检索结果-内蒙古大学图书馆

Question-Answering Pair Matching Based on Question Classification and Ensemble Sentence Embedding

computer systems science & engineering 2023年第9期46卷 3471-3489页

作者： Jae-Seok Jang Hyuk-Yoon Kwon Department of Computer Science and Engineering Seoul National University of Science and TechnologySeoul01811Korea Department of Industrial Engineering/Graduate School of Data Science Seoul National University of Science and TechnologySeoul01811Korea

Question-answering(QA)models find answers to a given *** necessity of automatically finding answers is increasing because it is very important and challenging from the large-scale QA data *** this paper,we deal with the QA pair matching approach in QA models,which finds the most relevant question and its recommended answer for a given *** studies for the approach performed on the entire dataset or datasets within a category that the question writer manually *** contrast,we aim to automatically find the category to which the question belongs by employing the text classification model and to find the answer corresponding to the question within the *** to the text classification model,we can effectively reduce the search space for finding the answers to a given ***,the proposed model improves the accuracy of the QA matching model and significantly reduces the model inference ***,to improve the performance of finding similar sentences in each category,we present an ensemble embedding model for sentences,improving the performance compared to the individual embedding *** real-world QA data sets,we evaluate the performance of the proposed QA matching *** a result,the accuracy of our final ensemble embedding model based on the text classification model is 81.18%,which outperforms the existing models by 9.81%∼14.16%***,in terms of the model inference speed,our model is faster than the existing models by 2.61∼5.07 times due to the effective reduction of search spaces by the text classification model.

关键词： Question-answering text classification model data augmentation text embedding

来源：评论

学校读者我要写书评

暂无评论

GRAPHTRAIL: Translating GNN Predictions into Human-Interpretable Logical Rules 38

GRAPHTRAIL: Translating GNN Predictions into Human-Interpret...

引用

38th Conference on Neural Information Processing systems, NeurIPS 2024

作者： Armgaan, Burouj Dalmia, Manthan Medya, Sourav Ranu, Sayan Department of Computer Science University of Illinois Chicago United States Department of Computer Science & Engineering Yardi School of AI IIT Delhi India

Instance-level explanation of graph neural networks (GNNs) is a well-studied area. These explainers, however, only explain an instance (e.g., a graph) and fail to uncover the combinatorial reasoning learned by a GNN from the training data towards making its predictions. In this work, we introduce GRAPHTRAIL, the first end-to-end, post-hoc, global GNN explainer that translates the functioning of a black-box GNN model to a boolean formula over the (sub)graph-level concepts without relying on local explainers. GRAPHTRAIL is unique in automatically mining the discriminative subgraph-level concepts using Shapley values. Subsequently, the GNN predictions are mapped to a human-interpretable boolean formula over these concepts through symbolic regression. Extensive experiments across diverse datasets and GNN architectures demonstrate significant improvement over existing global explainers in mapping GNN predictions to faithful logical formulae. The robust and accurate performance of GRAPHTRAIL makes it invaluable for improving GNNs and facilitates adoption in domains with strict transparency requirements. © 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

Bias analysis of AI models for undergraduate student admissions

引用

Neural Computing and Applications 2025年第12期37卷 7785-7795页

作者： Van Busum, Kelly Fang, Shiaofen Department of Computer Science and Software Engineering Butler University IndianapolisIN United States Computing and Engineering Indiana University Luddy School of Informatics Indiana University Indianapolis IndianapolisIN United States

Bias detection and mitigation is an active area of research in machine learning. This work extends previous research done by the authors Van Busum and Fang (Proceedings of the 38th ACM/SIGAPP Symposium on Applied Computing, 2023) to provide a rigorous and more complete analysis of the bias found in AI predictive models. Admissions data spanning six years was used to create an AI model to determine whether a given student would be directly admitted into the school of science under various scenarios at a large urban research university. During this time, submission of standardized test scores as part of a student’s application became optional which led to interesting questions about the impact of standardized test scores on admission decisions. We developed and analyzed AI models to understand which variables are important in admissions decisions, and how the decision to exclude test scores affects the demographics of the students who are admitted. We then evaluated the predictive models to detect and analyze biases these models may carry with respect to three variables chosen to represent sensitive populations: gender, race, and whether a student was the first in his/her family to attend college. We also extended our analysis to show that the biases detected were persistent. Finally, we included several fairness metrics in our analysis and discussed the uses and limitations of these metrics. © The Author(s), under exclusive licence to Springer-Verlag London Ltd., part of Springer Nature 2024.

