Search

In recent years, there is a increase in researchers' interest on social evidence, particularly for fake news detection (FND). However, news posts on social media often include diverse modalities, e.g., text, image, etc., and diverse events related to politics, economics, etc., resulting in significant modality and event differences. How to jointly learn modal-invariant and event-invariant discriminative features effectively of news posts remains a challenge. This paper proposes a novel FND approach, called Cycle Mapping and Adversarial Event Classification Network (CMAECN). It consists of two parts: a multi-modal cycle feature mapping module (CMM) and an adversarial event classification module (AECM). In order to fully reduce modality difference, the CMM module is designed, which performs cross-modal generation between image and text modalities by using the generative model, conducts feature source identification between initial and generated features for each modality with the discriminative model, and reconstructs text or image features with the cross-modal fused features to avoid information loss with the reconstructor. In order to fully reduce event difference, the AECM module is designed to perform event adversarial classification between the event classification task and the event-independent classification task with a multi-task event classifier, where each dimension of the classifier output corresponds to a certain event category, and an additional dimension of the output represents the event-independent category. The network training of CMAECN is conducted by adopting an adversarial scheme. Comprehensive experiments are conducted on two public datasets, and CMAECN shows superior performance compared to the state-of-the-art multi-modal FND methods. [ABSTRACT FROM AUTHOR]

Existing semi-supervised domain adaptation (SSDA) approaches on visual classification usually assume that the labelled source data are only collected from single modality. However, since single source data cannot fully show the characteristics of the target data, source domain may be collected from multiple modalities (i.e. RGB and depth modalities). Traditional domain adaptation (DA) task makes an unrealistic scenario, where the label space in the source equals to the label space in the target. However, in real-world scenario, source and target domains may have different label spaces. Thus, the irrelevant categories in the source domain will cause two challenges: negative transfer and imbalanced distribution. In this paper, we design a novel deep SSDA framework in an end-to-end fashion, termed Local Weight Coupled Network (LWCN) for effective knowledge transfer, which aims to take advantage of the multi-modal information in the source domain and tackle the mentioned challenges, simultaneously. Specially, we construct the output layer including classification and regression, where the multi-class classifier and the multi-layer feature extractor can be learned jointly for mutual benefits. Empirical evaluations on five cross-domain benchmarks illustrate the competitive performance of our model with respect to the state-of-the-art, especially under the unequal categories scenario. [ABSTRACT FROM AUTHOR]

Heterogeneous defect prediction (HDP) refers to identifying more likely defect-proneness of software modules in a target project using heterogeneous metric data from other source projects, which solves the heterogeneous metric problem in cross-project defect prediction. Recently, several mixed-project HDP methods have been presented. However, these models neglect to address the linear inseparability and cross-project class imbalance issues simultaneously. These limitations usually lead to the unsatisfactory performance of HDP. In this paper, we propose an improved transfer learning approach for mixed-project HDP to deal with the above limitations, called data sampling and kernel manifold discriminant alignment (DSKMDA). DSKMDA firstly applies data sampling technique to handle the class imbalance issue. Then it uses kernel manifold discriminant alignment technique to handle the linear inseparability issue. Extensive experiments on 13 projects from three public benchmark datasets with four evaluation measures demonstrate that DSKMDA can produce better or comparable results against a range of competing methods. [ABSTRACT FROM AUTHOR]

Recently, actor-critic structure based neural networks are widely used in many reinforcement learning tasks. It consists of two main parts: (i) an actor module which outputs the probability distribution of action, and (ii) a critic module which outputs the predicted value based on the current environment. Actor-critic structure based networks usually need expert demonstration to provide an appropriate pre-training for the actor module, but the demonstration data is often hard or even impossible to obtain. And most of them, such as those used in the maze and robot control tasks, suffer from a lack of proper pre-training and unstable error propagation from the critic module to the actor module, which would result in poor and unstable performance. Therefore, a specially designed module which is called relatively optimal historical information learning (ROHI) is proposed. The proposed ROHI module can record the historical explored information and obtain the relatively optimal information through a customized merging algorithm. Then, the relatively optimal historical information is used to assist in training the actor module during the main learning process. We introduce two complex experimental environments, including the complex maze problem and flipping game, to evaluate the effectiveness of the proposed module. The experimental results demonstrate that the extended models with ROHI can significantly improve the success rate of the original actor-critic structure based models and slightly decrease the number of iteration required to reach the stable phase of value-based networks. [ABSTRACT FROM AUTHOR]

