检索结果-内蒙古大学图书馆

Model-Based Clustering of Inhomogeneous Paired Comparison data

1st international workshop on Similarity-Based pattern recognition (SIMBAD)

作者： Busse, Ludwig M. Buhmann, Joachim M. ETH Dept Comp Sci CH-8092 Zurich Switzerland

ISBN: (纸本)9783642244704;9783642244711

This paper demonstrates the derivation of a clustering model for paired comparison data. Similarities for non-Euclidean, ordinal data are handled in the model such that it is capable of performing an integrated analysis on real-world data with different patterns of missings. Rank-based pairwise comparison matrices with missing entries can be described and compared by means of a probabilistic mixture model defined on the symmetric group. Our EM-method offers two advantages compared to models for pairwise comparison rank data available in the literature: (i) it identifies groups in the pairwise choices based on similarity (ii) it provides the ability to analyze a data set of heterogeneous character w.r.t. to the structural properties of individal data samples. Furthermore, we devise an active learning strategy for selecting paired comparisons that are highly informative to extract the underlying ranking of the objects. The model can be employed to predict pairwise choice probabilities for individuals and, therefore, it can be used for preference modeling.

关键词： Artificial intelligence

来源：评论

学校读者我要写书评

暂无评论

Enhancing Unsupervised Few-Shot Medical Image Classification with Weight-Enhanced Contrastive learning 24

Enhancing Unsupervised Few-Shot Medical Image Classification...

引用

1st international Conference on Image Processing machine learning and pattern recognition

作者： Liu, Huantong Zhong, Jingze Shanghai Univ Shanghai Peoples R China

ISBN: (纸本)9798400707032

In the realm of contemporary deep learning, the pivotal challenges confronting few-shot and unsupervised learning reside in harnessing scarce labeled samples alongside abundant unlabeled ones, particularly in the context of medical image classification. This paper introduces an innovative approach that seamlessly integrates dynamic clustering with weights augmentation, aimed at bolstering the performance of few-shot medical image classification. Dubbed Weight-Enhanced Contrastive learning (WECL), our method ingeniously fuses contrastive representation learning with a dynamic memory module during unsupervised pre-training. This fusion facilitates efficient clustering and classification of diverse augmented renditions of the same image. Additionally, the weights augmentation tactic meticulously tunes the weights of both ResNet and teacher-student model branches, thereby mitigating sample bias and enhancing the pre-trained model's proficiency. Extensive experiments across multiple few-shot medical image classification datasets underscore the superiority of our WECL approach, outperforming current state-of-the-art baselines, and effectively addressing issues pertaining to data distribution disparities and sample scarcity.

关键词： Medical Image Classification Weights Augmentation Contrastive learning Unsupervised learning Few-Shot learning

来源：评论

学校读者我要写书评

暂无评论

2014 IEEE 1st international workshop on Artificial Intelligence for Requirements Engineering, AIRE 2014 - Proceedings

2014 IEEE 1st International Workshop on Artificial Intellige...

引用

2014 IEEE 1st international workshop on Artificial Intelligence for Requirements Engineering, AIRE 2014 - Proceedings

ISBN: (纸本)9781479963553

The proceedings contain 9 papers. The topics discussed include: pragmatic ambiguity detection in natural language requirements;weka meets TraceLab: toward convenient classification: machine learning for requirements engineering problems: a position paper;transferring research into the real world: how to improve re with AI in the automotive industry;customizable rule-based verification of requirements ontology;content-based recommendation techniques for requirements engineering;on requirements representation and reasoning using answer set programming;a case study of applying data mining to sensor data for contextual requirements analysis;using AI to model quality attribute tradeoffs;and applying knowledge representation and reasoning to (simple) goal models.

关键词：

来源：评论

学校读者我要写书评

暂无评论

A Survey on Decentralization and Virtualization of Medical Trials: An approach through Ensemble learning models and Convolutional Neural Networks 1

A Survey on Decentralization and Virtualization of Medical T...

引用

1st international Conference on Computer Science and Emerging Technologies, CSET 2023

作者： Chakravarty, Diksha Thomas, Amy Mariam Vivek, V. Bangalore India

ISBN: (纸本)9798350341737

- In sectors such as medicine where accuracy and time are the most prioritized factors the use of technology and its tools have led to huge advancements and contributed to saving millions of lives. Technology has been bifurcated into various categories to suit multiple requirements of the medical sector. The use of machine learning along with well-designed models and sister technologies such as Computer Vision has opened up a whole new dimension in medical sciences. machine learning deals with the building of algorithms that can predict using fed patterns which further specializes in Deep learning. Deep learning's success in a variety of pattern recognition applications has sparked excitement and raised expectations that deep learning, or Artificial Intelligence (AI), could revolutionize the health care that we know today. Right from diagnosis to treatment it can revolutionize the way we construct and operate the medical sector. Over the last decade, Convolutional Neural Networks have achieved breakthroughs in a range of pattern recognition domains. CNN's are generally employed to solve complex image-driven pattern recognition tasks. They have proven to be a great tool in image recognition and classification and are tuned according to the needs of the user. Ensemble learning, as a research area, strives to bring data fusion, data modeling, and data mining together in a single framework. Ensemble learning combines the informative knowledge gained from the previous outcomes to produce knowledge, discovery, and improved predictive performance using adaptive voting methods. In this study, the methodologies of numerous models of image recognition and classification, as well as ensemble learning from multiple authors, are evaluated. In addition, this study shall discuss any feasible enhancements that could be implemented to the existing models as well as how the ideologies of the models could be used for efficient diagnosis in the field of medicine. © 2023 IEEE.

