检索结果-内蒙古大学图书馆

Comparison of Novel Long Short-Term Memory Over Recurrent Neural Networks for Predicting the Melting Pattern of Ice in Antarctic Sea Based on Accuracy

Comparison of Novel Long Short-Term Memory Over Recurrent Ne...

引用

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

作者： Shah, Chirag Suresh Jaisharma, K. Saveetha University Researh Scholar Department of Computer Science and Engineering Saveetha School of Engineering Saveetha Institute of Medical and Technical Sciences Tamilnadu Chennai India Saveetha University Saveetha Institute of Medical and Technical Sciences Department of Computer Science and Engineering Saveetha School of Engineering Tamilnadu Chennai India

ISBN: (纸本)9798331527518

Revolutionary technique that blends revolutionary Novel Long Short-Term Memory (NLSTM) with Recurrent Neural Network (RNN), this research seeks to revolutionize the prediction of ice melting. This research paper's dataset, which has 212 rows and 3 columns (attributes), was obtained from ***. This has to do with Antarctic mass, year, and 1-sigma uncertainty. The clincalc web application was utilized to determine a total sample size of 92 for the paper. Gpower sets a value of α=0.05 and power=0.80 to find a sample size of 20 for each group. Based on the data obtained from the two groups NLSTM has a much greater accuracy (98.64%) as compared to that of RNN accuracy (96.78%). Statistical significance was noted between NLSTM and RNN with p = 0.002 (Independent Sample T-Test, p © 2025 IEEE.

关键词： Long short-term memory

来源：评论

学校读者我要写书评

暂无评论

Obesity Prediction and Nutrition Recommendation using KNN and Random Forest algorithm for Increased Accuracy 8

Obesity Prediction and Nutrition Recommendation using KNN an...

引用

8th IET Smart Cities Symposium, SCS 2024

作者： Vineetha, B. Surendran, R. Logapriya, E. Department of computer science and Engineering Saveetha School of Engineering Saveetha Institute of Medical and Technical Science Chennai India

ISBN: (数字)9781837243105

ISBN: (纸本)9781837242672

Aim: The purpose of this study is to predict obesity using KNN algorithm compared with Random Forest algorithm. This research paper focuses on the creation of a novel method for obesity prediction and recommending nutrition for individuals. Making use of KNN and Random Forest machine learning algorithms, this study builds an accurate obesity prediction model for a data source. Materials and Methods: KNN algorithm and Random Forest algorithm are the two algorithms including 10 samples each with, and 80% G power, alpha value of 0.05 and beta value of 0.2 is being taken into account. Findings: While KNN usually has a lower accuracy rate of 67%, Random Forest method consistently obtains a 95% rate. Evidence suggests that, when comparing accuracy, the Random Forest method is superior to the KNN algorithm. © The Institution of engineering & Technology 2024.

关键词： Random forests

来源：评论

学校读者我要写书评

暂无评论

Intelligent growth of holy basil: an iot-based hydroponics system empowered by the CRL-SHS Algorithm

引用

Earth science Informatics 2025年第2期18卷 1-20页

作者： Kumaragurubaran, T. Chiranjeevi, Rahul D, Elangovan K, Deepak Kumar Department of Computer Science and Engineering Saveetha School of Engineering Saveetha University SIMATS Tamil Nadu Chennai India Department of Computer Science and Engineering R.M.D. Engineering College Tamil Nadu Kavaraipettai India Department of Computer Science and Engineering Saveetha Engineering College Thandalam Tamil Nadu Chennai India Department of Computer Science and Engineering Panimalar Engineering College Poonamalee Tamil Nadu Chennai India

The soilless method of growing plants, known as hydroponics, has gained attention for its ability to increase crop yields while conserving resources. This paper presents a novel approach integrating Internet of Things technology with a Convolutional Reinforcement Learning-based Seahorse Search algorithm to optimize plant growth efficiency in hydroponic environments. The Convolutional Reinforcement Learning-based Seahorse Search architecture combines a Convolutional Neural Network for Extracting critical environmental features. Reinforcement Learning for adaptive control, and the Seahorse Search algorithm for hyperparameter optimization. The Convolutional Neural Network component enables precise feature extraction, ensuring accurate data-driven decisions. Reinforcement Learning introduces a self-adaptive mechanism, dynamically adjusting environmental conditions to meet plant growth needs, while the Sea Horse Search algorithm fine-tunes parameters to reduce resource wastage and prevent overfitting. Together, these components create a responsive and sustainable hydroponic system capable of optimizing plant health and yield. Experimental results using five hydroponic datasets such as the Lettuce growth days dataset, Hydroponic plants nutrition dataset, Hydroponic and soil compound dataset, nutrient solution and Pakchoi dataset, and Vertical farming dataset demonstrate the system’s ability. The proposed model attains a higher accuracy value of 99.11% in environmental condition management, with significant resource savings in water and nutrient usage. This proposed hydroponic system offers a scalable, automated solution for sustainable agriculture, enabling precise and efficient crop production. Through this work, we contribute to advancing Internet of Things-driven smart farming and highlight the potential of Artificial Intelligence-based optimization to revolutionize resource management in agriculture. © The Author(s), under exclusive licence to Springer-Verlag GmbH Ger