In recent years, graph convolutional networks have achieved great success in unsupervised domain adaptation task. Although these works make effort to reduce the distribution difference between domains, they do not take into account the issue of distribution difference reduction in the class level. In this paper, we propose a Dual-aligned Unsupervised Domain Adaptation with Graph Convolutional Networks (DUDA-GCN) framework to align domain distributions and the distributions of two domains corresponding to each category jointly. The framework contains two parts, i.e., a cross-domain feature extractor and a dual aligner of distribution. The former adopts a two-channel sub-network, with each channel to fully explore the relation among within-domain samples, based on GCN with shared weights to learn common feature representations of two domains. The dual aligner contains an adversarial domain discriminator and a category aligner, where the domain discriminator is designed to reduce the distribution difference across domains. A pseudo-label generator is designed to generate pseudo-labels for unlabeled samples. With the generated pseudo-labels of unlabeled samples and the real labels of labeled samples, the category aligner aligns the sample distributions across domains of the same category. Extensive empirical evaluation on three real-world datasets shows that DUDA-GCN can perform better than state-of-the-art related domain adaptation methods. [ABSTRACT FROM AUTHOR]

Learning an expressive representation from multi-view data is a crucial step in various real-world applications. In this paper, we propose a semi-supervised multi-view representation learning approach named Co-Embedding. Unlike conventional multi-view representation learning methods use joint to concatenate different views, which ignores information exchange between views and limits the methods' ability to utilize complementarity of multi-view data, Co-Embedding fulfills mutual help between views by coordinating and exchanging information between views. Specifically, we first build a weighted deep metric learning based multi-view representation learning framework in Co-Embedding. This framework models multi-view information through coordinate alignment, which is customized to exploit the complementary information from well-learned representations to help with modeling the under-learned representations. Then, by exploiting consensus property and neighborhood information, we design a multi-view label propagation algorithm that labels unlabeled data for Co-Embedding. Experimental results on seven benchmark multi-view datasets demonstrate the effectiveness of Co-Embedding. [ABSTRACT FROM AUTHOR]

API tutorials are important learning resources as they explain how to use certain APIs in a given programming context. An API tutorial can be split into a number of units. Consecutive units that describe a same topic are often called a tutorial fragment. We consider the API explained by a tutorial fragment as an API tag. Generating API tags for a tutorial fragment can help understand, navigate, and retrieve the fragment. Existing approaches often do not perform well on API tag generation due to high manual effort and low accuracy. Like API tutorials, Stack Overflow (SO) is also an important learning resource that provides the explanations of APIs. Thus, SO posts also contain API tags. Besides, API tags of SO posts are abundant and can be extracted easily. In this paper, we propose a novel approach ATTACK (stands for A PI T ag for T utorial frA gments using C rowd K nowledge), which can automatically generate API tags for tutorial fragments from SO posts. ATTACK first constructs Q & A p a i r , t a g s e t pairs by extracting API tags of SO posts. Then, it trains a deep neural network with the attention mechanism to learn the semantic relatedness between Q&A pairs and the associated API tags, taking into consideration both textual descriptions and code in a Q&A pair. Finally, the trained model is used to generate API tags for tutorial fragments. We evaluate ATTACK on public Java and Android datasets containing 43,132 Q & A p a i r , t a g s e t pairs. Experimental results show that ATTACK is effective and outperforms the state-of-the-art approaches in terms of F-Measure. Our user study further confirms the effectiveness of ATTACK in generating API tags for tutorial fragments. We also apply ATTACK to document linking and the results confirm the usefulness of API tags generated by ATTACK. [ABSTRACT FROM AUTHOR]

Matching video clips of people across non-overlapping surveillance cameras (video-based person re-identification) is of significant importance in many real-world applications. In this paper, we address the video-based person re-identification by developing a Local and Global Aligned Spatiotemporal Attention (LGASA) network. Our LGASA network consists of five cascaded modules, including 3D convolutional layers, residual block, spatial transformer network (STN), multi-stream recurrent network and multiple-attention module. Specifically, the 3D convolutional layers are used to capture local short-term fast-varying motion information encoded in multiple adjacent original frames. The residual block is used to extract mid-level feature maps. STN is applied to align the mid-level feature maps. The multi-stream recurrent network is designed to exploit the useful local and global long-term temporal dependency from the aligned mid-level feature maps. The multiple-attention module is designed to aggregate feature vectors of the same body part (or global) from different frames within each video into a single vector according to their importance. Experimental results on three video pedestrian datasets verify the effectiveness of the proposed local and global aligned spatiotemporal attention network. [ABSTRACT FROM AUTHOR]

