检索结果-内蒙古大学图书馆

Cancer Evolution: Mathematical models and Computational Inference

SYSTEMATIC BIOLOGY 2015年第1期64卷 E1-E25页

作者： Beerenwinkel, Niko Schwarz, Roland F. Gerstung, Moritz Markowetz, Florian Swiss Fed Inst Technol Dept Biosyst Sci & Engn CH-4058 Basel Switzerland SIB CH-4058 Basel Switzerland European Bioinformat Inst European Mol Biol Lab Hinxton CB10 1SA Cambs England Wellcome Trust Sanger Inst Hinxton CB10 1SA Cambs England Univ Cambridge Canc Res UK Cambridge Inst Cambridge CB20RE England

Cancer is a somatic evolutionary process characterized by the accumulation of mutations, which contribute to tumor growth, clinical progression, immune escape, and drug resistance development. Evolutionary theory can be used to analyze the dynamics of tumor cell populations and to make inference about the evolutionary history of a tumor from molecular data. We review recent approaches to modeling the evolution of cancer, including population dynamics models of tumor initiation and progression, phylogenetic methods to model the evolutionary relationship between tumor subclones, and probabilistic graphical models to describe dependencies among mutations. Evolutionary modeling helps to understand how tumors arise and will also play an increasingly important prognostic role in predicting disease progression and the outcome of medical interventions, such as targeted therapy.

关键词： Cancer cancer progression evolution population genetics probabilistic graphical models

来源：评论

学校读者我要写书评

暂无评论

Learning Bayesian network classifiers by risk minimization

引用

INTERNATIONAL JOURNAL OF APPROXIMATE REASONING 2012年第2期53卷 248-272页

作者： Kelner, Roy Lerner, Boaz Ben Gurion Univ Negev Dept Ind Engn & Management IL-84105 Beer Sheva Israel Ben Gurion Univ Negev Dept Elect & Comp Engn IL-84105 Beer Sheva Israel

Bayesian networks (BNs) provide a powerful graphical model for encoding the probabilistic relationships among a set of variables, and hence can naturally be used for classification. However, Bayesian network classifiers (BNCs) learned in the common way using likelihood scores usually tend to achieve only mediocre classification accuracy because these scores are less specific to classification, but rather suit a general inference problem. We propose risk minimization by cross validation (RMCV) using the 0/1 loss function, which is a classification-oriented score for unrestricted BNCs. RMCV is an extension of classification-oriented scores commonly used in learning restricted BNCs and non-BN classifiers. Using small real and synthetic problems, allowing for learning all possible graphs, we empirically demonstrate RMCV superiority to marginal and class-conditional likelihood-based scores with respect to classification accuracy. Experiments using twenty-two real-world datasets show that BNCs learned using an RMCV-based algorithm significantly outperform the naive Bayesian classifier (NBC), tree augmented NBC (TAN), and other BNCs learned using marginal or conditional likelihood scores and are on par with non-BN state of the art classifiers, such as support vector machine, neural network, and classification tree. These experiments also show that an optimized version of RMCV is faster than all unrestricted BNCs and comparable with the neural network with respect to run-time. The main conclusion from our experiments is that unrestricted BNCs, when learned properly, can be a good alternative to restricted BNCs and traditional machine-learning classifiers with respect to both accuracy and efficiency. (C) 2011 Elsevier Inc. All rights reserved.

关键词： Bayesian networks Classification probabilistic graphical models Structure learning

来源：评论

学校读者我要写书评

暂无评论

Multidimensional Team Communication Modeling for Adaptive Team Training: A Hybrid Deep Learning and graphical Modeling Framework 1

引用

22nd International Conference on Artificial Intelligence in Education (AIED) - Mind the Gap - AIED for Equity and Inclusion

作者： Min, Wookhee Spain, Randall Saville, Jason D. Mott, Bradford Brawner, Keith Johnston, Joan Lester, James North Carolina State Univ Raleigh NC 27695 USA US Army Combat Capabil Dev Command Orlando FL 32826 USA

ISBN: (数字)9783030782924

ISBN: (纸本)9783030782924;9783030782917

Team communication modeling offers great potential for adaptive learning environments for team training. However, the complex dynamics of team communication pose significant challenges for team communication modeling. To address these challenges, we present a hybrid framework integrating deep learning and probabilistic graphical models that analyzes team communication utterances with respect to the intent of the utterance and the directional flow of communication within the team. The hybrid framework utilizes conditional random fields (CRFs) that use deep learning-based contextual, distributed language representations extracted from team members' utterances. An evaluation with communication data collected from six teams during a live training exercise indicate that linear-chain CRFs utilizing ELMo utterance embeddings (1) outperform both multi-task and single-task variants of stacked bidirectional long short-term memory networks using the same distributed representations of the utterances, (2) outperform a hybrid approach that uses non-contextual utterance representations for the dialogue classification tasks, and (3) demonstrate promising domaintransfer capabilities. The findings suggest that the hybrid multidimensional team communication analysis framework can accurately recognize speaker intent and model the directional flow of team communication to guide adaptivity in team training environments.

