检索结果-内蒙古大学图书馆

Data science and Business Systems (ICDSBS), International Conference on

作者： Rahul Arora Manan Agarwal Vinoth N A S Uma Maheswari KM Brindha R Computer Science and Engineering SRM Institute of Science and Technology Kattankulathur Chennai Department of Computing Technologies School of Computing College of Engineering and Technology SRM Institute of Science and Technology Kattankulathur Chennai

ISBN: (数字)9798331585600

ISBN: (纸本)9798331585617

The objective of the paper is to propose an approach real-time facial sentiment analysis framework called MoodLens, classifying emotions into seven categories: angry, disgust, fear, happy, neutral, sad, and surprise. By using Convolutional Neural Networks (CNNs) for automated feature extraction, the whole process can be done without any human input. This also solves the accuracy, scalability, and adaptability issues that traditional machine learning models have. Sculpted for application in the real world, MoodLens incorporates real-time streaming techniques, taking human-computer interaction to the next level in healthcare, entertainment, security, education, and *** framework is a strong and effective way to do sentiment analysis. It uses advanced preprocessing techniques, dropout layers to prevent overfitting, and a scalability design to classify things accurately in changing environments.

关键词： Human computer interaction YOLO Sentiment analysis Adaptation models Accuracy Scalability Computational modeling Feature extraction Real-time systems Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Deep Learning-Based Scientific Document Summarization Considering Citation

引用

SN computer science 2025年第4期6卷 1-13页

作者： Jyoti, Divya Mahato, Dharmendra Prasad Srivastava, Jyoti Department of Computer Science and Engineering National Institute of Technology Hamirpur Himachal Pradesh Hamirpur 177 005 India

The rapid expansion of scientific publications poses a significant challenge for researchers who need to stay updated on recent advancements and efficiently extract crucial information. Summarizing scientific documents can address this challenge by providing concise overviews that highlight the main contributions of the original texts. This study introduces a cutting-edge deep learning approach that employs the sBERT model to enhance the accuracy and relevance of document summaries. Our method treats extractive summarization as a classification task, implementing a dual BERT model configuration to effectively pinpoint and extract critical information. Tested on the CL-SciSumm dataset, our approach demonstrates marked improvements over existing techniques, achieving higher ROUGE scores. These results affirm the potential of our model to generate accurate and dependable summaries, assisting researchers in managing and interpreting the growing volume of scientific literature. © The Author(s), under exclusive licence to Springer Nature Singapore Pte Ltd. 2025.

关键词： BERT and sBERT model Citation based summarization Deep learning Natural language processing Scientific documentation Spacy

来源：评论

学校读者我要写书评

暂无评论

Fixed-point quantum continuous search algorithm with optimal query complexity

引用

science China(Physics,Mechanics & Astronomy) 2025年第6期68卷 80-91页

作者： Shan Jin Yuhan Huang Shaojun Wu Guanyu Zhou Chang-Ling Zou Luyan Sun Xiaoting Wang Institute of Fundamental and Frontier Sciences University of Electronic Science and Technology of ChinaChengdu 610051China Key Laboratory of Quantum Information CASUniversity of Science and Technology of ChinaHefei 230022China Center for Quantum Information Institute for Interdisciplinary Information SciencesTsinghua UniversityBeijing 100084China Department of Electronic and Computer Engineering The Hong Kong University of Science and TechnologyHong Kong 999077China Hefei National Laboratory Hefei 230088China

Continuous search problems(CSPs),which involve finding solutions within a continuous domain,frequently arise in fields such as optimization,physics,and *** discrete search problems,CSPs require navigating an uncountably infinite space,presenting unique computational *** this work,we propose a fixed-point quantum search algorithm that leverages continuous variables to address these challenges,achieving a quadratic *** by the discrete search results,we manage to establish a lower bound on the query complexity of arbitrary quantum search for CSPs,demonstrating the optimality of our *** addition,we demonstrate how to design the internal structure of the quantum search oracle for specific ***,we develop a general framework to apply this algorithm to a range of problem types,including optimization and eigenvalue problems involving continuous variables.