Recently, person re-identification technique has been successfully applied to many fields, such as suspect tracking and lost human location. As video always contains more valuable information, more and more researchers focus on video based person re-identification, especially in image-to-video person re-identification (IVPR). However, most of existing IVPR models are under the supervised framework. In fact, marking enough training samples will cost numbers of labors, which limits the practical value of them. At the same time, the 2D features extracted from pedestrian image and 3D features extracted from pedestrian video are heterogeneous, which brings significant challenge for IVPR task. To effective solve the above problems, we propose an unsupervised domain adaption image-to-video person re-identification model by cross-modal feature generating and target information preserving transfer network (CMGTN). On one hand, the designed generator in our model can not only transform target domain unlabeled sample features into source domain feature space, but also can preserve target identity information. On the other hand, we eliminate the gap between pedestrian images and videos by embedding a cross-modal loss term. To evaluate the performance of our approach, we conduct extensive experiments on PRID-2011, iLIDS-VID and MARS datasets, and compare our approach with existing state-of-the-art IVPR models including four unsupervised methods and three supervised methods. Experimental results demonstrate the effectiveness of our approach. [ABSTRACT FROM AUTHOR]

This work focuses on unsupervised visual domain adaptation which is still challenging in visual recognition. Most of the attention has been dedicated to seeking the domain-invariant features of cross-domain data, but they ignores the valuable discriminative information in the source domain. In this paper, we propose a Discriminative Dictionary Evolution (DDE) approach to seek discriminative features robust to domain shift. Specifically, DDE gradually adapts a discriminative dictionary learned from the source domain to the target domain through a dictionary evolving procedure, in which self-selected atoms of the dictionary are updated with ℓ 2 , 1 -norm-based regularization. DDE produces domain-invariant representations for cross-domain visual recognition meanwhile promotes the discriminativeness of the dictionary. Empirical results on real-world data sets demonstrate the advantages of the proposed approach over existing competitive methods. [ABSTRACT FROM AUTHOR]

Recently, person re-identification (PR-ID) has attracted numerous of research interest because of its broad applications. However, most of the existing PR-ID models always follow the supervised framework, which requires substantial labeled data. In fact, it is often very hard to get enough labeled training samples in many practical application scenarios. To overcome this limitation, some semi-supervised PR-ID methods have been presented more recently. Although some of these semi-supervised models achieve satisfied results, there is still much space to improve. In this paper, we propose a novel semi-supervised PR-ID by similarity-embedded cycle GANs (SECGAN). Our SECGAN model can learn cross-view features with limited labeled data by using cycle GANs. Simultaneously, to further enhance the ability of cycle GANs so that it can extract more discriminative and robust features, similarity metric subnet and specific features extracting subnet are embedded into cycle GANs. Extensive experiments have been conducted on three public PR-ID benchmark datasets, and the experimental results show that our proposed SECGAN approach outperforms several typical supervised methods and the existing state-of-the-art semi-supervised methods including traditional and deep learning semi-supervised methods. [ABSTRACT FROM AUTHOR]

Video-based person re-identification (re-id), which aims to match people through videos captured by non-overlapping camera views, has attracted lots of research interest recently. In this paper, we first propose a novel hybrid 2D and 3D convolution-based recurrent neural network (HCRN) for video-based person re-id task. Specifically, the 3D convolutional module can explore the local short-term fast-varying motion information, while the recurrent layer can leverage the global long-term spatial–temporal information. Based on HCRN, we design a scale-fusion framework to make full use of features of different scales to further improve the performance of video-based person re-id. More concretely, the scale-fusion framework preserves a complete subnetwork similar to HCRN for each scale to extract features and exchanges information between all subnetworks at several stages of the framework. Besides, we propose a training method called species invasion to further improve the performance of HCRN and scale-fusion framework by utilizing a large amount of unlabeled data. Experimental results on the publicly available PRID 2011, iLIDS-VID and MARS multi-shot pedestrian re-id datasets demonstrate the effectiveness of the proposed HCRN, scale-fusion framework and species invasion training method. [ABSTRACT FROM AUTHOR]