关键词： Artificial intelligence Cloud decision making Convolutional neural network Hydroponic farm Hydroponics system Reinforcement learning Seahorse search

来源：评论

学校读者我要写书评

暂无评论

IoT-Driven Energy Consumption Optimization in Smart Homes 2

IoT-Driven Energy Consumption Optimization in Smart Homes

引用

2nd IEEE International Conference on Trends in Quantum Computing and Emerging Business Technologies, TQCEBT 2024

作者： Loganayagi, S. Hemavathi, R. Jayalakshmi, D. Vimal, V.R. Kanthan Narayanan, Lakshmi Saveetha Institute of Medical and Technical Sciences Saveetha University Saveetha School of Engineering Department of Computer Science and Engineering Chennai India Saveetha University Saveetha School of Engineering Department of Cse Thandalam Chennai India Saveetha Institute of Medical and Technical Sciences SIMATS Saveetha University Saveetha School of Engineering Department of Computer Science and Engineering Chennai India

ISBN: (纸本)9798350384277

Smart homes that optimize energy use have emerged from the Internet of Things (IoT). IoT-driven energy optimization in smart homes has several benefits, as this study shows. Our study found that IoT-driven energy consumption optimization can save smart homes 15% of their energy use. This reduction saves homes $150 a year on energy expenditures. These savings also reduce CO2 emissions by 500 kg, meeting sustainability targets. A staggering 90% of users are satisfied with IoT-driven energy optimization. The usercentric approach allows homeowners to customize their energy choices, as 70% of users do so. The study also shows that these systems meet 80% of customers' energy goals. IoT-driven systems anticipate energy usage with 95% accuracy, demonstrating their reliability. The excellent 98% device compatibility ensures easy device connection. The study also shows that privacy and security safeguards work, with no data breaches. With a 500 kg CO2 reduction, IoT-driven energy usage optimization has a significant environmental impact. A 20% energy waste reduction over five years is expected for long-term sustainability. Up to 10% of energy is renewable, contributing to a greener future. © 2024 IEEE.

关键词： Smart homes

来源：评论

学校读者我要写书评

暂无评论

Machine Learning-Enhanced Shatavari Recommendations for Menstrual Health 8

Machine Learning-Enhanced Shatavari Recommendations for Mens...

引用

8th IET Smart Cities Symposium, SCS 2024

作者： Logapriya, E. Surendran, R. Department of computer science and Engineering Saveetha School of Engineering Saveetha Institute of Medical and Technical Science Chennai India

ISBN: (纸本)9781837242672

The application of machine learning techniques to provide personalized Shatavari (Asparagus racemosus) menstrual cramps and other symptoms cause great suffering to many women. This research focuses on the application of machine learning algorithms for personalised Shatavari diet based on women menstrual cycle symptoms. To the attempt is made to analyse the results of various menstrual problems and their response to the consumption of Shatavari. This research aimed at predicting by how much the women’s diet would change if they were to take Shatavari pills. Some of the ML models that have been adopted include the Decision Trees, Random Forest, Logistic Regression, Support Vector Machine (SVM), and Neural Networks. While training and evaluating the model performance it featured characteristics, Precision, accuracy, recall, F1-score. Neural Network model was more effective to other approaches and successfully predict that usage of Shatavari would be helpful to minimize menstruation discomfort by 97%. The findings are very positive for the treatment of menstruation ailments and overall women care and they add to the evidence about the need for unique formulation of Shatavari. Possible to combine traditional knowledge about herbal nutrition with modern approaches such as machine learning to improve women wellness as presented in this recommendation. © The Institution of engineering & Technology 2024.