关键词： Students

来源：评论

学校读者我要写书评

暂无评论

CMCF-Net: An End-to-End Context Multiscale Cross-Fusion Network for Robust Copy-Move Forgery Detection

引用

IEEE Transactions on Multimedia 2024年 26卷 6090-6101页

作者： Xiong, Lizhi Xu, Jianhua Yang, Ching-Nung Zhang, Xinpeng Nanjing University of Information Science and Technology School of Computer Science Nanjing210044 China Jiangsu Collaborative Innovation Center on Atmospheric Environment and Equipment Technology Engineering Research Center of Digital Forensics Ministry of Education Nanjing210044 China Nanjing University of Information Science and Technology Software Engineering Nanjing210044 China National Dong Hwa University Department of Computer Science and Information Engineering Hualien974301 Taiwan Fudan University School of Computer Science Shanghai200433 China

Image copy-move forgery detection (CMFD) has become a challenging problem due to increasingly powerful editing software that makes forged images increasingly realistic. Existing algorithms that directly connect multiple scales of features in the encoder part may not effectively aggregate contextual information, resulting in poor performance. In this paper, an end-to-end context multiscale cross-fusion network (CMCF-Net) is proposed to detect image copy-move forgery. The proposed network consists of a multiscale feature extraction fusion (MSF) module and a multi-information fusion decoding (MFD) module. Multiscale information is efficiently extracted and fused in the MSF module utilizing stacked-scale feature fusion, which improves the network's forgery localization ability on objects of different scales. The MFD module employs contextual information combination and weighted fusion of multiscale information to guide the network in obtaining relevant clues from correlated information at multiple different scales. Experimental results and analysis have demonstrated that the proposed CMCF-Net achieves the best localization results with higher robustness. © 1999-2012 IEEE.

关键词： Feature extraction

来源：评论

学校读者我要写书评

暂无评论

Accurate and Efficient Code Matching Across Android Application Versions Against Obfuscation 31

Accurate and Efficient Code Matching Across Android Applicat...

引用

31st IEEE International Conference on Software Analysis, Evolution and Reengineering, SANER 2024

作者： Feng, Runhan Zhang, Zhuohao Zhou, Yetong Yan, Ziyang Zhang, Yuanyuan Shanghai Jiao Tong University Department of Computer Science and Engineering Shanghai China School of Computer Science and Engineering University of Electronic Science and Technology of China Chengdu China

ISBN: (纸本)9798350330663

In an effort to enhance the attractiveness of apps, developers consistently and frequently release updates to introduce new features and address known issues. Although frequent updates are beneficial for improving user experience, they also increase the workload for reverse engineers since existing analysis results may become obsolete after the release of a new version. Matching code across app versions can help reverse engineers quickly migrate existing analysis results to new versions, verifying whether their prior findings still hold in the new version. This allows them to focus more on the modified portions of the code, thus increasing reverse engineering efficiency. Nevertheless, existing techniques cannot effectively match the code of apps protected by obfuscation techniques, which are pervasively adopted in prac-tice. To address the challenges introduced by code obfuscation, this study presents MatchScope, a novel automated approach designed to match code at the method level across versions of Android app binaries. MatchScope effectively leveraging different levels of fine-grained code features, including class structures and method opcodes, etc., for similarity comparison, thus achieving high accuracy. To further enhance the matching efficiency, we design an index-aware matching algorithm, significantly reducing the scope and number of pairwise comparisons required compared with existing work. The critical insight of our algorithm lies in that the obfuscation tools usually rely on an incrementing index to generate obfuscated names for classes in a deterministic way. Our evaluation on 20 open-source and 60 real-world apps demonstrates the effectiveness of MatchScope. The precision and recall of MatchScope on the ground truth achieve 97.49 % and 92.34 %, respectively, which are 19.50 % and 30.74 % higher than the state-of-the-art tool. © 2024 IEEE.

关键词： Reverse engineering

来源：评论

学校读者我要写书评

暂无评论

Double Auction Strategies for Peer-to-Peer Energy Trading in Smart Grids: A Game Theoretic and Blockchain Approach 6

Double Auction Strategies for Peer-to-Peer Energy Trading in...

