检索结果-内蒙古大学图书馆

38th Conference on Neural Information Processing Systems, NeurIPS 2024

作者： Wu, Fang Jin, Shuting Li, Siyuan Li, Stan Z. Computer Science Department Stanford University United States School of Computer Science and Technology Wuhan University of Science and Technology China School of Engineering Westlake University China

Machine learning catalyzes a revolution in chemical and biological science. However, its efficacy heavily depends on the availability of labeled data, and annotating biochemical data is extremely laborious. To surmount this data sparsity challenge, we present an instructive learning algorithm named InstructMol to measure pseudo-labels' reliability and help the target model leverage large-scale unlabeled data. InstructMol does not require transferring knowledge between multiple domains, which avoids the potential gap between the pretraining and fine-tuning stages. We demonstrated the high accuracy of InstructMol on several real-world molecular datasets and out-of-distribution (OOD) benchmarks. © 2024 Neural information processing systems foundation. All rights reserved.

关键词：

来源：评论

学校读者我要写书评

暂无评论

DeBERTa-GRU: Sentiment Analysis for Large Language Model

引用

computers, Materials & Continua 2024年第6期79卷 4219-4236页

作者： Adel Assiri Abdu Gumaei Faisal Mehmood Touqeer Abbas Sami Ullah Department of Informatics for Business College of BusinessKing Khalid UniversityAbha61421Saudi Arabia Department of Computer Science College of Computer Engineering and SciencesPrince Sattam bin Abdulaziz UniversityAl-Kharj11942Saudi Arabia School of Electrical and Information Engineering Zhengzhou UniversityZhengzhou450001China Department of Computer Science and Technology Beijing University of Chemical TechnologyBeijing100029China Department of Computer Science Government College University FaisalabadFaisalabadPunjab38000Pakistan

Modern technological advancements have made social media an essential component of daily *** media allow individuals to share thoughts,emotions,and *** analysis plays the function of evaluating whether the sentiment of the text is positive,negative,neutral,or any other personal emotion to understand the sentiment context of the *** analysis is essential in business and society because it impacts strategic *** analysis involves challenges due to lexical variation,an unlabeled dataset,and text distance *** execution time increases due to the sequential processing of the sequence ***,the calculation times for the Transformer models are reduced because of the parallel *** study uses a hybrid deep learning strategy to combine the strengths of the Transformer and Sequence models while ignoring their *** particular,the proposed model integrates the Decoding-enhanced with Bidirectional Encoder Representations from Transformers(BERT)attention(DeBERTa)and the Gated Recurrent Unit(GRU)for sentiment *** the Decoding-enhanced BERT technique,the words are mapped into a compact,semantic word embedding space,and the Gated Recurrent Unit model can capture the distance contextual semantics *** proposed hybrid model achieves F1-scores of 97%on the Twitter Large Language Model(LLM)dataset,which is much higher than the performance of new techniques.

关键词： DeBERTa GRU Naive Bayes LSTM sentiment analysis large language model

来源：评论

学校读者我要写书评

暂无评论

An Integrated Approach Based on Fuzzy Logic and Machine Learning Techniques for Reliable Wine Quality Prediction 4

An Integrated Approach Based on Fuzzy Logic and Machine Lear...

引用

4th International Conference on Evolutionary Computing and Mobile Sustainable Networks, ICECMSN 2024

作者： Kaur, Narinder Kaur, Gaganjot Aruchamy, Prasanth Chaudhary, Neelu Department of Computer Science & Engineering Chandigarh University Mohali Gharuan India Department of Computer Science & Engineering Raj Kumar Goel Institute of Technology UP Ghaziabad India Department of Computer Science & Engineering Vel Tech Rangarajan Dr. Sagunthala R&D Institute of Science and Technology Chennai India Department of Computer Science & Engineering Manav Rachna University Haryana Faridabad India

In recent decades, wine quality has been one of the predominant problems in many wine industries. However, the analysis of wine quality is inherently complex owing to its multivariate characteristics and the instigation of several sensory features. Most of the existing prediction methods lagged in providing higher detection accuracy for multi-dimensional datasets. To overcome this, a novel Adaptive Wine Quality Prediction (AWQP) approach has been proposed to assess the quality of the wine in an accurate manner. The proposed AWQP methodology entails the development of a Hybrid detection model that encompasses the fuzzy logic principles with the predictive influence of machine learning techniques. Primarily, the sensory features like aroma, taste, and color are delineated by exemplifying the linguistic variables. The fuzzy rules are then determined to collare the qualitative relationships among these different variables. Subsequently, the finest machine learning algorithm can be carried out to train and test the prediction model. The proposed AWQP methodology ameliorates the comprehensibility of the decision-making process through the hybridization of fuzzy logic and the finest machine learning. This proper hybridization facilitates the proposed method to achieve a superior detection accuracy of 98.75% when compared to the existing prediction methods. © 2025 The Authors. Published by Elsevier B.V.

