检索结果-内蒙古大学图书馆

International Conference on Cybernetics and Intelligent System (ICORIS)

作者： Karli Eka Setiawan Gregorius Natanael Elwirehardja Bens Pardamean Computer Science Department BINUS Graduate Program Master of Computer Science Program Bina Nusantara University Jakarta Indonesia Bioinformatics and Data Science Research Center Bina Nusantara University Jakarta Indonesia

ISBN: (纸本)9781665453967

Precise predictions of indoor climate conditions are required in the implementation of Smart Solar Dryer Dome (SDD). Trend development of prediction models is discussed in this review from 15 selected research papers (2018–2022) on indoor climate prediction which was obtained from research paper databases The output shows that the most used model for predicting indoor climate is Artificial Neural Network (ANN), especially Recurrent Neural Network (RNN) such as LSTM and GRU. However, there are some potential methods such as Transformer, Combined Support Vector Machines (SVM)-Deep Learning, and sequence-to-sequence which could outperform other commonly used models. Based on findings various opportunities exist to improve the precision of indoor climate prediction, which can bring power consumption efficiency and others benefit to Smart SDD users. Such studies may further be explored to produce more accurate machine learning models.

关键词： Support vector machines Systematics Recurrent neural networks Power demand Bibliographies Artificial neural networks Predictive models

来源：评论

学校读者我要写书评

暂无评论

LSHBLOOM: MEMORY-EFFICIENT, EXTREME-SCALE DOCUMENT DEDUPLICATION

arXiv

引用

arXiv 2024年

作者： Khan, Arham Underwood, Robert Siebenschuh, Carlo Babuji, Yadu Ajith, Aswathy Hippe, Kyle Gokdemir, Ozan Brace, Alexander Chard, Kyle Foster, Ian Department of Computer Science University of Chicago ChicagoIL United States Data Science & Learning Division Argonne National Laboratory LemontIL United States

Deduplication is a major focus for assembling and curating training datasets for large language models (LLM) – detecting and eliminating additional instances of the same content – in large collections of technical documents. Unrestrained, duplicates in the training dataset increase training costs and lead to undesirable properties such as memorization in trained models or cheating on evaluation. Contemporary approaches to document-level deduplication are often extremely expensive in both runtime and memory. We propose LSHBloom, an extension to MinhashLSH, which replaces the expensive LSHIndex with lightweight Bloom filters. LSHBloom demonstrates the same deduplication performance as MinhashLSH with only a marginal increase in false positives (as low as 1e-5 in our experiments);demonstrates competitive runtime (270% faster than MinhashLSH on peS2o);and, crucially, uses just 0.6% of the disk space required by MinhashLSH to deduplicate peS2o. We demonstrate that this space advantage scales with increased dataset size—at the extreme scale of several billion documents, LSHBloom promises a 250% speedup and a 54× space advantage over traditional MinHashLSH scaling deduplication of text datasets to many billions of documents. Copyright © 2024, The Authors. All rights reserved.

关键词： Large datasets

来源：评论

学校读者我要写书评

暂无评论

Multi-Objective Reinforcement Learning for the Control of Storm-Water Systems Under Distributional Shift

Multi-Objective Reinforcement Learning for the Control of St...

引用

Computational Intelligence for Energy, Transport and Environmental Sustainability (CIETES Companion), IEEE Symposia on

作者： Daisy Welham Sara Sharifzadeh Liam Butler Chedly Tizaoui Department of Computer Science Swansea University Swansea UK Data Science Team Dŵr Cymru Welsh Water Cardiff Wales Department of Chemical Engineering Swansea University Swansea UK

ISBN: (数字)9798331519681

ISBN: (纸本)9798331519698

This paper establishes two reinforcement learning agents, Q-learning and actor-critic, for the control of a simulated storm-water tank. Their performances are compared to existing programmable logic controller (PLC) systems, and a multi-objective optimisation problem is formed for minimising the pumping costs subject to safety constraints. We demonstrate that both Q-learning and actor-critic agents are successful in saving pumping costs in a range of dynamic conditions. In the case of overflow prevention, when there is a distributional shift towards high-rain conditions, some failure cases have been observed affecting capability robustness. We then implement an ensemble method which uses majority-voting of the PLC, actorcritic, and Q-learning agents, and demonstrate that the ensemble method is capable of both reducing pumping costs and reducing overflow under high-rain conditions. We demonstrate that the ensemble method is able to reduce pumping costs by about 30% compared to the PLC controller under normal conditions, while also reducing overflow rates from, 2.1% to 0.27% in high-rain conditions compared to the use of actor-critic alone.