引用

6th International Conference on Energy, Power and Environment, ICEPE 2024

作者： Naik, Madhusudan Pradhan, Nihar Ranjan Pratap Singh, Akhilendra National Institute of Technology Meghalaya Department of Computer Science and Engineering Shillong India VIT-AP University School of Computer Science and Engineering Amaravati India

ISBN: (纸本)9798350379297

The concept of using blockchain technology for peer-to-peer energy trading (P2PET) in smart grids, particularly employing a double auction-based game theoretic method, is a fascinating application of both blockchain and economic principles. This paper deals with a blockchain-enabled framework for P2PET in a smart grid by using a double auction-based game theory approach. The proposed permissioned blockchain network was developed using Hyperledger Fabric, Minifab and designed by configuring the *** file. Three smart contracts, or Chaincode namely Double Auction Order, CreateP2PEnergy, and Matching, have been written using Golang and deployed into the blockchain network, thereby creating automatic energy trading transactions between energy producers and consumers. The client communication, such as consumer queries, user input, and producer trading concepts, was designed using the Postman API. Additionally, the double auction framework's performance has been measured by Hyperledger Caliper. Analysis shows that when 1000 transactions were initiated in each round by clients using Hyperledger Caliper and REST API services, the Proof of Elapsed Time (PoET) mechanism successfully processed 946 transactions, in contrast to the 627 transactions managed by Proof of Work (PoW). The outcomes from multiple rounds of testing underscore the enhanced performance of this framework. In summary, the combination of blockchain technology, P2P energy trading, smart grids, and game theory offers a promising approach to revolutionizing the energy sector. By decentralizing energy trading, empowering consumers, and leveraging market mechanisms, this approach has the potential to enhance efficiency, promote renewable energy adoption, and foster a more sustainable energy ecosystem. © 2024 IEEE.

关键词： Blockchain

来源：评论

学校读者我要写书评

暂无评论

Enhanced Video Streaming Based Computing Encoder for Digital Visual Processing 2

Enhanced Video Streaming Based Computing Encoder for Digital...

引用

2nd IEEE International Conference on Distributed Computing and Electrical Circuits and Electronics, ICDCECE 2023

作者： Kumar, Tarun Shukla, Amita Galgotias University Department of Computer Science and Engineering Greater Noida India Noida Institute of Engineering and Technology Department of Computer Science and Business System Greater Noida India

ISBN: (纸本)9798350347456

The rise of video streaming for digital visual processing has been a boon for the industry of visual processing. Video streaming technology has made it easier for companies to capture, analyze, and interpret visual data faster than ever before. It has allowed for the storage and transmission of large amounts of visual data at high speeds, providing businesses with the ability to process and interpret this data in real-time. Video streaming technology can be used in a wide variety of applications, including facial recognition, 3D mapping, and object recognition. By streaming video data, companies can quickly and accurately identify individuals, recognize objects, and track movement. This technology can also be used in security applications, such as surveillance and monitoring, as well as in medical imaging, such as MRI and CT scans. Video streaming technology has also allowed companies to create more efficient visual processing systems. The streaming video data can be used to automate the process of image recognition and object classification. This has allowed companies to reduce the amount of time and effort needed to interpret visual data. Additionally, streaming video data can be used to create virtual reality experiences, providing users with an immersive experience when viewing digital images. © 2023 IEEE.

关键词： Video streaming

来源：评论

学校读者我要写书评

暂无评论

Personalized Shatavari Nutrition Recommendations for Menopausal Women using Machine Learning Algorithms 5

Personalized Shatavari Nutrition Recommendations for Menopau...

引用

5th International Conference on Smart Electronics and Communication, ICOSEC 2024

作者： Logapriya, E. Surendran, R. Soniya, M. Sathish Kumar, P.J. Saveetha School of Engineering Saveetha Institute of Medical and Technical Sciences Department of Computer Science and Engineering Chennai India R M.K Engineering College Department of Computer Science and Engineering Tamil Nadu Kavaraipettai India Panimalar Engineering College Department of Computer Science and Engineering Chennai India

ISBN: (纸本)9798331504403

This research provides a comprehensive approach in designing a recommendation system for women's health issues, with emphasis on menopause and optimal eating plan with Shatavari nutrition plan incorporated. The data includes symptom severity;heart rate;temperature;sleep;and food and water intake, which has been obtained from patients' medical records, structured surveys, the users themselves, and through healthcare apps. Handling of missing values involves the completion of the missing records by forward method and imputation. Kindly find below the steps taken to pre-process the given data set for its numerical and categorical features respectively;Standard Scaler for numerical data set and OneHotEncoder for the binary data set. Variable creation is used to improve the performances of the models by creating new variables from other variables. Raw, selected and vectorized features are used to perform and compare multiple machine learning algorithms such as linear regression, logistic regression, support vector machines, decision random forest and gradient boosting, XGBoost, LightGBM, and CatBoost. The present study reveals that LightGBM has the maximum accuracy of 96%. The proposed system combines collaborative and content-based filtering, which will provide accurate and individual health and lifestyle recommendations for enhancing the quality of women's life during menopausal transition. © 2024 IEEE.

