INTERANTIONAL JOURNAL OF SCIENTIFIC RESEARCH IN ENGINEERING AND MANAGEMENT最新文献_第5页

Enhancing Image Classification Using Few-Shot Learning Prototypical Networks with ResNet-18: Detection, Accuracy Enhancement, and Optimization 利用 ResNet-18 的少量学习原型网络加强图像分类：检测、准确度提升和优化

INTERANTIONAL JOURNAL OF SCIENTIFIC RESEARCH IN ENGINEERING AND MANAGEMENT Pub Date : 2024-07-24 DOI: 10.55041/ijsrem36755

Dr. S. M. Kulkarni, S. S. Pawar, A. A. Dekhane, S. L. Suryawanshi

{"title":"Enhancing Image Classification Using Few-Shot Learning Prototypical Networks with ResNet-18: Detection, Accuracy Enhancement, and Optimization","authors":"Dr. S. M. Kulkarni, S. S. Pawar, A. A. Dekhane, S. L. Suryawanshi","doi":"10.55041/ijsrem36755","DOIUrl":"https://doi.org/10.55041/ijsrem36755","url":null,"abstract":"Image classification, especially in scenarios with limited data, presents significant challenges. Few shot learning (FSL) aims to address these challenges by training models that can generalize from a few examples. This paper explores the integration of prototypical networks with ResNet-18 for feature extraction to enhance image classification accuracy. Prototypical networks are designed to create a prototype representation for each class, which can then be used to classify new examples based on their distance to these prototypes. By leveraging ResNet-18's powerful feature extraction capabilities, we aim to improve the quality of these prototypes, thereby enhancing classification performance.We propose various methods for accuracy enhancement and optimization, including hyperparameter tuning, regularization techniques, and advanced methods like attention mechanisms and metric learning. Hyperparameter tuning involves adjusting the model's parameters to find the optimal settings that yield the best performance. Regularization techniques, such as dropout and weight decay, help prevent overfitting and improve the model's generalization capabilities. Advanced methods like attention mechanisms can focus on the most relevant parts of the image, while metric learning aims to learn a distance metric that better reflects the similarities between images.Our experiments on datasets like Mini-ImageNet and Omniglot demonstrate significant improvements in classification performance. These datasets are commonly used benchmarks in the few-shot learning community, allowing us to compare our results with existing methods. The integration of prototypical networks with ResNet-18, along with the proposed optimization techniques, provides a robust approach for tackling the challenges of image classification in few-shot learning scenarios. Key Words: Few-shot learning, ResNet-18, Prototypical Networks.","PeriodicalId":504501,"journal":{"name":"INTERANTIONAL JOURNAL OF SCIENTIFIC RESEARCH IN ENGINEERING AND MANAGEMENT","volume":"47 6","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-07-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141809842","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

PRE - POST COVID 19 STOCK ANALYSIS OF ONGC COVID 19 前 - 后的 ONGC 股票分析

INTERANTIONAL JOURNAL OF SCIENTIFIC RESEARCH IN ENGINEERING AND MANAGEMENT Pub Date : 2024-07-24 DOI: 10.55041/ijsrem36772

Vikram Kumar, Dr. Somya Vatsnayan

{"title":"PRE - POST COVID 19 STOCK ANALYSIS OF ONGC","authors":"Vikram Kumar, Dr. Somya Vatsnayan","doi":"10.55041/ijsrem36772","DOIUrl":"https://doi.org/10.55041/ijsrem36772","url":null,"abstract":"The financial markets have been significantly influenced by Covid19. Investors have reallocated their portfolios as a result of changing expectations for risk and return. In both academia and industry, building a portfolio via wise stock selection has been seenas a problem. The stock market's inherent uncertainties are to blame for this. Stock selection in a portfolio is impacted by anticipated price movement. The predictability of stock price changes has been disputed for a very long time, however. The random walk hypothesis (Fama, 1995) states that since stock price changes are unpredictable and lack memory, the past cannot foretell the future. Therefore, if the market is efficient, the stock price at the moment represents all the information. Since insider trading is required, it is impossible to outperform the market and is compatible with EMH. Therefore, the quest for effective forecasting techniques does not lead to consistent, long-term trendsthat can be predicted. According to the findings, investors have begun redistributing their portfolios across other equities in response to the current financial crisis related to COVID-19. But not all investors experience the same situation when switching from risky to risk- free investments.","PeriodicalId":504501,"journal":{"name":"INTERANTIONAL JOURNAL OF SCIENTIFIC RESEARCH IN ENGINEERING AND MANAGEMENT","volume":"4 12","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-07-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141806448","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