关键词： Nutrition

来源：评论

学校读者我要写书评

暂无评论

Image Enhancment Hair Removal in Dermoscopy Images Using Dual Swin Autoencoder 2

Image Enhancment Hair Removal in Dermoscopy Images Using Dua...

引用

2nd International Conference on Self Sustainable Artificial Intelligence Systems, ICSSAS 2024

作者： Malathi, M. Fernandez, F. Mary Harin Saveetha Institue of Medical and Technical Science Saveetha University Saveetha School of Engineering Department of Computer Science and Engineering Chennai India

ISBN: (纸本)9798350368413

Skin cancer requires effective treatment for better survival rates especially in the case of melanoma. To reduce the image noise and removing hair the conventional method takes high computational complexity and decayed accuracy. In order to tackle this issue, an innovative dual Swin autoencoder has been proposed in this paper. This autoencoder utilizes Swin Transformer blocks to effectively eliminate noise from images and get rid of unwanted hair, all while maintaining important characteristics of the skin and lesions. Experimental results demonstrate that our model outperforms conventional techniques, with PSNR values up to 81.71 dB and SSIM up to 0.99, effectively eliminating hair and retaining important skin features. This approach enhances the accuracy of melanoma diagnosis in dermoscopy images. © 2024 IEEE.

关键词： Oncology

来源：评论

学校读者我要写书评

暂无评论

Multi-Scale CNN Architectures for Enhanced Feature Extraction in Image Sentiment Analysis 3

Multi-Scale CNN Architectures for Enhanced Feature Extractio...

引用

3rd International Conference on Electronics and Renewable Systems, ICEARS 2025

作者： Indhumathi, S. Fernandez, F. Mary Harin Saveetha Institute of Medical and Technical Sciences Saveetha University Saveetha School of Engineering Department of Computer Science and Engineering Chennai India

ISBN: (纸本)9798331509675

Imagery sentiment analysis is a complex task, the aim of which is the interpretation of both local and global visual signs in order to correctly classify the emotions ordered by the image. CNNs old architectures commonly have trouble with the proper balance of the tip-oriented extraction and the overall understanding of the context. This research suggests a new multi-scale CNN architecture in order to improve the extraction of features by the hierarchical representations of the visual sentiment cues. The architecture operates by adding convolutional layers of nucleotide scale and pyramid pooling modules to take features from different spatial resolutions. Besides, the attention mechanism is applied to the sentimental areas, thus dealing with the noise from unimportant parts. That is modeling a CNN with multi-scale convolutional layers in the construction of an image where each kernel size will deliver what is a fine signal versus a contextual pattern. A pooling module based on a pyramid structure is collected to the resolution to obtain features from different spatial scales, so the model will have the view of the entire image at a context level. Then, in addition, the proposed attention mechanism that finds and highlights regions with more probability of emotional content is used. The suggested architecture is prepared and tested on a number of datasets, that reflect the real-world sentiment contexts and the evaluation metrics like accuracy, precision, recall, and F1-score, which show the effectiveness of the proposed architecture. The proposed approach seeks to give a more in-depth definition of visual sentiment which, in turn, will lead to more precise classification as well as allow for better understanding with attention-based visualizations. The results are anticipated to drive the image sentiment analysis field further, making it relevant for social media analytics, advertising, and human-computer interaction. © 2025 IEEE.

关键词： Emotion Recognition

来源：评论

学校读者我要写书评

暂无评论

Enhancing Accuracy in Image Recognization for Deepfake Images Using Convolutional Networks Comparing with K-Nearest Neighbors (KNN) Algorithm

Enhancing Accuracy in Image Recognization for Deepfake Image...

引用

2025 International Conference on Multi-Agent Systems for Collaborative Intelligence, ICMSCI 2025

作者： Chanakya, A.S. Surendran, R. Madhusundar, N. Saveetha School of Engineering Saveetha Institute of Medical and Technical Sciences Department of Computer Science and Engineering Chennai India