关键词： quantum search quantum algorithm quantum computation optimization spectral problems

来源：评论

学校读者我要写书评

暂无评论

Diabetic Retinopathy Detection using Deep Learning Model ResNet15 2

Diabetic Retinopathy Detection using Deep Learning Model Res...

引用

2nd International Conference on Disruptive Technologies, ICDT 2024

作者： Rai, Bipin Kumar Ojha, Harsh Srivastava, Ishani Dayananda Sagar University Department of Computer Science & Engineering Bengaluru India Abes Institute of Technology Department of Computer Science & Engineering- Data Science U.P. Ghaziabad India

ISBN: (纸本)9798350371055

Diabetic Retinopathy stands out as one of the common retinal diseases which has a significant threat to vision and can lead to blindness. Several problems occurred due to DR can be stopped by controlling blood glucose and timely treatment, but the manual assessment of retinal images is often a lengthy process and can prevail human errors. This paper attempts in the direction of finding an automated way to classify the set of retinal images. The study starts with the collection of retinal images and pre- processing them, followed by training of a CNN model, Utilizing Deep Learning methodologies, CNN classifies the degree of diabetic retinopathy (DR) into four categories: 0 (no DR) to 4 (proliferative DR). We evaluate our model's performance using a diverse dataset. We use Convolutional Neural Networks (CNNs) strength to DR detection, which involves three major difficulties: Classification, Segmentation and Detection. We work towards developing a GUI based system to store and maintain a series of predictions with the patient's id and name. Twilio API have been used to make SMS connectivity to patients possible in case they are not accessible. Many research activities in the previous years have many limitations and disadvantages like models used other than Resnet have low accuracy, not capturing intricate features and patterns in images, Vanishing Gradient Descent and low performance for image classification. Resnet models that are previously used in DR detection with Resnet152, we can conclude that it has the lowest Training loss, lowest Test loss and it has the lowest loss Difference. We have worked on a RESNET-152 pre-trained deep learning model, which consists of thousands of layers and allows for the capturing of intricate features, high performance of image classification, and high accuracy, in order to overcome the issues and limitations of earlier developed models. © 2024 IEEE.

关键词： Deep Learning Diabetic Retinopathy (DR) Fundus images Ophthalmologist Proliferative DR Resnet152

来源：评论

学校读者我要写书评

暂无评论

Text Categorization into Emotional States Using Enhanced Deep Learning Model

Text Categorization into Emotional States Using Enhanced Dee...

引用

Cognitive Computing in engineering, Communications, sciences and Biomedical Health Informatics (IC3ECSBHI), International Conference on

作者： P. Sasanka C. Bharath D. Hemateja Saravanan Chandrasekaran Department of Computer Science and Engineering SRM Institute of Science and Technology Ramapuram Chennai India

ISBN: (数字)9798331518523

ISBN: (纸本)9798331518530

The growing reliance on online platforms in our daily lives has made sentiment analysis a crucial area in pattern recognition and natural language processing (NLP). In this field, user feedback on various topics is evaluated and analyzed. One of the key challenges in modern NLP is identifying emotions within text. The proposed model aims to extract emotions from text using BERT, which is built by integrating transformers with bidirectional encoder representations. While sentiment analysis helps by identifying the polarity of messages, certain applications require emotion detection, as sentiment analysis alone may not capture the full range of emotions. To truly understand human psychology, it is essential to analyze data as quickly as it is generated. BERT is employed to train the model, and a comparative study showed that BERT outperforms other state-of-the-art models by better understanding word context. The findings of this study revealed that the BERT model achieved an impressive accuracy of 96%.

关键词： Deep learning Training Sentiment analysis Emotion recognition Accuracy Text categorization Bidirectional control Transformers Encoding Context modeling

来源：评论

学校读者我要写书评

暂无评论

A Progressive Approach of an Augmented Steganography for Network Peers

A Progressive Approach of an Augmented Steganography for Net...