关键词： Team communication analytics probabilistic graphical models Deep learning Distributed language representations Natural language processing

来源：评论

学校读者我要写书评

暂无评论

Human identification with dental panoramic radiographic images

引用

IET BIOMETRICS 2018年第4期7卷 349-355页

作者： Oktay, Ayse Betul Istanbul Medeniyet Univ Fac Engn & Nat Sci Dept Comp Engn Istanbul Turkey

Dental panoramic radiographic images are commonly used as biometrics for human identification. In this study, a novel method is presented for identifying humans by matching 2D panoramic dental X-ray images. Each tooth is first identified and labeled with support vector machines and probabilistic graphical models. Missing teeth and dental restorations are also detected. Then, matching scores between images are calculated according to tooth-wise appearance and geometric similarities by taking dental restorations into account. The method is tested on a dataset including 206 dental panoramic X-ray images of 170 different subjects. The proposed method has 81% rank-1 accuracy and 89% rank-2 accuracy.

关键词： dentistry support vector machines probability image restoration diagnostic radiography human identification dental panoramic radiographic images biometrics 2D panoramic dental X-ray images support vector machines probabilistic graphical models dental restorations teeth restorations tooth-wise appearance geometric similarities

来源：评论

学校读者我要写书评

暂无评论

Graph Entropy-Based Learning Analytics 23rd

Graph Entropy-Based Learning Analytics

引用

23rd International Conference on Artificial Intelligence in Education (AIED)

作者： Al-Zawqari, Ali Vandersteen, Gerd Vrije Univ Brussel Dept ELEC B-1050 Brussels Belgium

ISBN: (纸本)9783031116476;9783031116469

The goal of this research is to strengthen the teaching strategy with quantitatively measured learning analytics. The entropy-based learning analytics aims to measure and understand students' progress by quantitatively measuring the difference between the content to be learned, the tutors' expectation of understanding, and the student's knowledge. This quantification will take similar steps than taken by Shannon for his information theory using a mathematical formalism to quantitatively measure knowledge (equivalent to Shannon's entropy) and knowledge transfer (equivalent to Shannon's mutual information). Knowledge graphs will be used to represent the content to be learned, the tutors' expectations, and the student's knowledge. Early results reveal that advanced analytical algorithms and graph entropy specified for educational applications is necessary for this research project to succeed.

关键词： probabilistic graphical models Shannon theory Graph entropy Learning analytics

来源：评论

学校读者我要写书评

暂无评论

Updating credal networks is approximable in polynomial time

引用

INTERNATIONAL JOURNAL OF APPROXIMATE REASONING 2012年第8期53卷 1183-1199页

作者： Maua, Denis D. de Campos, Cassio P. Zaffalon, Marco IDSIA CH-6928 Manno Switzerland

Credal networks relax the precise probability requirement of Bayesian networks, enabling a richer representation of uncertainty in the form of closed convex sets of probability measures. The increase in expressiveness comes at the expense of higher computational costs. In this paper, we present a new variable elimination algorithm for exactly computing posterior inferences in extensively specified credal networks, which is empirically shown to outperform a state-of-the-art algorithm. The algorithm is then turned into a provably good approximation scheme, that is, a procedure that for any input is guaranteed to return a solution not worse than the optimum by a given factor. Remarkably, we show that when the networks have bounded treewidth and bounded number of states per variable the approximation algorithm runs in time polynomial in the input size and in the inverse of the error factor, thus being the first known fully polynomial-time approximation scheme for inference in credal networks. (C) 2012 Elsevier Inc. All rights reserved.

关键词： probabilistic graphical models Credal networks Approximation scheme Valuation algebra

来源：评论

学校读者我要写书评

暂无评论

A novel semi-naive Bayesian restaurant recommendation method considering feature correlation 5

A novel semi-naive Bayesian restaurant recommendation method...