关键词： Q-learning Costs Rain Prevention and mitigation programmable logic devices Green products Robustness Safety Ensemble learning Optimization

来源：评论

学校读者我要写书评

暂无评论

Mining Latent Topical Key Phrase from Content to Context in Unrestricted Healthcare data

Mining Latent Topical Key Phrase from Content to Context in ...

引用

International Conference on computer Supported Cooperative Work in Design

作者： Tie Hua Zhou Tian Yu Jin Xi Wei Wang Ling Wang Keun Ho Ryu School of Computer Science Northeast Electric Power University Jilin China Department of Computer Science Chungbuk National University Cheongju Korea Data Science Laboratory Ton Duc Thang University Ho Chi Minh City Vietnam

ISBN: (数字)9798350349184

ISBN: (纸本)9798350349191

Healthcare data, collected from multiple sources and representing various perspectives, has become an increasingly challenging and important research topic. It permeates many aspects of evidence-based clinical practice, the healing process, and health policy formulation. However, the characteristics of redundancy, diversity, volume, inconsistency, and incompleteness make it difficult to capture valuable semantic properties and linguistic relationships. In this paper, we propose a CNN-based Bidirectional Extension of Phrase Boundary (BEPB) approach to reduce over-reliance on term frequency and mine multiple latent topical key phrases, aiming to improve the quality of key phrases in downstream Natural Language Processing (NLP) applications. The experiments indicate that the BEPB trained from unstructured user-contributed content on health social media sites has successfully mined more relevant topical phrases, particularly in the areas of clinical symptoms and co-occurrence patterns. To summarize, these findings serve as a key contributor to the advancements in Evidence-Based Medicine (EBM), paving the way for improvements in the prevention, diagnosis, treatment, and nursing of diseases.

关键词： Social networking (online) Prevention and mitigation Semantics Redundancy Medical services Named entity recognition Personnel

来源：评论

学校读者我要写书评

暂无评论

Real-time road occupancy and traffic measurements using unmanned aerial vehicle and fundamental traffic flow diagrams

引用

Personal and Ubiquitous Computing 2023年第5期27卷 1669-1680页

作者： Ahmed, Afzal Outay, Fatima Farooq, Muhammad Umer Saeed, Sehrish Adnan, Muhammad Ismail, Muhammad Ali Qadir, Adnan Department of Urban & Infrastructure Engineering NED University of Engineering & Technology Karachi Pakistan Zayed University Dubai United Arab Emirates Department of Computer Science and Information Technology NED University of Engineering & Technology Karachi Pakistan Exascale Open Data Analytics Lab NED University of Engineering & Technology Karachi Pakistan Hasselt University Diepenbeek Belgium

The technologies and sensors developed for standard traffic streams often fail to accurately measure the heterogeneous traffic with no lane discipline. This research proposes an efficient framework to measure traffic flow parameters in real-time using an unmanned aerial vehicle (UAV). The proposed framework in this research estimates road area occupancy from the images extracted from the video recorded using a UAV by applying the image segmentation technique. The road area occupancy is converted into traffic density using an occupancy-density model developed for the local heterogeneous traffic. A speed-density fundamental diagram is developed, which estimates the average speed corresponding to the measured traffic density. The fundamental relation between flow rate, average speed, and density is applied to compute traffic flow rate using the estimated average speed and density. The proposed framework is computationally less demanding in comparison with other computer vision and artificial intelligence–based techniques. The proposed framework accurately measured road area occupancy with an average accuracy of 97.6%. Traffic density was measured with an average accuracy of 88.8%. Similarly, the average speed and flow rate were measured with an average accuracy of 85.4% and 84.2%, respectively. © 2023, The Author(s), under exclusive licence to Springer-Verlag London Ltd., part of Springer Nature.