关键词： learning systems

来源：评论

学校读者我要写书评

暂无评论

Acquisition of Typical Occupancy Schedules for Commercial Buildings from Social Networks 1

Acquisition of Typical Occupancy Schedules for Commercial Bu...

引用

1st ACM international workshop on Urban Building Energy Sensing, Controls, Big data Analysis, And Visualization (UrbSys)

作者： Lu, Xing Feng, Fan O'Neill, Zheng Univ Alabama Dept Mech Engn Tuscaloosa AL 35487 USA

ISBN: (纸本)9781450370141

Occupancy behavior in buildings has been a hotspot of research as building systems become more sophisticated. Traditional sensing technologies suffer from high costs, inevitable sensor errors, and scalability issues and thus are not widely implemented in buildings. In this notes paper, two different approaches for extracting the typical occupancy schedules for the input to the building energy simulation are explored based on the data from social networks. The first approach is to use text classification algorithms to identify whether people are present in space where they are making posts on social networks. To achieve this, word embedding and machine learning algorithms for the classification are used. On top of that, we could extract the typical occupancy schedules by assuming certain people counting rules. The second approach is to utilize the processed GPS location tracking data provided by social network giants such as Facebook and Google Map. Web scraping techniques are used to obtain the data and extract the building typical occupancy schedules. Two preliminary case studies demonstrate these two approaches as a proof of concept using a museum building, the Art Institute of Chicago. The results show that the extracted building occupancy schedules from different social network data sources (Twitter, Facebook, and Google Map) share a similar trend but slightly distinct with each other, which requires more explorations and further validations and corrections. However, the methodology and promising results from this preliminary study will lay the foundation of the occupancy sensing through the social network data mining, which aims to provide another data source for occupancy sensing in buildings at the building level. This will provide another alternative to estimate the occupancy at the community- and urban-scale.

关键词： Building Occupancy data Acquisition machine learning machine learning

来源：评论

学校读者我要写书评

暂无评论

NEURAL NETWORKS - THEIR APPLICATIONS AND PERSPECTIVES IN INTELLIGENT MACHINING

引用

COMPUTERS IN INDUstRY 1991年第2-3期17卷 101-119页

作者： BARSCHDORFF, D MONOstORI, L UNIV GESAMTHSCH PADERBORN W-4790 PADERBORNGERMANY

In intelligent manufacturing systems, unprecedented and unforeseen situations are expected to be solved, within certain limits, even on the basis of incomplete and imprecise information. One has tried to achieve this goal since years. Gradually, it seems to be realizable through partial solutions, integrated in today's flexible manufacturing systems. These complexes are fairly complicated material and data processing systems, in which different sensors and actuators are thoroughly distributed. The most important requirements for the intelligent techniques to be applied in these systems are the abilities for integration of multiple sensor information, for real-time functioning, for effective knowledge representation and for learning or adaptivity. Artificial neural networks seem to be able to fulfil some of these requirements and to contribute to the implementation of partial solutions, which can lead to the realization of truly intelligent manufacturing systems. The paper gives a summary of known neural networks applications and perspectives in intelligent manufacturing. Special emphasis is given on intelligent machining, namely on the following fields: multisensor fusion and integration, learning of process models, adaptive control, monitoring, diagnostics and quality control. For the sake of perspicuity a short survey of different artificial neural network structures and learning algorithms is also given, together with frequent applications of neural network techniques in fields different from intelligent manufacturing.

关键词： CONTROL MONITORING DIAGNOstICS machine TOOLS MANUFACTURING PROCESSES INTELLIGENT MACHINING DIGITAL SIGNAL PROCESSING pattern recognition NEURAL NETWORKS EXPERT SYstEMS

来源：评论

学校读者我要写书评

暂无评论

Evaluating the Max-Min Hill-Climbing Estimation of Distribution Algorithm on B-Functions 6th

Evaluating the Max-Min Hill-Climbing Estimation of Distribut...

引用

6th international workshop on Artificial Intelligence and pattern recognition (IWAIPR)

作者： Madera, Julio Ochoa, Alberto Univ Camaguey Dept Comp Sci Camaguey Cuba Inst Cibernet Matemat & Fis Havana Cuba

ISBN: (纸本)9783030011321;9783030011314

In this paper we evaluate a new Estimation of Distribution Algorithm (EDA) constructed on top of a very successful Bayesian network learning procedure, Max-Min Hill-Climbing (MMHC). The aim of this paper is to check whether the excellent properties reported for this algorithm in machine learning papers, have some impact on the efficiency and efficacy of EDA based optimization. Our experiments show that the proposed algorithm outperform wellknown state of the art EDA like BOA and EBNA in a test bed based on B-functions. On the basis of these results we conclude that the proposed scheme is a promising candidate for challenging real-world applications, specifically, problems related to the areas of data mining, Patter recognition and Artificial Intelligence.