关键词： Adversarial machine learning

来源：评论

学校读者我要写书评

暂无评论

SkyStrokes: Bridging Artistic Expression and Communication Barriers Through Aerial Digital Canvas 6

SkyStrokes: Bridging Artistic Expression and Communication B...

引用

6th IEEE International Conference on Emerging Smart Computing and Informatics, ESCI 2024

作者： Bommala, Harikrishna Suvekha, Raman Reddy, Orra Venkata Srikanth Kumar, Neerati Shiva Rokati, Sai Kiran KG Reddy College of Engineering and Technology Department of Computer Science and Engineering Telanagana Moinabad India KG Reddy College of Engineering and Technology 4th Year Department of Computer Science and Engineering Hyderabad India

ISBN: (纸本)9798350306613

This endeavors to address pressing societal concerns by introducing innovative solutions. It confronts the challenges of communication barriers encountered by the hearing-impaired, the excessive reliance on smartphones, and the environmental impact of paper waste. Through the implementation of 'on-Air writing,' an application is proposed that tracks object movements, enabling users to draw on a computer screen by simply waving their hands. By merging technology and social responsibility, this initiative strives to empower the deaf and mute community while advocating for eco-conscious practices. In doing so, it not only revolutionizes communication but also encourages a more inclusive, tech-savvy, and environmentally aware world. © 2024 IEEE.

关键词： computer Vision Gesture recognition Hand tracking Human-computer Interaction Machine Learning Social Issues

来源：评论

学校读者我要写书评

暂无评论

Automated Depression Detection from Tweets: A Comparison of NLP Techniques 8

Automated Depression Detection from Tweets: A Comparison of ...

引用

8th International Artificial Intelligence and Data Processing Symposium, IDAP 2024

作者： Balci, Emirhan Sarac, Esra Adana Science and Technology University Department of Computer Engineering Adana Turkey

ISBN: (纸本)9798331531492

This paper aims to classify suicidal ideation as a symptom of depression from social media posts by applying the state-of-the-art classification model BERT (Bidirectional Encoder Representations from Transformers) and three traditional machine learning algorithms for binary classification. Since depression is one of the most prevalent mental health disorders amongst psychiatric disorders, the authors intended to present an experimental analysis of the machine learning classifier results as a comparison of novel depression detection techniques. We utilized undiagnosed user posts from Twitter as our dataset and tested the fine-tuned BERT model by applying hold-out and 10-fold cross-validation techniques. Since the dataset is highly unbalanced, Support Vector Machine (SVM), Naive Bayes, and Random Forest algorithms were employed on the same dataset with and without the oversampling method SMOTE (Synthetic Minority Oversampling Technique). The results demonstrate that traditional machine learning classifiers cannot infer sentiment from data containing various linguistic cues, such as depression symptoms. On the other hand, the state-of-the-art model BERT achieves 99.29% and 99.56% macro and micro-F-measure values, respectively, surpassing traditional machine learning algorithms in terms of these metrics. As a robust solution to depression detection from textual data, the BERT model is more trustworthy than the traditional machine learning classifiers to detect specific cues related to depression and similar mental disorders. This study contributes to the relevant research areas of natural language processing by indicating the performance difference between the BERT model and several traditional machine learning algorithms as a generalized approach for classification tasks. © 2024 IEEE.

关键词： Tweets

来源：评论

学校读者我要写书评

暂无评论

RTPD-YOLO: Reconciling YoLo-V8 Model for Real-Time Potholes Detection

RTPD-YOLO: Reconciling YoLo-V8 Model for Real-Time Potholes ...