引用

5th International Conference on Computer Engineering and Application (ICCEA)

作者： Yu, Yulong Zhang, Hongyu Cent South Univ Sch Business Changsha Peoples R China

ISBN: (纸本)9798350386783;9798350386776

This paper introduces a novel recommendation model, namely, the maximum feature dependency semi-naive Bayesian (MDN) model, which is aimed at addressing feature correlation in restaurant selection. Existing restaurant recommendation methods ignore feature correlation and negative samples to improve recommendation accuracy. In contrast, the proposed recommendation model establishes maximum dependency features for each attribute, using both positive and negative samples to enhance its accuracy. The model first performs logarithmic division transformation on naive Bayesian classification to obtain user behavior patterns that can be sorted. Then, it combines frequent pattern mining to optimize user behavior patterns and obtain a semi-naive Bayesian recommendation model for restaurant recommendations. Experiments on a Yelp dataset show that the restaurant recommendation algorithm based on naive Bayesian user behavior patterns outperforms other basic recommendation methods. The proposed recommendation model stably improves recommendation methods based on naive Bayesian user behavior patterns, whereas other semi-naive Bayesian methods reduce accuracy due to the consideration of feature correlation. The restaurant recommendation method based on user behavior patterns also improves the interpretability of restaurant recommendations.

关键词： Semi-naive Bayesian Restaurant recommendation User behavior pattern Frequent pattern probabilistic graphical models Interpretability

来源：评论

学校读者我要写书评

暂无评论

probabilistic Inference of the Packet Delivery Ratio in Industrial Wireless Sensor Networks

Probabilistic Inference of the Packet Delivery Ratio in Indu...

引用

International Joint Conference on Neural Networks (IJCNN)

作者： Florencio, Heitor Neto, Adriao Doria Fed Univ Rio Grande Norte UFRN Dept Automat & Comp Engn Natal RN Brazil

ISBN: (纸本)9781509060146

The combined use of graphical models and probabilistic techniques has been shown to be highly effective in applications involving the uncertain data from industrial environments. In addition, industrial data analysis has been a necessary and interesting approach for systems in industry 4.0. The present paper proposes a probabilistic graphical model to infer the probability of the path presenting a packet delivery ratio (PDR) above a threshold. The technique used was Bayesian Network, however selection and discretization techniques were applied prior to data processing. These data were collected froma real WirelessHART network. The results show the applicability of the probabilistics method to real problems of industrial network.

关键词： probabilistic graphical models Indutrial Wireless Sensor Network Bayesian models WirelessHART Network

来源：评论

学校读者我要写书评

暂无评论

PROOF-READING GUIDANCE IN CELL TRACKING BY SAMPLING FROM TRACKING-BY-ASSIGNMENT models 12

PROOF-READING GUIDANCE IN CELL TRACKING BY SAMPLING FROM TRA...

引用

IEEE 12th International Symposium on Biomedical Imaging

作者： Schiegg, Martin Heuer, Ben Haubold, Carsten Wolf, Steffen Koethe, Ullrich Hamprecht, Fred A. Heidelberg Univ HCI IWR Speyerer Str 6 D-69115 Heidelberg Germany

ISBN: (纸本)9781479923748

Automated cell tracking methods are still error-prone. On very large data sets, uncertainty measures are thus needed to guide the expert to the most ambiguous events so these can be corrected with minimal effort. We present two easy-to-use methods to sample multiple proposal solutions from a tracking-by-assignment graphical model and experimentally evaluate the benefits of the uncertainty measures derived. Expert time for proof-reading is reduced greatly compared to random selection of predicted events.

关键词： Cell tracking uncertainty machine learning probabilistic graphical models

来源：评论

学校读者我要写书评

暂无评论

Unifying Guilt-by-Association Approaches: Theorems and Fast Algorithms

引用

European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases (ECML PKDD)

作者： Koutra, Danai Ke, Tai-You Kang, U. Chau, Duen Horng (Polo) Pao, Hsing-Kuo Kenneth Faloutsos, Christos Carnegie Mellon Univ Sch Comp Sci Pittsburgh PA 15213 USA Natl Taipei Univ Technol Dept Comp Sci & Informat Engn Taipei Taiwan

ISBN: (纸本)9783642237836;9783642237829

If several friends of Smith have committed petty thefts, what would you say about Smith? Most people would not be surprised if Smith is a hardened criminal. Guilt-by-association methods combine weak signals to derive stronger ones, and have been extensively used for anomaly detection and classification in numerous settings (e.g., accounting fraud, cyber-security, calling-card fraud). The focus of this paper is to compare and contrast several very successful, guilt-by-association methods: Random Walk with Restarts, Semi-Supervised Learning, and Belief Propagation (BP). Our main contributions are two-fold: (a) theoretically, we prove that all the methods result in a similar matrix inversion problem;(b) for practical applications, we developed FABP, a fast algorithm that yields 2 x speedup, equal or higher accuracy than BP, and is guaranteed to converge. We demonstrate these benefits using synthetic and real datasets, including YahooWeb, one of the largest graphs ever studied with BP.

关键词： Belief Propagation Random Walk with Restart Semi-Supervised Learning probabilistic graphical models inference

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：