引用

2024 International Conference on Electrical, Electronics and Computing Technologies, ICEECT 2024

作者： Kumar, B. Pavan Monika, N. Sravani, T.C. Sasi Preetham, C. Varuntej Reddy, V. Kumar Gupta, Manish Siddharth Institute of Engineering & Technology Department of Computer Science and Engineering Andhra Pradesh Chittoor India

ISBN: (纸本)9798350378092

In order to interface and exchange data with various devices and frameworks via the internet, real objects or things implanted with sensors, programming, and other advancements are referred to as the Internet of Things (IoT). While there are IoT devices [13] that are publicly accessible to everyone in the company, it is still important to understand the risks associated with digital attacks and security threats. Before a message is sent, it is entirely converted from plain to encoded in cryptography. The reverse also applies to plaining messages received following exchanges of letters. One method of hiding sensitive data is to embed it within a text, image, video, or sound file via a technique called steganography. Information can be hidden using bits that address comparable variety pixels that are repeatedly rehashed in an image record. A picture record that seems identical to the original image but contains 'commotion' samples of typical, decoded information will be produced by subtly applying the encoded information to this repeating information. The goal of this project is to jumble up the information on IoT networks by using an image information concealing technique to hide the message inside a picture record. We plan to integrate the use of convolutional neural networks with standard image information hiding techniques in order to significantly increase the amount of data that can be transmitted through an image. We will investigate various convolutional bounds in order to achieve the highest payload. The information will be encrypted and unscrambled using the recently developed profound learning computation. As a result, the convolutional organizations involved in this project will be set up to increase the payload of the data that needs to be encrypted and safely decoded in order to view the initial message. © 2024 IEEE.

关键词： Convolution

来源：评论

学校读者我要写书评

暂无评论

A Deep Learning Based Approach for Detecting Mango Leaf Diseases 6

A Deep Learning Based Approach for Detecting Mango Leaf Dise...

引用

6th International Conference on Computational Intelligence and Networks, CINE 2024

作者： Pradhan, Astik Kumar Rout, Jitendra Kumar Ghosh, Debanjan Sinha, Shailendra Kumar Arka Jain University School of computer Science and IT Jamshedpur India National Institute of Technology School of computer Science and Engineering Raipur India

ISBN: (纸本)9798331516796

The mango industry faces substantial annual economic losses due to diseases and pests. Farmers encounter difficulties in differentiating between various mango diseases, as their symptoms can closely resemble each other and even occur simultaneously. By making it possible to quickly and accurately detect and identify mango leaf illnesses, this study aims to offer a remedy. To accomplish this, the research employs state-of-the-art Convolutional Neural Networks (CNNs), which are advanced learning algorithms capable of automatically extracting and understanding complex features from raw images. Specifically, the study utilizes EfficientNet, a model based on CNNs, to facilitate the early identification and classification of mango leaf diseases. Additionally, advanced data augmentation techniques such as rotation, reflection, translation, and scaling are applied to enhance the collected dataset. This enhanced data is then used to train the EfficientNet model, which achieves an amazing 98% accuracy rate in identifying and categorizing mango leaf diseases. These findings underscore the CNN method's effectiveness in detecting a wide range of mango leaf diseases, positioning it as an invaluable and practical tool for both farmers and agricultural experts. © 2024 IEEE.

关键词： Fruits

来源：评论

学校读者我要写书评

暂无评论

Energy Efficient Analysis of Electric Vehicle Motor under Drive Cycle Influence

Energy Efficient Analysis of Electric Vehicle Motor under Dr...

引用

2023 IEEE International Conference on Power Electronics, Smart Grid, and Renewable Energy, PESGRE 2023

作者： Malarselvam, V. Carunaiselvane, C. Srm Institute of Science and Technology Electrical Vehicle Technology Lab Department of Automobile Engineering Chennai India

ISBN: (纸本)9798350310573

Electric vehicle is one way to prevent harmful emission contributing to air pollution and that makes globally migrate from conventional to electrified transportation. To ensure optimal performance, longevity, and cost-effectiveness of the electric vehicle architecture, the primary and crucial component is the selection of an electric motor. The selected motor should possess suitable power and other characteristics specifically tailored for traction purposes, considering the driving cycle. The efficiency of the motor varies when it is operated subjected to traction application than operated in isolation mod which is due to various dynamic factors. Generally, the efficiency is predicted based on the vehicle dynamics considering the maximum traction power required. But the average efficiency of the vehicle over the entire driving cycle is more important than the efficiency over single operating point. Moreover, when the motor is operated for traction purpose it is operated at 60% power for maximum run time than the rated power. This paper aims to compare and analyze the energy efficiency of traction motor including and excluding for New European driving cycle (NEDC). These articles presents a 250 W, 415 rpm, 54S/60P Electric vehicle BLDC motor and study its performance over the driving cycle. Furthermore, this research attempts to suggest de-rated motor for the same vehicle through energy efficient analysis. © 2023 IEEE.

