448 Hits in 8.5 sec

Advance Probabilistic Binary Decision Tree using SVM

Anita Meshram, Roopam Gupta, Sanjeev Sharma
2014 International Journal of Computer Applications  
In the testing phase, the algorithm depends on rooted binary directed acyclic graph to make a decision. So the classification of DAG is usually faster than OaO.  ...  Here proposed an algorithm Advance Probabilistic Binary Decision Tree (APBDT) using SVM for solving large class problem and it performs better when increase the size of the database.  ...  The proposed decision tree based OAA(DTOAA) aimed, to increasing the classification speed of OAA by use of posterior probability estimates of binary SVM output.  ... 
doi:10.5120/18956-0256 fatcat:rmgsnihrm5ct5ffh5o6fso4ssq

Hierarchical Ensemble Methods for Protein Function Prediction

Giorgio Valentini
2014 ISRN Bioinformatics  
In this paper, we provide a comprehensive review of hierarchical methods for protein function prediction based on ensembles of learning machines.  ...  According to this general approach, a separate learning machine is trained to learn a specific functional term and then the resulting predictions are assembled in a "consensus" ensemble decision, taking  ...  Acknowledgements The author thanks the reviewers for their comments and suggestions and acknowledges partial support from the PRIN project "Automi e linguaggi formali: aspetti matematici e applicativi"  ... 
doi:10.1155/2014/901419 pmid:25937954 pmcid:PMC4393075 fatcat:i6w56fpbqnekjozm2kdpik635e

Gene Ontology consistent protein function prediction: the FALCON algorithm applied to six eukaryotic genomes

Yiannis AI Kourmpetis, Aalt DJ van Dijk, Cajo JF ter Braak
2013 Algorithms for Molecular Biology  
The optimization is done using the Differential Evolution algorithm.  ...  For example, a protein may be predicted to belong to a detailed functional class, but not in a broader class that, due to the vocabulary structure, includes the predicted one.  ...  Acknowledgements We thank Roeland van Ham, James Holzwarth and the two reviewers for constructive remarks on the manuscript. YK was supported by the Biorange grant SP3.  ... 
doi:10.1186/1748-7188-8-10 pmid:23531338 pmcid:PMC3691668 fatcat:epybp63cqzbyfhpiowncuqfc7u

Pruning GP-Based Classifier Ensembles by Bayesian Networks [chapter]

C. De Stefano, G. Folino, F. Fontanella, A. Scotto di Freca
2012 Lecture Notes in Computer Science  
In addition, a comparison with a pareto optimal strategy of pruning has been performed. Table 3. Comparison results for the selection strategies.  ...  The framework is based on two modules: an ensemble-based Genetic Programming (GP) system, which produces a high performing ensemble of decision tree classifiers, and a Bayesian Network (BN) approach to  ...  Ensemble techniques have been also used for improving GP-based classification systems [2, 6, 9] .  ... 
doi:10.1007/978-3-642-32937-1_24 fatcat:ufkreir2mnbdrjde5fecupxmhq

An overview of ensemble methods for binary classifiers in multi-class problems: Experimental study on one-vs-one and one-vs-all schemes

Mikel Galar, Alberto Fernández, Edurne Barrenechea, Humberto Bustince, Francisco Herrera
2011 Pattern Recognition  
Classification problems involving multiple classes can be addressed in different ways.  ...  One of the most popular techniques consists in dividing the original data set into two-class subsets, learning a different binary model for each new subset.  ...  Decision directed acyclic graph (DDAG) [59] : DDAG method constructs a rooted binary acyclic graph where each node is associated to a list of classes and a binary classifier.  ... 
doi:10.1016/j.patcog.2011.01.017 fatcat:u6cytpcyzrcnrbhjwkcfqg26ji

A Bayesian Approach for Combining Ensembles of GP Classifiers [chapter]

C. De Stefano, F. Fontanella, G. Folino, A. Scotto di Freca
2011 Lecture Notes in Computer Science  
The first one applies a boosting technique to a GP-based classification algorithm in order to generate an effective decision trees ensemble.  ...  The second module uses a Bayesian network for combining the responses provided by such ensemble and select the most appropriate decision trees.  ...  A BN is a probabilistic graphical model that allows the representation of a joint probability distribution of a set of random variables through a Direct Acyclic Graph (DAG).  ... 
doi:10.1007/978-3-642-21557-5_5 fatcat:iboxqqhtxvakrinloftfakbb7e

Tree-based Ensemble Classifier Learning for Automatic Brain Glioma Segmentation

Samya Amiri, Mohamed Ali Mahjoub, Islem Rekik
2018 Neurocomputing  
Third, we train DMT using different scales for input patches and superpixels.  ...  Unlike previous works that simply aggregate or cascade classifiers for addressing image segmentation and labeling tasks, we propose to embed strong classifiers into a tree structure that allows bi-directional  ...  Bayesian network Bayesian networks are probabilistic graphical models based on directed acyclic graphs (DAGs) and Thomas Bayes' theorem in probability theory giving a graphical representation of the probabilistic  ... 
doi:10.1016/j.neucom.2018.05.112 fatcat:2uv3siv2f5ajpjewjitxm7fdk4