关键词： Estimation of distribution algorithms B-functions Bayesian networks Dependency learning Evolutionary optimization

来源：评论

学校读者我要写书评

暂无评论

Research on a Software System Size Evaluation Model Combining BiLstM-CRF with Domain Adaptation learning 24

Research on a Software System Size Evaluation Model Combinin...

引用

1st international Conference on Image Processing machine learning and pattern recognition

作者： Sun, Sen Tao, Yizheng China Acad Engn Phys Inst Comp Applicat Beijing Peoples R China

ISBN: (纸本)9798400707032

In the field of software engineering, accurate and efficient measurement of system size is crucial for project management and decision-making. Traditional software system size evaluation methods, such as expert decision-making methods, COCOMO models, function point methods, and analogical methods, have problems such as relying on manual labor, low efficiency, and susceptibility to subjective factors. To address these issues, this paper proposes a Dalse (Domain Adaptive learning System Evaluation) automated method for evaluating software function size. Based on the IFPUG function point standard, this method constructs a BiLstM-CRF model for identifying function types, and introduces domain adaptive learning technology to solve the problem of insufficient labeled text data for domain-specific annotation. This enables the method to flexibly adapt to the characteristics of software in specific domains, achieving automated extraction and quantitative calculation of software function points. Experimental validation shows that the method's precision, recall, and F1 score reached 83%, 80%, and 82%, respectively, which is a 5% improvement compared to the accuracy of pure BiLstM-CRF function point recognition.

关键词： Software Scale Evaluation BiLstM-CRF Domain Adaptation learning

来源：评论

学校读者我要写书评

暂无评论

learning from Partially Labeled Sequences for Behavioral Signal Annotation 7th

Learning from Partially Labeled Sequences for Behavioral Sig...

引用

7th international workshop on machine learning and data mining for Sports Analytics (MLSA)

作者： Aniszewska-stepien, Anna Herault, Romain Hacques, Guillaume Seifert, Ludovic Gasso, Gilles INSA Rouen Normandy St Etienne Du Rouvray France Rouen Univ Normandy Mont St Aignan France

ISBN: (纸本)9783030649111;9783030649128

Herewith, we present a learning procedure that allows to deal with a partially labeled sequence dataset, i.e. when each sequence in the train dataset may contain labeled as well as unlabeled chunks. In our application case, this occurs when motor activity has been manually annotated (due to the recognition based on the video recording) and independently registered by the measuring system of high precision (touch sensors): human annotation misses some events that have been captured by the sensors. In the general setting, we aim at predicting the labels for a new fully unlabeled movement sequence, while the training has been performed on the partially labeled dataset. For this purpose we propose to use classical sequence model (hidden Markov model) that is furnished with a constrained Viterbi algorithm, which gives us a quick access to the hard approximation of the correct labeling sequences. We demonstrate, that this simple modification that constrained Viterbi provide, allows the HMM model to be trained on sparse data, and overall results in surprisingly high log-likelihood and accuracy level in annotating the partially labeled behavioral sequences in climbing. The same time we show the way to access correct labeling of the unannotated signal that can be helpful in various sport science studies for movement pattern sequential prediction.

关键词： Partially labeled data Hidden Markov model Constrained viterbi algorithm Behavioral signal labeling Climbing pattern discovery

来源：评论

学校读者我要写书评

暂无评论

Virtual astronomy, information technology, and the new scientific methodology

Virtual astronomy, information technology, and the new scien...

引用

7th international workshop on Computer Architecture for machine Perception

作者： Djorgovski, SG CALTECH Div Phys Math & Astron Pasadena CA 91125 USA

ISBN: (纸本)0769522556

All sciences, including astronomy, are now entering the era of information abundance. The, exponentially increasing volume and complexity of modern, data sets promises to transform the scientific practice, but also, poses a number of common technological challenges. The Virtual Observatory concept is the astronomical community's response to these challenges: it aims to harness the progress in information technology in the service of astronomy, and at the same time provide a valuable testbed for information technology and applied computer science. Challenges broadly fall into two categories: data handling (or "data farming"), including issues such as archives, intelligent storage, databases, interoperability, fast networks, etc., and data mining, data understanding, and knowledge discovery, which include issues such as automated clustering and classification, multivariate correlation searches, pattern recognition, visualization in highly hyperdimensional parameter spaces, etc., as well as various applications of machine learning in these contexts. Such techniques are forming a methodological foundation for science with massive and complex data sets in general, and are likely to have a much broather impact on the modern society, commerce, information economy, security, etc. There is a powerful emerging synergy between the computationally enabled science and the science-driven computing, which will drive the progress in science, scholarship, and many other venues in the 21st century.

关键词： astronomy data management information technology knowledge acquisition knowledge representation scientific visualization

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：