关键词： BLDC motor driving cycle finite element analysis vehicle driving performance simulation

来源：评论

学校读者我要写书评

暂无评论

Efficient Information Extraction from Medical Records 1st

Efficient Information Extraction from Medical Records

引用

1st International Research Conference on Computing Technologies for Sustainable Development, IRCCTSD 2024

作者： Anand, G. Paavai Sundaram, Rishi Raja, S. Karthikk Kalpana, P. Department of Computer Science and Engineering SRM Institute of Science and Technology Vadapalani Chennai India

ISBN: (纸本)9783031823886

Over the years, information extraction has undergone numerous significant changes. This also holds true for the medical field. Medical transcripts are valuable resources filled with information that can help medical professionals and patients with insights. However, extracting this data manually is not efficient. Automating or minimalizing the process is required to perform efficiently. We aim to achieve No Code data extraction from medical transcripts to achieve inclusivity. To achieve this, we fine-tune the model using curated datasets. LLMs are powerful learners, but lack specific knowledge. By feeding them medical data like research papers and clinical documents, we adjust their internal understanding towards the medical domain. This lets them recognize medical terms, grasp relationships between diseases and treatments, and understand the context of medical queries. So, when we ask a question, the fine-tuned LLM can search through the data it learned to give us an accurate and relevant answer [5, 6]. We evaluate the model’s performance using RAG metrics for data retrieval. Compared to Mixtral 7B, our proposed model shows 8% improvement in MMLU performance. Translation being the only weak section of our proposed model as it is trained for data retrieval and not translation. The fine-tuning process is carried out with fixed parameters listed in the PEFT section. Methodology of training is PEFT and for learning, reinforcement learning with a learning rate of 1e−4. This value is experimental and the relation between learning rate and results vary due to difference in nature and quality of the training data. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

关键词： Reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

Identification of Real or Fake Currency Using Semi-supervised Generative Adversarial Networks (SSGANs) 6th

Identification of Real or Fake Currency Using Semi-supervise...

引用

6th International Conference on Computational Intelligence in Pattern Recognition, CIPR 2024

作者： Mukherjee, Anupam Ghosh, Anupam Department of Computer Science & Engineering Siliguri Institute of Technology West Bengal Siliguri India Department of Computer Science & Engineering Netaji Subhash Engineering College West Bengal Kolkata India

ISBN: (纸本)9789819780891

Traditional paper money and contemporary electronic money are two significant forms of exchange. However, due to the new and improved methods that counterfeiters are using, it is now becoming an increasingly important issue. This work suggests a machine-assisted method that has been created to distinguish between fake and real currency. Generative adversarial networks (GANs), a cutting-edge machine learning model, are used for this purpose. To develop a model that can be used for supervised prediction, GANs utilize semi-supervised learning. Semi-supervised learning is an exceptional approach in building a model that is fit for supervised prediction. It offers a multitude of benefits, allowing the training of unlabeled data while achieving highly accurate predictions. In this work, detecting counterfeit money using GANs involves a multi-step methodology that leverages the power of deep learning for accurate identification. Preprocess the data by resizing, normalizing, and augmenting the images to ensure uniformity and improve model generalization. The discriminator minimizes the loss when making correct classifications, while the generator minimizes the loss when the discriminator classifies its generated images as genuine. Indian banknotes were subjected to this procedure. Modern feature identification and image processing techniques were employed to develop the overall strategy of a valid input. The trials demonstrate that a high-precision system may be created to identify real paper money. During the training process, the generator network and the discriminator networks update their loss functions. Once training is complete, the model can accurately identify real money notes from fake ones with an 80% accuracy rate. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： Generative adversarial networks

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：