引用

2024 IEEE International Conference on Machine Learning and Applied Network Technologies, ICMLANT 2024

作者： Balakrishna, Sivadi Solanki, Vijender Kumar Department of Advanced Computer Science & Engineering Vignan’s Foundation for Science Technology & Research Guntur A.P Vadlamudi India Department of Computer Science & Engineering Stanley College of Engineering and Technology for Women Telangana Hyderabad India

ISBN: (纸本)9798350389180

Indian road privilege systems employ low-quality building materials, making pothole detection and avoidance difficult. Preventing accidents requires quickly identifying and fixing potholes. Roadside potholes may damage cars and endanger drivers. This article introduces a new method for realtime pothole identification called RTPD-YOLOv8. It can identify several forms of potholes, including cracks, oil stains, patches, and stones. To begin processing, pothole videos are first transformed into picture frames. The Ultratics are applied to these frames before processing to lessen distortions. The next step is to use the pre-processed photos to detect the pothole and identify its region using the Sobel edge detector and YOLOv8. Thereafter. The proposed RTPD-YOLO will detect the potholes efficiently. The Simulation results show that Accuracy, Precision, Recall, and F1-score were used to measure the performance of the RTPD-YOLO approaches. A total Accuracy of 99.72% is attained by the RTPD-YOLO. Also, the RTPD-YOLO model gets a 98.86% precision, 94.64 percent recall, and 91.35% F1-score. When compared to other models, like Faster RCNN, SSD, and mask R CNN, the RTPD-YOLO network achieves a higher Accuracy range. © 2024 IEEE.

关键词： Edge detection

来源：评论

学校读者我要写书评

暂无评论

Application of Deep Learning in the Development of a Drug Discovery System 15th

Application of Deep Learning in the Development of a Drug Di...

引用

15th International Conference on Soft Computing and Pattern Recognition, SoCPaR 2023 and 14th World Congress on Nature and Biologically Inspired Computing, NaBIC 2023

作者： Praveen Kumar, B.V. Nageswara Rao, A. Tummala, Navya Deepthi Bikku, Thulasi Reddy Madhavi, K. Computer Science and Engineering Usha Rama College of Engineering and Technology Andhra Pradesh Telaprolu India Computer Science and Engineering CMR Institute of Technology Telangana Hyderabad India Computer Science and Engineering Vignan’s Nirula Institute of Technology and Science for Women Andhra Pradesh Guntur India Computer Science and Engineering Amrita School of Computing Amaravati Amrita Vishwa Vidyapeetham Andhra Pradesh Amaravati522503 India School of Computing Mohan Babu University Tirupati India

ISBN: (纸本)9783031810824

Efficient drug discovery hinges on the accurate prediction of interactions between potential drugs and target proteins. Deep learning models have shown promise in this domain;however, their adoption can be challenging for individuals not well-versed in both computer science and cheminformatics. In response to this challenge, we present DeepPurpose, an extensive and user-friendly deep learning library tailored for drug-target interaction (DTI) prediction. DeepPurpose offers a range of features, including the ability to customize interaction types through chemical and protein encoders, as well as access to over 50 neural architectures. As the effectiveness of existing aseptic medications diminishes, the identification of viable DTIs becomes pivotal in the drug development process. DeepPurpose employs an encoder-decoder structure to model interactions, taking compound SMILES strings and pairs of protein amino acid sequences as inputs. The system generates a binding affinity score for the given compound-protein pair. To achieve this, DeepPurpose encodes input proteins and compounds using multiple deep learning encoders, combining their deep embeddings and passing them through a deep neural network decoder to determine binding propensity. DeepPurpose represents a valuable tool in the field of drug discovery, providing researchers with accessible and effective means to predict and explore potential drug-target interactions. Its user-friendly design and comprehensive functionality aim to bridge the gap between deep learning and DTI prediction for a wider audience of researchers and practitioners. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

关键词： Deep neural networks

来源：评论

学校读者我要写书评

暂无评论

Enhancing Investment Forecasting through the Second-Level Thinking Concept

Enhancing Investment Forecasting through the Second-Level Th...

