2016 5th Brazilian Conference on Intelligent Systems (BRACIS)最新文献_第5页

Estimating the Number of Clusters Based on Sequential Clustering Algorithms 基于顺序聚类算法的聚类数量估计

2016 5th Brazilian Conference on Intelligent Systems (BRACIS) Pub Date : 2016-10-01 DOI: 10.1109/BRACIS.2016.050

E. M. Real

引用次数: 0

Fault Detection in Hard Disk Drives Based on Mixture of Gaussians 基于混合高斯分布的硬盘故障检测

2016 5th Brazilian Conference on Intelligent Systems (BRACIS) Pub Date : 2016-10-01 DOI: 10.1109/BRACIS.2016.036

Lucas P. Queiroz, Francisco Caio M. Rodrigues, J. Gomes, Felipe T. Brito, Iago C. Brito, Javam C. Machado

引用次数: 6

Efficient Minimal Learning Machines with Reject Option 具有拒绝选项的高效最小学习机

2016 5th Brazilian Conference on Intelligent Systems (BRACIS) Pub Date : 2016-10-01 DOI: 10.1109/BRACIS.2016.078

A. C. D. Oliveira, J. Gomes, A. Neto, A. Souza

引用次数: 3

Supervised Microalgae Classification in Imbalanced Dataset 不平衡数据集的监督微藻分类

2016 5th Brazilian Conference on Intelligent Systems (BRACIS) Pub Date : 2016-10-01 DOI: 10.1109/BRACIS.2016.020

Iago Correa, Paulo L. J. Drews-Jr, M. S. Souza, V. Tavano

引用次数: 12

Towards Fixation Prediction: A Nonparametric Estimation-Based Approach through Key-Points 注视预测:基于关键点的非参数估计方法

2016 5th Brazilian Conference on Intelligent Systems (BRACIS) Pub Date : 2016-10-01 DOI: 10.1109/BRACIS.2016.077

Saulo A. F. Oliveira, A. Neto, J. Gomes

引用次数: 1

A Dynamic Gesture Recognition System to Translate between Sign Languages in Complex Backgrounds 复杂背景下手语翻译的动态手势识别系统

2016 5th Brazilian Conference on Intelligent Systems (BRACIS) Pub Date : 2016-10-01 DOI: 10.1109/BRACIS.2016.082

D. Neiva, C. Zanchettin

引用次数: 2

Physical Topology Design of Optical Networks Aided by Many-Objective Optimization Algorithms 基于多目标优化算法的光网络物理拓扑设计

2016 5th Brazilian Conference on Intelligent Systems (BRACIS) Pub Date : 2016-10-01 DOI: 10.1109/BRACIS.2016.080

Elliackin M. N. Figueiredo, Danilo R. B. Araújo, C. J. A. B. Filho, Teresa B Ludermir

引用次数: 4

Markov Decision Processes Specified by Probabilistic Logic Programming: Representation and Solution 由概率逻辑规划指定的马尔可夫决策过程:表示与解

2016 5th Brazilian Conference on Intelligent Systems (BRACIS) Pub Date : 2016-10-01 DOI: 10.1109/BRACIS.2016.068

T. P. Bueno, D. Mauá, L. N. Barros, Fabio Gagliardi Cozman

引用次数: 1

A Meta-Learning Method to Select Under-Sampling Algorithms for Imbalanced Data Sets 不平衡数据集欠采样选择的元学习方法

2016 5th Brazilian Conference on Intelligent Systems (BRACIS) Pub Date : 2016-10-01 DOI: 10.1109/BRACIS.2016.076

Romero F. A. B. de Morais, P. Miranda, Ricardo Martins

{"title":"A Meta-Learning Method to Select Under-Sampling Algorithms for Imbalanced Data Sets","authors":"Romero F. A. B. de Morais, P. Miranda, Ricardo Martins","doi":"10.1109/BRACIS.2016.076","DOIUrl":"https://doi.org/10.1109/BRACIS.2016.076","url":null,"abstract":"Imbalanced data sets originating from real world problems, such as medical diagnosis, can be found pervasive. Learning from imbalanced data sets poses its own challenges, as common classifiers assume a balanced distribution of examples' classes in the data. Sampling techniques overcome the imbalance in the data by modifying the examples' classes distribution. Unfortunately, selecting a sampling technique together with its parameters is still an open problem. Current solutions include the brute-force approach (try as many techniques as possible), and the random search approach (choose the most appropriate from a random subset of techniques). In this work, we propose a new method to select sampling techniques for imbalanced data sets. It uses Meta-Learning and works by recommending a technique for an imbalanced data set based on solutions to previous problems. Our experimentation compared the proposed method against the brute-force approach, all techniques with their default parameters, and the random search approach. The results of our experimentation show that the proposed method is comparable to the brute-force approach, outperforms the techniques with their default parameters most of the time, and always surpasses the random search approach.","PeriodicalId":183149,"journal":{"name":"2016 5th Brazilian Conference on Intelligent Systems (BRACIS)","volume":"30 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122252505","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 6

Random Walk in Feature-Sample Networks for Semi-supervised Classification 面向半监督分类的特征样本网络随机行走

2016 5th Brazilian Conference on Intelligent Systems (BRACIS) Pub Date : 2016-10-01 DOI: 10.1109/BRACIS.2016.051

F. Verri, Liang Zhao

{"title":"Random Walk in Feature-Sample Networks for Semi-supervised Classification","authors":"F. Verri, Liang Zhao","doi":"10.1109/BRACIS.2016.051","DOIUrl":"https://doi.org/10.1109/BRACIS.2016.051","url":null,"abstract":"Positive-unlabeled learning is a semi-supervised task in which only some positive-labeled and many unlabeled samples are available. The goal of its transductive setting is to label all unlabeled data at once. In this paper, we developed a technique to grade positive-class pertinence levels of each sample, and we interpret the grades to classify the unlabeled ones. In our method, a sparse binary matrix represents the input data, which determines the feature–sample network whose vertices represent samples and attributes. The limiting probabilities of a random walk in the network estimate the pertinence levels. The results are evaluated regarding both class discrimination and classification accuracy. Computer simulations reveal that our model performs well in positive-unlabeled learning, especially with few labeled samples. Notably, the outcomes compare to the results from supervised methods, which profit from most data labeled. Additionally, the technique has linear time and space complexity if the input dataset is already in a sparse representation. The low computational cost of the construction and update of the feature–sample network allows for extensions of the technique to several learning problems, including online learning and dimensionality reduction.","PeriodicalId":183149,"journal":{"name":"2016 5th Brazilian Conference on Intelligent Systems (BRACIS)","volume":"27 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128469176","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 3