ISBN: (数字)9798331509828

ISBN: (纸本)9798331509828

The primary goal of this research is to improve the accuracy of deepfake picture recognition using Convolutional Neural Networks (CNN) and compare the results to a K-Nearest Neighbors (KNN) approach. This entails researching approaches like transfer learning, feature extraction, and model evaluation to increase the robustness and efficiency of deepfake detection systems. For the training and evaluation phases, a Kaggle dataset with both authentic and deepfaked photos was used. The dataset included samples from several deepfake generating techniques and scenarios. This work involved the implementation of two independent groups, each applying different algorithmic approaches for picture identification. Group 1 used Convolutional Neural Networks (CNNs), while Group 2 used a neural network design with the K-Nearest Neighbors (KNN) algorithm. Each group had 20 samples, for a total sample size of 40. This sample size was chosen to ensure enough statistical power for subsequent studies. *** was used for a priori power analysis, and the statistical power (G-power) was set at 80%. The chosen significance level (alpha, α) was 0.05, and the type II error rate (beta, β) was 0.2, A 95% confidence interval was selected to assess the precision of the estimated effectiveness of the two algorithmic approaches. Result: The accuracy of the K-Nearest Neighbors (KNN) algorithm was 72.8%, whereas the accuracy of the Convolutional Neural Networks (CNNs) was 97.15%. A two-tailed significance test confirmed the superiority of the CNN model over the KNN model, with a statistically significant p-value of 0.001 (p 0.05) for accuracy. © 2025 IEEE.

关键词： k-nearest neighbors

来源：评论

学校读者我要写书评

暂无评论

Enhancing E-Commerce Predictive Analytics using Ensemble Models for Consumer Behaviour and Market Forecasting 1

Enhancing E-Commerce Predictive Analytics using Ensemble Mod...

引用

1st International Conference on Intelligent Systems and Computational Networks, ICISCN 2025

作者： Balaji, S. Manikavelan, D. Saveetha University Saveetha School of Engineering Saveetha Institute of Medical and Technical Sciences Department of Computer Science and Engineering Chennai India

ISBN: (纸本)9798331529246

Predictive analytics is becoming more and more crucial in the fast-changing e-commerce sector, which helps businesses predict consumer behavior, enhance marketing, and optimize operational efficiency. This project aims to improve predictive analytics in e-commerce using machine learning techniques. AI is a fast-developing field that plays a significant role in data analysis and forecasting, allowing businesses to make informed decisions. The research utilizes algorithmic learning that involves ensemble models such as Kernel Support Vector Machine, Kernel Principal Component Analysis, and a Mentor-Learner based Optimization model. The main aim is to understand the potential growth and prospects of online shopping in the future. To achieve this, secondary data on ecommerce valuation was analyzed to generate revenue insights and understand online customer behavior better. Then, Python was applied to predict the sector's contribution from 2024 to 2026. The study also incorporates factors such as the proportion of Indian online consumers and e-commerce revenue to improve market price prediction accuracy. By combining academic research with industry practices, this paper offers valuable recommendations for online shopping managers and executives seeking to implement neural networks for analytical forecasting. The MAPE for KPCA-SVM-MLBO was 0.189%, significantly lower than the MAPE values of 0.769% and 0.547% for PCA-SVM-PSO and SVM-MLBO, respectively. © 2025 IEEE.

关键词： Predictive analytics

来源：评论

学校读者我要写书评

暂无评论

Optimizing Air Pollution Prediction in Urban Environments using a Hybrid RNN-PBO Model with IoT Data

Optimizing Air Pollution Prediction in Urban Environments us...

引用

2025 International Conference on Visual Analytics and Data Visualization, ICVADV 2025

作者： Mohandas, Prajul Subramanian, P. Surendran, R. Saveetha School of Engineering Saveetha Institute of Medical and Technical Sciences Department of Computer Science and Engineering Chennai602105 India Saveetha School of Engineering Saveetha Institute of Medical and Technical Sciences Department of Computer Science and Engineering Tamil Nadu Chennai602105 India

ISBN: (纸本)9798331521394

This research study proposes a novel hybrid model based on Recurrent Neural Network (RNN) combinatorial with Polar Bear Optimization (PBO) to improve air pollution prediction and control measures over cities. The model aims at learning the temporal relationships and interdependence among air-quality data. To implement the model styled with real-time Internet of Things (IoT) sensor data, tracking pollutants and other environmental factors in the urban region of study, the sequential nature of the RNNs allows it to build model patterns yet has a slow convergence and local minima when learning. To address this challenge, the PBO algorithm is introduced to optimize the parameter weight of the RNN, thus enhancing learning capacity and prediction accuracy. The exploration and exploitation of the PBO allows an effective mechanism of saving time with fast convergence and better process of complex data. The hybrid method was validated on an actual urban data set, giving improved prediction accuracy and stability over traditional RNNsand other optimization schemes. The proposed model aspires to deliver an efficient and trustworthy methodology for real-time air quality forecasting smart city management and a healthier city. © 2025 IEEE.

关键词： Air quality

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：