关键词： Image segmentation

来源：评论

学校读者我要写书评

暂无评论

Impact of Application Choice on iTracker Model Performance Evaluation: A Perspective 2

Impact of Application Choice on iTracker Model Performance E...

引用

2nd International Engineering Conference on Electrical, Energy, and Artificial Intelligence, EICEEAI 2023

作者： Moghlebey, Adam Bou Haydar, Hady Moubayed, Abdallah Injadat, MohammadNoor School of Computing and Augmented Intelligence Arizona State University GilbertAZ United States Lebanese American University Electrical and Computer Engineering Department Byblos Lebanon Zarqa University Data Science and Artificial Intelligence Department Zarqa Jordan

ISBN: (纸本)9798350373363

Eye tracking is underpinned by the eye-mind hypothesis, which posits that individuals tend to direct their gaze toward the information they are currently cognitively engaged with. This aspect has garnered significant attention within deep learning research due to its versatile applications and research. The utilization of Artificial Intelligence (AI) in conjunction with webcams for tracking eye movements has gained increasing popularity in recent years and is anticipated to continue flourishing. This literature review aims to assess the current state and depth of the research in this domain and its various applications. We examine the different methods that employ eye tracking across a diverse range of applications, and by re-exploring existing research, we are able to filter out any overlooked potential or setbacks associated with the use of AI-webcam based systems, with the ultimate goal of maximizing their potential contributions to humanity. In our study, we analyze and evaluate a pioneering eye tracking algorithm, iTracker trained on a crowd-sourced data-set, GazeCapture within the context of the applications it may be used for. While we acknowledge inherent limitations, this widely adopted model forms a robust foundation for future refinements. Furthermore, our investigation encompasses diverse applications with varying accuracy requirements, making it invaluable for individuals seeking to integrate eye tracking into their projects. © 2023 IEEE.

关键词： Eye tracking

来源：评论

学校读者我要写书评

暂无评论

Semantic grounding of LLMs using knowledge graphs for query reformulation in medical information retrieval

Semantic grounding of LLMs using knowledge graphs for query ...

引用

IEEE International Conference on Big data

作者： Antonela Tommasel Ira Assent ISISTAN CONICET-UNCPBA Argentina Department of Computer Science Aarhus University DIGIT Aarhus University Centre for Digitalisation Big Data and Data Analytics Denmark

ISBN: (数字)9798350362480

ISBN: (纸本)9798350362497

The widespread adoption of electronic health records has generated a vast amount of patient-related data, mostly presented in the form of unstructured text, which could be used for document retrieval. However, querying these texts in full could present challenges due to their unstructured and lengthy nature, as they may contain noise or irrelevant terms that can interfere with the retrieval process. Recently, large language models (LLMs) have revolutionized natural language processing tasks. However, despite their promising capabilities, their use in the medical domain has raised concerns due to their lack of understanding, hallucinations, and reliance on outdated knowledge. To address these concerns, we evaluate a Retrieval Augmented Generation (RAG) approach that integrates medical knowledge graphs with LLMs to support query refinement in medical document retrieval tasks. Our initial findings from experiments using two benchmark TREC datasets demonstrate that knowledge graphs can effectively ground LLMs in the medical domain.

