We can face with the patternrecognition problems where the influence of hidden context leads to more or less radical changes in the target concept. This paper proposes the mathematical and algorithmic framework for t...
详细信息
In this paper we propose Manifold-Regularized Minimax Probability machine, called MRMPM. We show that Minimax Probability machine can properly be extended to semi-supervised version in the manifold regularization fram...
详细信息
In many real-world applications there are usually abundant unlabeled data but the amount of labeled training examples are often limited, since labeling the data requires extensive human effort and expertise. Thus, exp...
详细信息
This paper describes a scalable solution for identifying influential subscribers in for example telecom networks. The solution estimates one weighted value of inuence out of several Social Network Analysis(SNA) metric...
详细信息
A semi-supervised version of Fisher's linear discriminant analysis is presented. As opposed to virtually all other approaches to semi-supervision, no assumptions on the data distribution are made, apart from the o...
详细信息
The proceedings contain 11 papers. The topics discussed include: connect the dots: the reconstruction of region boundaries from contour sampling points;digital shape analysis with maximal segments;discrete curvature e...
ISBN:
(纸本)9783642323126
The proceedings contain 11 papers. The topics discussed include: connect the dots: the reconstruction of region boundaries from contour sampling points;digital shape analysis with maximal segments;discrete curvature estimation methods for triangulated surfaces;on morphological hierarchical representations for image processing and spatial data clustering;volumetric analysis of digital objects using distance transformation: performance issues and extensions;geometric analysis of 3D electron microscopy data;machinelearning as a preprocessing phase in discrete tomography;fast planarity estimation and region growing on GPU;writing reusable digital topology algorithms in a generic image processing framework;and a new image-mining technique for automation of Parkinson's disease research.
The proceedings contain 59 papers. The special focus in this conference is on Rough Sets and Applications, machinelearning in patternrecognition and Image Processing, machinelearning in Multimedia Computing, Bioinf...
ISBN:
(纸本)9783642353253
The proceedings contain 59 papers. The special focus in this conference is on Rough Sets and Applications, machinelearning in patternrecognition and Image Processing, machinelearning in Multimedia Computing, Bioinformatics and Cheminformatics, data Classification and Clustering, Cloud Computing and Recommender Systems, Case-Based Reasoning and data Processing, Authentication, Digital Forensics and Plagiarism Detection. The topics include: Rough sets-based machinelearning over non-deterministic data;learning a table from a table with non-deterministic information;parameterised fuzzy Petri nets for approximate reasoning in decision support systems;rough sets-based rules generation approach;automatic color image segmentation based on illumination invariant and superpixelization;wavelet based statistical adapted local binary patterns for recognizing avatar faces;solving avatar captchas automatically;comparative analysis of image fusion techniques in remote sensing;density based fuzzy thresholding for image segmentation;subjectivity and sentiment analysis of Arabic;support vector machine approach for detecting events in video streams;study of feature categories for musical instrument recognition;a genetic-CBR approach for cross-document relationship identification;improved action recognition using an efficient boosting method;towards smart Egypt -the role of large scale WSNs;language for writing descriptors of outline shape of molecules;web service based approach for viral hepatitis ontology sharing and diagnosing;sampleboost for capsule endoscopy categorization and abnormality detection;advanced parallel genetic algorithm with gene matrix for global optimization;semi-possibilistic biclustering applied to discrete and continuous data;a comparative study of localization algorithms in WSNs;test cases automatic generator (TCAG);support vector machines with weighted powered kernels for data classification;an enhanced cloud-based view materialization approach for peer-to-
This paper studies a new machinelearningstrategy called co-transfer learning. Unlike many previous learning problems, we focus on how to use labeled data of different feature spaces to enhance the classification of ...
详细信息
ISBN:
(纸本)9781450315555
This paper studies a new machinelearningstrategy called co-transfer learning. Unlike many previous learning problems, we focus on how to use labeled data of different feature spaces to enhance the classification of different learning spaces simultaneously. For instance, we make use of both labeled images and labeled text data to help learn models for classifying image data and text data together. An important component of co-transfer learning is to build different relations to link different feature spaces, thus knowledge can be co-transferred across different spaces. Our idea is to model the problem as a joint transition probability graph. The transition probabilities can be constructed by using the intra-relationships based on affinity metric among instances and the inter-relationships based on co-occurrence information among instances from different spaces. The proposed algorithm computes ranking of labels to indicate the importance of a set of labels to an instance by propagating the ranking score of labeled instances via the random walk with restart. The main contribution of this paper is to (i) propose a co-transfer learning (CT-Learn) framework that can perform learning simultaneously by co-transferring knowledge across different spaces;(ii) show the theoretical properties of the random walk for such joint transition probability graph so that the proposed learning model can be used effectively;(iii) develop an efficient algorithm to compute ranking scores and generate the possible labels for a given instance. Experimental results on benchmark data (image-text and English-Chinese-French classification data sets) have shown that the proposed algorithm is computationally efficient, and effective in learning across different spaces. In the comparison, we find that the classification performance of the CT-Learn algorithm is better than those of the other tested transfer learning algorithms. Copyright 2012 ACM.
Advances in Eelectroencephalography (EEG) and functional magnetic resonance imaging (fMRI) have opened up the possibility for real time data classification. A small amount of labelled training data is usually availabl...
详细信息
Microarrays are standard tools for measuring thousands of gene expression levels simultaneously. They are frequently used in the classification process of tumor tissues. In this setting a collected set of samples ofte...
详细信息
暂无评论