关键词： Support vector regression

来源：评论

学校读者我要写书评

暂无评论

An Overview of Artificial Intelligence Diagnostic systems for Hepatitis B Virus 4

An Overview of Artificial Intelligence Diagnostic Systems fo...

引用

4th International Conference on Technological Advancements in Computational sciences, ICTACS 2024

作者： Singh, Dalwinder Kaur, Harjeet School of Computer Science and Engineering Lovely Professional University Punjab India Lovely Professional University Department of Computer Science and Engineering Punjab Phagwara India

ISBN: (纸本)9798350387490

A virus that can attack a human's liver and cause serious liver damage is primarily known as Hepatitis B Virus or HVB. It influences nearly all functionalities performed by the liver. It is basically a short term infection;however, it will become a chronic disorder and remains for a long term in some individuals. The severe effect of this illness is that it can cause chronic liver diseases, and with time it results in cancer of the liver. Hence, it puts a great risk to the life of a patient. The early detection of the virus is necessary, which assists in saving the patient's life and brings more treatment procedures that can adapt by the doctor for treatment. However, this life-threatening disease's therapy and diagnosis are both costly and can have dangerous adverse effects. As a result, developing a diagnostic system that lowers the cost of therapy and diagnosis is critical. In the medical field, several artificial intelligence technologies have been employed to assist doctors and experts in the diagnosing process. The significant aim of this paper is to thoroughly review all these diagnostic systems developed or proposed by various researchers. Later on, the systems are compared with one another to evaluate which system or model is best to use in solving the problem. The parameter that is taken into account to complete the comparison. As a result, it is noticed that the system developed by using a genetic neural network is accurate for the diagnosis of HBV and has an accuracy of 99.14%. © 2024 IEEE.

关键词： Artificial Intelligence Comparative Study Diagnostic systems Early Detection Hepatitis B Virus Intelligent systems Medical Expert systems

来源：评论

学校读者我要写书评

暂无评论

Algorithmic Innovations in Machine Learning for Drug Discovery and Pharmaceutical Innovations 2

Algorithmic Innovations in Machine Learning for Drug Discove...

引用

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

作者： Sangeetha, M. Harin Fernandez, F. Mary Ganesh Ramachandran, A. Saravanan, S.K. Bhanumathi, M. University of Technology and Applied Sciences Department of Computer Engineering Nizwa Oman Saveetha Institute of Medical and Technical Sciences Saveetha University Saveetha School of Engineering Department of Computer Science and Engineering Chennai India Saveetha Institute of Medical and Technical Sciences Saveetha University Saveetha School of Engineering Department of Computer Science Chennai India Saveetha School of Engineering Simats BigData and Network Security Department of Computers Science and Engineering Tamil Nadu Chennai India Department of Computer Science and Engineering Chennai India

ISBN: (纸本)9798350384277

Algorithmic advancements in machine learning have transformed drug discovery and pharmaceutical innovation. This study examines how machine learning might speed medication development and pharmaceutical innovation. Data from genomes, chemistry, and clinical trials is collected to start the research. Careful data cleansing, normalization, and feature engineering prepare the data for model building. Random Forest, SVM, NN, Decision Tree, and Logistic Regression are tested for prediction. ROC AUC results show that Random Forest is the most accurate model at 86%. Ensemble approaches have drug discovery potential. Beyond accuracy, models are assessed for precision, false positive, and false negative rates. Drug discovery relies on precision to measure positive predictions. The Random Forest model makes the most accurate positive predictions at 0.87. However, false positives and negatives also matter in drug discovery. The Decision Tree model contains the falsest positives (35), underlining the need to reduce them to save resources. The Neural Network approach reduces false negatives to 12, ensuring good medication candidates are not overlooked. The results show that different machine learning models have trade-offs, underscoring the necessity for a personalized strategy depending on objectives and restrictions. Algorithmic breakthroughs in pharmaceutical innovation reduce medication development time by 40% and cost by 25%, according to the study. © 2024 IEEE.

关键词： Drug discovery

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：