AI Hardware Resource Monitoring in the Data Center Environment 数据中心环境中的人工智能硬件资源监控

INTERANTIONAL JOURNAL OF SCIENTIFIC RESEARCH IN ENGINEERING AND MANAGEMENT Pub Date : 2024-07-24 DOI: 10.55041/ijsrem36782

Nanduri Vijaya Saradhi

{"title":"AI Hardware Resource Monitoring in the Data Center Environment","authors":"Nanduri Vijaya Saradhi","doi":"10.55041/ijsrem36782","DOIUrl":"https://doi.org/10.55041/ijsrem36782","url":null,"abstract":"Deploying an AI (Artificial Intelligence) model in the data center initiates more responsibilities to the backend services such as Monitoring. It is required to monitor the performance of AI systems regularly to ensure that they meet the requirements and will not encounter any system performance issues. This whitepaper focuses on the importance of monitoring AI systems, the monitoring model, how to measure the performance of the system hardware resources such as CPU, Memory, disk and GPU, and tools to be used to monitor the system resources. Organisations can take necessary proactive maintenance actions before an incident is caused due to performance bottlenecks in the AI systems, proving the importance of monitoring the AI system. The goal of continuous monitoring of AI systems is to ensure the effective operation of AI systems throughout their lifecycle to meet several objectives such as performance, anomaly detection, security monitoring, data compliance and continuous improvements. Performance measurement of critical resources such as GPU, Memory and Storage by using suitable tools and configuring the alerts when the thresholds are reached on the identified resource threads. These measurements will be utilized to strengthen the AI system that will be stable for any performance bottlenecks.","PeriodicalId":504501,"journal":{"name":"INTERANTIONAL JOURNAL OF SCIENTIFIC RESEARCH IN ENGINEERING AND MANAGEMENT","volume":"4 4","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-07-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141809348","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Human-Computer Interaction Through Digital Virtual Navigation System 通过数字虚拟导航系统实现人机交互

INTERANTIONAL JOURNAL OF SCIENTIFIC RESEARCH IN ENGINEERING AND MANAGEMENT Pub Date : 2024-07-24 DOI: 10.55041/ijsrem36729

Saravana Kumar R, John Rohith J S, Sabarivasan S M, Mangalapriya S

引用次数: 0

CUSTOMER SEGMENTATION USING MACHINE LEARNING 利用机器学习进行客户细分

INTERANTIONAL JOURNAL OF SCIENTIFIC RESEARCH IN ENGINEERING AND MANAGEMENT Pub Date : 2024-07-22 DOI: 10.55041/ijsrem36658