引用

2024 IEEE International Conference on Systems, Man, and Cybernetics, SMC 2024

作者： Zhao, Dan Wang, Chao Shen, Zhenyi The College of Computer Science and Technology Zhejiang University China Swinburne University of Technology The Faculty of Science Engineering and Technology Victoria Australia

ISBN: (纸本)9781665410205

Forecasting models that rely on machine learning often lack a transparent trading logic, which can erode trust when they are applied to real-world trading scenarios. To tackle this problem, a model grounded in an investment concept known as 'second-level thinking' is proposed. This approach enhances both the interpretability of the model and the prediction accuracy. The model's first-level thinking process makes a broad prediction based on features related to the macroeconomic environment. These features are used to group samples into clusters, each representing a different state of the macroeconomic environment. For each cluster, an associative memory is constructed to improve interpretability and simplify the prediction process, mirroring the function of human memory. A theorem is developed to explore the relationship between the capacity of the associative memory and the threshold used in the clustering step. This helps establish an appropriate threshold for reliably memorizing patterns within each cluster. The second-level thinking process then refines these predictions by fitting the residuals from the first-level process using technical features, further enhancing prediction performance. Experiments conducted on two datasets of Chinese government bonds demonstrate that our proposed model outperforms others in terms of prediction accuracy and trading performance. © 2024 IEEE.

关键词： Prediction models

来源：评论

学校读者我要写书评

暂无评论

Customized Creation of ERC 20 Standard Cryptocurrency on the Ethereum Network 1

Customized Creation of ERC 20 Standard Cryptocurrency on the...

引用

1st International Conference on Optimization Techniques for Learning, ICOTL 2023

作者： Rahin Batcha, R. Vignesh, S. Subathra Saravanan, D. Arunkumar, G. Sathyabama Institute of Science and Technology Department of Computer Science and Engineering Chennai India Madanapalle Institute of Technology & Science Department of Computer Science and Engineering Madanapalle India

ISBN: (纸本)9798350328042

Cryptocurrencies like Bitcoin and Ethereum use the blockchain technology, that is known for providing significant security advantages. But there is more to Bitcoin and Ethereum than that. They use digital signatures to verify users. A digital signature uses cryptography extensively. There are two different kinds of keys: private and public. The public keys is the location of a bitcoin wallet that a bitcoin user holds. At this address, anyone desiring to send Bitcoins should do so. Blockchain is a method of storing data that renders it challenging or impossible to alter, hack, or deceive the entire system. In this current endeavour, we develop a digital currency (a crypto tokens) that complies with the ERC 20 norm, can be traded on Ethereum platform, and is able to be sold in a public sale. The Electronic Token Exchange (ERC) 20 standard specifies the characteristics of tokens, including token transfer, total token supply, and other factors. Solidity-built smart contracts will be used to govern and control how the tokens are used. A test-driven process is used to build tokens. The production of test cases prior to the building of smart contracts is referred to as "test-driven"development. © 2023 IEEE.

关键词： Smart contract

来源：评论

学校读者我要写书评

暂无评论

Optimized Routing in FANET Using Marine Predators Algorithm 4

Optimized Routing in FANET Using Marine Predators Algorithm

引用

4th IEEE International Conference on Electrical, computer, and Energy Technologies, ICECET 2024

作者： Minocha, Sachin Singh, Birmohan Kaur, Manpreet Namasudra, Suyel Lorenz, Pascal School of Computer Science Engineering and Technology Bennett University Greater Noida India Sant Longowal Institute of Engineering and Technology Department of Computer Science and Engineering Longowal India Sant Longowal Institute of Engineering and Technology Department of Electrical and Instrumentation Engineering Longowal India National Institute of Technology Agartala Department of Computer Science and Engineering Tripura India University of Haute Alsace Colmar France

ISBN: (纸本)9798350395914

Flying Adhoc Network (FANET) has gained attention due to regressive real-time use of Unmanned Aerial Vehicles (UAVs). Optimized routing in FANETs is more challenging than mobile ad-hoc networks and vehicular ad-hoc networks due to restricted bandwidth and higher mobility of UAVs. Apart from the Mobile Adhoc Network (MANET) routing protocols, nature-inspired techniques-based optimized routing has shown promising routing performance in FANET. This paper proposes a Marine Predators Algorithm (MPA) based optimized routing protocol for FANET. The proposed routing technique selects a better routing path because of the high exploration and exploitation capabilities of MPA. The fitness function based on link stability and energy efficiency helps to select a routing path with better Quality of Service (QoS). The performance of the proposed technique has been analyzed based on packet arrival ratio, end-to-end delay, and throughput, and it is compared to five state-of-the-art techniques by varying the speed and number of the nodes. The simulation results depict that the proposed work is more efficient than the existing state-of-the-art schemes. © 2024 IEEE.

关键词： Vehicular ad hoc 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：