A general graph-based semi-supervised learning with novel class discovery

Feiping Nie; Shiming Xiang; Yun Liu; Changshui Zhang
May 2010
Neural Computing & Applications;May2010, Vol. 19 Issue 4, p549
Academic Journal
In this paper, we propose a general graph-based semi-supervised learning algorithm. The core idea of our algorithm is to not only achieve the goal of semi-supervised learning, but also to discover the latent novel class in the data, which may be unlabeled by the user. Based on the normalized weights evaluated on data graph, our algorithm is able to output the probabilities of data points belonging to the labeled classes or the novel class. We also give the theoretical interpretations for the algorithm from three viewpoints on graph, i.e., regularization framework, label propagation, and Markov random walks. Experiments on toy examples and several benchmark datasets illustrate the effectiveness of our algorithm.


Related Articles

  • Smooth Harmonic Transductive Learning. Ying Xie; Bin Luo; Rongbin Xu; Sibao Chen // Journal of Computers;Dec2013, Vol. 8 Issue 12, p3079 

    In this paper, we present a novel semi-supervised smooth harmonic transductive learning algorithm that can get closed-form solution. Our method introduces the unlabeled class information to the learning process and tries to exploit the similar configurations shared by the label distribution of...

  • Confidence Estimation for Graph-based Semi-supervised Learning. Tao Guo; Guiyang Li // Journal of Software (1796217X);Jun2012, Vol. 7 Issue 6, p1307 

    To select unlabeled example effectively and reduce classification error, confidence estimation for graph-based semi-supervised learning (CEGSL) is proposed. This algorithm combines graph-based semi-supervised learning with collaboration-training. It makes use of structure information of sample...

  • Gradient Tree Boosting for Training Conditional Random Fields. Dietterich, Thomas G.; Hao, Guohua; Schenfelter, Adam // Journal of Machine Learning Research;10/1/2008, Vol. 9 Issue 10, p2113 

    Conditional random fields (CRFs) provide a flexible and powerful model for sequence labeling problems. However, existing learning algorithms are slow, particularly in problems with large numbers of potential input features and feature combinations. This paper describes a new algorithm for...

  • Learning From Crowds. Raykar, Vikas C.; Shipeng Yu; Zhao, Linda H.; Valadez, Gerardo Hermosillo; Florin, Charles; Bogoni, Luca; Moy, Linda // Journal of Machine Learning Research;4/1/2010, Vol. 11 Issue 4, p1297 

    For many supervised learning tasks it may be infeasible (or very expensive) to obtain objective and reliable labels. Instead, we can collect subjective (possibly noisy) labels from multiple experts or annotators. In practice, there is a substantial amount of disagreement among the annotators,...

  • Research on the Document Image Segmentation Based on the LDA Model. Hanyang Jiang // Advances in Information Sciences & Service Sciences;Feb2012, Vol. 4 Issue 3, p12 

    To aim the problem of document image segmentation, we propose a topic model based method to segment the document images into several areas, such as text, background, tables and figures. In the past, the segmentation of document images focused on threshold based method or supervised learning...

  • 3-D Depth Reconstruction from a Single Still Image. Saxena, Ashutosh; Chung, Sung; Ng, Andrew // International Journal of Computer Vision;Jan2008, Vol. 76 Issue 1, p53 

    We consider the task of 3-d depth estimation from a single still image. We take a supervised learning approach to this problem, in which we begin by collecting a training set of monocular images (of unstructured indoor and outdoor environments which include forests, sidewalks, trees, buildings,...

  • Learning to Combine Bottom-Up and Top-Down Segmentation. Levin, Anat; Weiss, Yair // International Journal of Computer Vision;Jan2009, Vol. 81 Issue 1, p105 

    Bottom-up segmentation based only on low-level cues is a notoriously difficult problem. This difficulty has lead to recent top-down segmentation algorithms that are based on class-specific image information. Despite the success of top-down algorithms, they often give coarse segmentations that...

  • MULTI-LABEL CLASSIFICATION USING ERROR CORRECTING OUTPUT CODES. KAJDANOWICZ, TOMASZ; KAZIENKO, PRZEMYS┼üAW // International Journal of Applied Mathematics & Computer Science;Dec2012, Vol. 22 Issue 4, p829 

    A framework for multi-label classification extended by Error Correcting Output Codes (ECOCs) is introduced and empirically examined in the article. The solution assumes the base multi-label classifiers to be a noisy channel and applies ECOCs in order to recover the classification errors made by...

  • Analysis of Various Clustering and Classification Algorithms in Datamining. Valsala, Sandhia; Thomas, Bindhya; George, Jissy Ann // International Journal of Computer Science & Network Security;Nov2012, Vol. 12 Issue 11, p54 

    Clustering and classification of data is a difficult problem that is related to various fields and applications. Challenge is greater, as input space dimensions become larger and feature scales are different from each other. The term "classification" is frequently used as an algorithm for all...


Read the Article


Sorry, but this item is not currently available from your library.

Try another library?
Sign out of this library

Other Topics