Kiran D, A. C

{"title":"CUSTOMER SEGMENTATION USING MACHINE LEARNING","authors":"Kiran D, A. C","doi":"10.55041/ijsrem36658","DOIUrl":"https://doi.org/10.55041/ijsrem36658","url":null,"abstract":"Businesses need to identify and segment their consumer base in order to effectively customize their strategies and improve customer satisfaction in the highly competitive market landscape of today. The goal of this study is to employ machine learning techniques to create a strong consumer segmentation model that will classify customers according to their demographics, behaviors, and purchase histories. Through the use of multiple clustering methods, including K-means, DBSCAN, and Hierarchical Clustering, the model seeks to find unique customer segments with shared attributes. To accomplish optimal segmentation, the segmentation process entails three steps: feature selection to identify the most significant features, model training, and data preprocessing to manage missing values and outliers. In-depth segment analysis is also included in the report to offer practical insights for better client retention tactics, tailored recommendations, and focused marketing efforts. The results of the study illustrate how machine learning may be used to find hidden patterns in consumer data, giving organizations the ability to make data- driven decisions. Organizations may improve their marketing efforts, allocate resources more efficiently, and eventually increase customer engagement and profitability by putting this customer segmentation strategy into practice. Key Words: K-Means Clustering, Hierarchical Clustering, Density-Based Spatial Clustering of Applications with Noise (DBSCAN), Cluster Plotting, Heatmaps, customer relationship management (CRM) system.","PeriodicalId":504501,"journal":{"name":"INTERANTIONAL JOURNAL OF SCIENTIFIC RESEARCH IN ENGINEERING AND MANAGEMENT","volume":"29 5","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-07-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141817038","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Review on IoT-Driven Infusion Monitoring & Management System 物联网驱动的输液监控和管理系统回顾

INTERANTIONAL JOURNAL OF SCIENTIFIC RESEARCH IN ENGINEERING AND MANAGEMENT Pub Date : 2024-07-22 DOI: 10.55041/ijsrem36700

Srashti Shukla, Dr. Yashpal Singh, Dr. Meghana Mishra

{"title":"Review on IoT-Driven Infusion Monitoring & Management System","authors":"Srashti Shukla, Dr. Yashpal Singh, Dr. Meghana Mishra","doi":"10.55041/ijsrem36700","DOIUrl":"https://doi.org/10.55041/ijsrem36700","url":null,"abstract":"Effective intravenous (IV) infusion management and monitoring are essential in today's healthcare environment to guarantee patient safety and improve the standard of care. In order to overcome the drawbacks of conventional IV drip monitoring techniques, this study describes the creation of a contactless, Internet of Things-driven infusion monitoring and management system. With the use of cutting-edge sensor technology and Internet of Things connection, the suggested system combines real-time IV drip liquid level monitoring and control, delivering precise and ongoing updates. Healthcare practitioners may now monitor and analyse patient data remotely thanks to the gathered data's constant updating and transmission to the Think Speak cloud platform. With its combination of real- time data collecting, cloud-based analytics, and emergency alert features, the proposed contactless IoT-driven infusion monitoring and management system represents a substantial improvement in healthcare technology. Using the power of IoT and cloud computing, this invention seeks to enhance overall patient outcomes, optimize IV treatment administration, and decrease human monitoring efforts. To increase patient safety, the system has an emergency alarm mechanism based on GSM modules. This feature ensures that caregivers are notified promptly of any irregularities or critical IV drip levels, allowing for prompt treatment. By providing real-time signals, the device lessens the risks related to delayed reactions to IV fluid failure or depletion. Formulating a real-time model for IV drip liquid level monitoring is the main goal of the system. By employing non-invasive sensors to identify and record the drip rate and fluid levels, this gadget assures precise tracking without having physical contact with the IV apparatus. Keywords: Internet of Things, ESP32 microcontroller, Biomedical, GSM, Relay, LCD Display.","PeriodicalId":504501,"journal":{"name":"INTERANTIONAL JOURNAL OF SCIENTIFIC RESEARCH IN ENGINEERING AND MANAGEMENT","volume":"53 18","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-07-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141814814","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Evaluating Word Sense Disambiguation Techniques for Punjabi Language: A Comparative Analysis 评估旁遮普语的词义消歧技术：比较分析

INTERANTIONAL JOURNAL OF SCIENTIFIC RESEARCH IN ENGINEERING AND MANAGEMENT Pub Date : 2024-07-22 DOI: 10.55041/ijsrem36699

Gursewak Singh

引用次数: 0

WATER QUALITY PREDICTION USING MACHINE LEARNING TECHNIQUE 利用机器学习技术预测水质

INTERANTIONAL JOURNAL OF SCIENTIFIC RESEARCH IN ENGINEERING AND MANAGEMENT Pub Date : 2024-07-22 DOI: 10.55041/ijsrem36721