关键词： Grounding Law Retrieval augmented generation Semantics Pipelines Noise Knowledge graphs Benchmark testing Natural language processing Resource management

来源：评论

学校读者我要写书评

暂无评论

Paid with Models: Optimal Contract Design for Collaborative Machine Learning

arXiv

引用

arXiv 2024年

作者： Wang, Bingchen Wu, Zhaoxuan Liu, Fusheng Low, Bryan Kian Hsiang Institute of Data Science National University of Singapore Singapore Singapore-MIT Alliance for Research and Technology Singapore Department of Computer Science National University of Singapore Singapore

Collaborative machine learning (CML) provides a promising paradigm for democratizing advanced technologies by enabling cost-sharing among participants. However, the potential for rent-seeking behaviors among parties can undermine such collaborations. Contract theory presents a viable solution by rewarding participants with models of varying accuracy based on their contributions. However, unlike monetary compensation, using models as rewards introduces unique challenges, particularly due to the stochastic nature of these rewards when contribution costs are privately held information. This paper formalizes the optimal contracting problem within CML and proposes a transformation that simplifies the non-convex optimization problem into one that can be solved through convex optimization algorithms. We conduct a detailed analysis of the properties that an optimal contract must satisfy when models serve as the rewards, and we explore the potential benefits and welfare implications of these contract-driven CML schemes through numerical experiments. © 2024, CC BY.

关键词： Collaborative learning

来源：评论

学校读者我要写书评

暂无评论

Collaborative Filtering of Malicious Information from the MULTimedia data Using s

Collaborative Filtering of Malicious Information from the MU...

引用

computer science and Emerging Technologies (CSET), International Conference on

作者： Manikandan Parasuraman Abirami G Arockia Raj A Batri Krishnan Ashok Kumar Munnangi M Kumaresan School of Computer Science and Engineering Jain Deemed to be University Bengaluru India Department of Computer Science Sathyabama Institute of science and technology Chennai India Department of Computer Science and Engineering - Data Science Madanapalle Institute of Technology and Science Madanapalle India Velagapudi Ramakrishna Siddhartha Engineering College Vijayawada India

Due to high proliferation of unsolicited information in the text, image and video files, the social media analytics engine suffers from losing its user, user privacy and others. The unsolicited information is been a huge threat to the service provides as it affects the reputations of the services provided and hence it is necessary to develop an intelligent model. Modern machine learning techniques are applied to identify and filter unsolicited information. However, some of the conventional methods are designed in an insecure way while processing all the text file, image and video files.

关键词：

来源：评论

学校读者我要写书评

暂无评论

HiGPT: Heterogeneous Graph Language Model

arXiv

引用

arXiv 2024年

作者： Tang, Jiabin Yang, Yuhao Wei, Wei Shi, Lei Xia, Long Yin, Dawei Huang, Chao Musketeers Foundation Institute of Data Science Hong Kong Department of Computer Science City University of Hong Kong Hong Kong Baidu Inc. China

Heterogeneous graph learning aims to capture complex relationships and diverse relational semantics among entities in a heterogeneous graph to obtain meaningful representations for nodes and edges. Recent advancements in heterogeneous graph neural networks (HGNNs) have achieved state-of-the-art performance by considering relation heterogeneity and using specialized message functions and aggregation rules. However, existing frameworks for heterogeneous graph learning have limitations in generalizing across diverse heterogeneous graph datasets. Most of these frameworks follow the "pre-train" and "fine-tune" paradigm on the same dataset, which restricts their capacity to adapt to new and unseen data. This raises the question: "Can we generalize heterogeneous graph models to be well-adapted to diverse downstream learning tasks with distribution shifts in both node token sets and relation type heterogeneity?" To tackle those challenges, we propose HiGPT, a general large graph model with Heterogeneous graph instructiontuning paradigm. Our framework enables learning from arbitrary heterogeneous graphs without the need for any fine-tuning process from downstream datasets. To handle distribution shifts in heterogeneity, we introduce an in-context heterogeneous graph tokenizer that captures semantic relationships in different heterogeneous graphs, facilitating model adaptation. We incorporate a large corpus of heterogeneity-aware graph instructions into our HiGPT, enabling the model to effectively comprehend complex relation heterogeneity and distinguish between various types of graph tokens. Furthermore, we introduce the Mixture-of-Thought (MoT) instruction augmentation paradigm to mitigate data scarcity by generating diverse and informative instructions. Through comprehensive evaluations conducted in various settings, our proposed framework demonstrates exceptional performance in terms of generalization performance, surpassing current leading benchmarks. Copyright © 2024, T

关键词： Graph neural 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：