative machine learning techniques. The contribution, rooted in a blend of quantitative and qualitative research methods, showcases significant advances in technologies that assist BSL users. By integrating TensorFlow and MediaPipe, we designed a sophisticated multi-layer LSTM neural network that effectively handles real-time translation. Our approach encompassed several stages: gathering data, intensive preprocessing, meticulous training of the model, and thorough testing. The results from these tests confirmed the model's high accuracy and robust performance across various environments, marking a notable achievement in usability and technological application in real-life scenarios. Additionally, feedback from users indicated a more than 90% success rate in real-time gesture recognition, emphasizing the system's practical utility. This research not only bridges a vital communication gap for those who rely on BSL but also paves the way for future innovations, including integration into mobile and web platforms, thus expanding access and usability. The insights gained here highlight the profound impact machine learning can have on enhancing communication tools for the speech-impaired, resonating with ongoing trends in computational linguistics and interactive technologies © 2024 IEEE.

关键词： Interactive computer systems

来源：评论

学校读者我要写书评

暂无评论

Mitigating Security Vulnerabilities in Offline USSD Payments in non-Smartphones 11

Mitigating Security Vulnerabilities in Offline USSD Payments...

引用

11th IEEE International Conference on Software Defined Systems, SDS 2024

作者： Anunobi, Somtochukwu Memon, Mehak Howard, Christopher Lawal, Bashir University of Staffordshire School of Digital Technology Innovation and Business Department of Computing Stoke-on-Trent United Kingdom

ISBN: (纸本)9798331518325

This paper addresses security vulnerabilities in Unstructured Supplementary Service Data (USSD) based peer-to-peer (P2P) transactions, which often expose users to social engineering attacks and shoulder surfing due to unmasked Personal Identification Number (PIN) entries. We propose a novel P2P system that integrates unique wallet identifiers and partial PIN entry to mitigate these risks. Validation tests show a significant reduction in unauthorised access through shoulder surfing, with a success rate of just 0.05% under simulation conditions where the failed login throttle was ***, unique wallet identifiers significantly reduce the risk of social engineering attacks by eliminating the need to exchange personal information during transactions. Furthermore, we explore a voucher-based alternative for P2P money transfers within the USSD ***, our findings demonstrate that integrating unique wallet identifiers and partial PIN entry methods greatly enhances the security of USSD-based financial transactions, significantly reducing the risk of unauthorised access and social engineering attacks. These security improvements remain within acceptable USSD session timeout thresholds, strengthening user trust and paving the way for broader adoption of USSD-based financial services. © 2024 IEEE.

关键词： Economic and social effects

来源：评论

学校读者我要写书评

暂无评论

Optimizing Microservices Architecture: A Comprehensive Study on Resolving Complexity in Billing Systems 11

Optimizing Microservices Architecture: A Comprehensive Study...

引用

11th IEEE International Conference on Software Defined Systems, SDS 2024

作者： Lawal, Bashir Memon, Mehak Howard, Christopher Anunobi, Somtochukwu University of Staffordshire School of Digital Technology Innovation and Business Department of Computing Stoke-on-Trent United Kingdom

ISBN: (纸本)9798331518325

This study enhances microservices architecture optimization by introducing a novel approach that increases scalability and reduces complexity, particularly for billing systems. Through the implementation of a refined microservices framework, this research demonstrates significant improvements in system performance and manageability. Utilizing Java Spring Boot and Docker for the development and deployment of individual microservices, the system achieved marked enhancements in handling dynamic user demands and complex data workflows. Performance benchmarking revealed that the optimized system meets the needs of the industry-standard performance metrics, with advanced feature prototype showing a 20% improvement in memory efficiency and a 15% increase in CPU usage efficiency under peak load conditions compared to the traditional microservice architecture. Furthermore, the adoption of Domain-Driven Design (DDD) and event-driven communication has strategically addressed common issues related to service integration and data consistency, presenting a robust solution tailored to modern billing operations. This research contributes to the academic and practical discourse on microservices by providing a tested blueprint for scalable, flexible architecture that significantly mitigates the operational complexities often associated with large-scale billing systems. The implications of this study are broad, suggesting that similar approaches could be effectively applied in other sectors requiring dependable and scalable software solutions, thus paving the way for future innovations in microservices technology. © 2024 IEEE.

关键词： Large scale systems

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：