Er. P Nagalakshmi, Dr.P.Ganesh Kumar

引用次数: 0

Heart Disease Prediction Using Machine Learning Algorithms 利用机器学习算法预测心脏病

INTERANTIONAL JOURNAL OF SCIENTIFIC RESEARCH IN ENGINEERING AND MANAGEMENT Pub Date : 2024-07-22 DOI: 10.55041/ijsrem36570

Mahammad Sahil Khan, Asst.Prof. Archana Panda

{"title":"Heart Disease Prediction Using Machine Learning Algorithms","authors":"Mahammad Sahil Khan, Asst.Prof. Archana Panda","doi":"10.55041/ijsrem36570","DOIUrl":"https://doi.org/10.55041/ijsrem36570","url":null,"abstract":"Heart disease is a major issue that has become increasingly prevalent. According to current statistics, heart disease claims the life of one person every minute. In the last several years, one of the hardest problems facing the medical field is predicting heart disease. Reducing the death rate can be achieved with early detection of cardiac disease. Machine learning is the most effective approach to forecasting heart disease. This paper aims to create a lightweight, straightforward solution to detecting cardiac disease using machine learning. Machine learning can aid in heart disease prediction. This study analyzes several machine learning algorithms and performance indicators. This study compares cardiac disease detection methods using a publicly available dataset from the UCI machine learning repository. There are other datasets accessible, including the Switzerland and Cleveland databases. Here the dataset contains 303 patient records and 18 characteristics. The analysis shows that out of six machine learning algorithms, the Random Forest algorithm gives the best result with 94.50%. Keywords- cardiac disease detection, datasets, heart disease prediction, Machine Learning, Random Forest algorithm.","PeriodicalId":504501,"journal":{"name":"INTERANTIONAL JOURNAL OF SCIENTIFIC RESEARCH IN ENGINEERING AND MANAGEMENT","volume":"24 7","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-07-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141816442","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Innovative Approaches in Psoriasis Management: Integration of Novel Drug Delivery Systems, Herbal Medicine, and Lifestyle Modifications 牛皮癣治疗的创新方法：新型给药系统、草药和生活方式调整的整合

INTERANTIONAL JOURNAL OF SCIENTIFIC RESEARCH IN ENGINEERING AND MANAGEMENT Pub Date : 2024-07-22 DOI: 10.55041/ijsrem36650

Krunal Detholia

{"title":"Innovative Approaches in Psoriasis Management: Integration of Novel Drug Delivery Systems, Herbal Medicine, and Lifestyle Modifications","authors":"Krunal Detholia","doi":"10.55041/ijsrem36650","DOIUrl":"https://doi.org/10.55041/ijsrem36650","url":null,"abstract":"Psoriasis, a complex inflammatory skin disorder, is characterized by rapid skin cell proliferation resulting in thick, red patches covered with silvery scales. This review explores the multidimensional approach to managing psoriasis, which encompasses current therapies such as systemic medications, topical treatments, and phototherapy, all known to have potential side effects. Recognizing the associated increased risk of comorbidities like psoriatic arthritis, anxiety, and cardiovascular diseases among others, this paper also delves into the promising role of herbal medicines which are gaining popularity due to their accessibility, cost-effectiveness, and potential efficacy. Additionally, it highlights the emerging advancements in novel drug delivery systems including liposomes, nanostructured lipid carriers, and microneedles, aimed at enhancing treatment efficacy through improved drug targeting and reduced side effects. This comprehensive review seeks to provide valuable insights for the development of safer and more effective therapeutic strategies, offering a beacon of hope for those afflicted by this chronic condition and guiding future research in the field. Keywords: Psoriasis, novel drug delivery systems, liposomes, nanostructured lipid carriers, microneedles, herbal medicine, lifestyle modifications, biologic agents","PeriodicalId":504501,"journal":{"name":"INTERANTIONAL JOURNAL OF SCIENTIFIC RESEARCH IN ENGINEERING AND MANAGEMENT","volume":"35 14","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-07-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141816689","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0