A Hierarchical Multi-Label Classification Algorithm for Gene Function Prediction

2017 Algorithms  
This paper proposed a novel HMC algorithm for solving this problem based on the Gene Ontology (GO), the hierarchy of which is a directed acyclic graph (DAG) and is more difficult to tackle.  ...  Gene function prediction is a complicated and challenging hierarchical multi-label classification (HMC) task, in which genes may have many functions at the same time and these functions are organized in  ...  A directed acyclic graph (DAG) for the Gene Ontology (GO) and a rooted tree for the Functional Catalogue (FunCat) are two main hierarchy structures of gene functional classes [5] .  ... 
doi:10.3390/a10040138 fatcat:kjr24toqi5dc5epx2mbbh4adpe

Synergy of multi-label hierarchical ensembles, data fusion, and cost-sensitive methods for gene functional inference

Nicolò Cesa-Bianchi, Matteo Re, Giorgio Valentini
2011 Machine Learning  
Besides classical top-down hierarchical multilabel ensemble methods, in our experiments we consider two recently proposed multilabel methods: one based on the approximation of the Bayesian optimal classifier  ...  Gene function prediction is a complex multilabel classification problem with several distinctive features: the hierarchical relationships between functional classes, the presence of multiple sources of  ...  Acknowledgements We would like to thank the anonymous reviewers for their comments and suggestions.  ... 
doi:10.1007/s10994-011-5271-6 fatcat:5uvavmu6zjft3g7xvtc6sghsya

Scalable multi-output label prediction: From classifier chains to classifier trellises

Jesse Read, Luca Martino, Pablo M. Olmos, David Luengo
2015 Pattern Recognition  
In this paper, we present the classifier trellis (CT) method for scalable multi-label classification.  ...  In particular, the methods' strategies for discovering and modeling a good chain structure constitutes a mayor computational bottleneck.  ...  directed trellis graph using Algorithm 1.  ... 
doi:10.1016/j.patcog.2015.01.004 fatcat:wki64hvjsfcsfkuqjff7gkvkia

A Weighted Majority Vote Strategy Using Bayesian Networks [chapter]

Luigi P. Cordella, Claudio De Stefano, Francesco Fontanella, Alessandra Scotto di Freca
2013 Lecture Notes in Computer Science  
In this paper we propose a new weighted majority vote rule, that uses the joint probabilities of each class as weights for combining classifier outputs.  ...  A Bayesian Network automatically infers the joint probability distribution for each class.  ...  Bayesian Network Properties A BN makes it possible to represent the joint probability distribution of a set of random variables through the structure of a Direct Acyclic Graph (DAG).  ... 
doi:10.1007/978-3-642-41184-7_23 fatcat:2nmh4gahbbgnnh5vyspxpmzqve

Comparing Combination Rules of Pairwise Neural Networks Classifiers

Olivier Lézoray, Hubert Cardot
2007 Neural Processing Letters  
Based on the above, we provide future research directions which consider the recombination problem as an ensemble method.  ...  In this paper, we focus on pairwise decomposition approach to multiclass classification with neural networks as the base learner for the dichotomies.  ...  This decoder was originally described by Kressel [21] and reintroduced by Platt [33] where it was called Directed Acyclic Graph (DAG).  ... 
doi:10.1007/s11063-007-9058-5 fatcat:zbkbs6exj5hhdpx44mgllapc6u

A Ternary Brain-Computer Interface Based on Single-Trial Readiness Potentials of Self-initiated Fine Movements: A Diversified Classification Scheme

Elias Abou Zeid, Alborz Rezazadeh Sereshkeh, Benjamin Schultz, Tom Chau
2017 Frontiers in Human Neuroscience  
The ternary classification was decomposed into binary classifications using a decision-directed acyclic graph (DDAG).  ...  A pipeline of spatio-temporal filtering with per participant parameter optimization was used for feature extraction.  ...  We applied the ODCS with a decision directed acyclic graph (ODCS-DDAG).  ... 
doi:10.3389/fnhum.2017.00254 pmid:28596725 pmcid:PMC5443161 fatcat:crrghmfacbeylbirzsxh2ostyi

Overview of integrative analysis methods for heterogeneous data

Jaya Thomas, Lee Sael
2015 2015 International Conference on Big Data and Smart Computing (BIGCOMP)  
We categorize integrative methods for heterogeneous data analysis to Bayesian network based methods and multiple kernel based methods and describe them in detail with examples of successful applications  ...  Then, what are the existing methods for utilizing theses heterogeneous data to improve data analysis and how can we choose amongst these methods?  ...  More formally, BN is a directed acyclic graph G that represents the joint probability distribution over the set of random variables X 1 ,X 2 ,. . .  ... 
doi:10.1109/35021bigcomp.2015.7072811 dblp:conf/bigcomp/ThomasS15 fatcat:gxnirhy2cff7xm73xyoxi32iri

Classifier Chains: A Review and Perspectives [article]

Jesse Read, Bernhard Pfahringer, Geoff Holmes, Eibe Frank
2020 arXiv   pre-print
this approach in the domain of multi-label classification in the future.  ...  This approach involves linking together off-the-shelf binary classifiers in a chain structure, such that class label predictions become features for other classifiers.  ...  Acknowledgements Thanks to Tomasz Kajdanowicz and Willem Waegeman who pointed out useful references and provided insightful discussion on classifier chains during the preparation of this paper.  ... 
arXiv:1912.13405v2 fatcat:nzpdycyj5fgrtp445sxfivzamm
« Previous Showing results 1 — 15 out of 448 results