Metric learning with triplet loss is one of the most effective methods for face verification, which aims to minimize the distance of positive pairs while maximizing the distance of negative pairs in feature embedding space. The arduous hard triplets mining and insufficient inter-class and intra-class variations are the two limitations of the previous methods. In this paper, we propose an improved triplet loss based on deep neural network for end-to-end metric learning, which can effectively cut down the number of the possible triplets and increase the proportion of hard triplets. It considers not only the relative distance between positive and negative pairs with the same probe images, but also the absolute distance between positive and negative pairs with different probe images, resulting in a smaller intra-class variation and a larger inter-class variation by adding a new constraint to push away negative pairs from positive pairs. In particular, a dynamic margin is proposed based on the distribution of positive and negative pairs in a batch, which can avoid the under-sampling or over-sampling problems. Our method is evaluated on LFW and YTF datasets, which are the most widely used benchmarks. The experimental results indicate that the proposed method can greatly outperform the standard triplet loss, and obtain the state-of-the-art performance with less time. [ABSTRACT FROM AUTHOR]

Person re-identification (re-id) plays an important role in video surveillance and forensics applications. In many cases, person re-id should be conducted between video clips, i.e., given a query pedestrian video from one camera, the re-id system should retrieve the video clips containing the same person from other cameras. However, person re-id between videos, which we call video-based person re-id, has not been well studied. In this paper, we propose a visual-appearance-level and spatial-temporal-level dictionary learning (VSDL) approach for video-based person re-id. Specifically, we first employ two kinds of models to represent each walking cycle in the video, i.e., visual-appearance features of all frames within the walking cycle, and a spatial-temporal feature vector. By separately learning a visual-appearance-level dictionary and a spatial-temporal-level dictionary from two kinds of representations, each walking cycle can be represented as a coding coefficient. To enhance the discriminative ability of the obtained coding coefficients, we design a representation coefficient discriminant term for VSDL. Experiments on the public iLIDS-VID and PRID 2011 datasets demonstrate the effectiveness of VSDL. [ABSTRACT FROM AUTHOR]

Different views of one object usually represent different aspects of the object, and a single view is unlikely to comprehensively describe the object. In multi-view learning, comprehensive utilization of multi-view information is helpful. In this paper, we propose a novel supervised latent subspace learning method called multi-view intact discriminant space learning (MIDSL) by efficiently integrating complementary multi-view information of different views. MIDSL learns a latent intact discriminant space by employing Fisher discrimination criterion to fully use class label information, which can well guide exploiting useful discriminant information, of labeled training samples. MIDSL can simultaneously minimize the within-class scatter and maximize the between-class scatter of the feature representations of different objects in the learned latent intact discriminant space. Aiming to utilize unlabeled samples to help mining more useful information for better learning latent intact discriminant space, we extend MIDSL method in semi-supervised scenario and propose semi-supervised multi-view intact discriminant space learning (SMIDSL) method. We further extend MIDSL and SMIDSL methods by kernel technique and propose kernelized multi-view intact discriminant space learning (KMIDSL) and kernelized semi-supervised multi-view intact discriminant space learning (KSMIDSL) methods. Experimental results on Caltech 101, LFW, MNIST and RGB-D datasets demonstrate the effectiveness of our proposed methods. [ABSTRACT FROM AUTHOR]

Logarithm difference is the subtraction of the center pixel and its neighbor in the face local region. The commonly-used assumption of the illumination invariant measure is that illumination intensities of neighbor pixels are approximately equal in the face local region. However, logarithm difference based illumination invariant measure performs unsatisfactorily under severe illumination variations, since severe varying illumination cause differences of illumination intensities are large in the face local region. In this paper, the general logarithm difference model (GLDM) is proposed to tackle severe illumination variations. In spired by the fact that a logarithm difference is the subtraction of two neighbor pixels, which may be a positive or negative numerical value, we divide a face local region into a positive region and a negative region, and the GLDM is developed by integrating positive and negative logarithm differences. Then, the multiscale logarithm difference edge-maps (MSLDE) [7] is employed as the test-bed, and the proposed GLDM is introduced into MSLDE to form General MSLDE (GMSLDE). Further, the proposed GMSLDE method is integrated with the advanced deep learning model VGG to obtain GMSLDE-VGG. Finally, the performance of GMSLDE and GMSLDE-VGG are verified not only on the Extended Yale B and CMU PIE face databases with severe illumination variations, but also on the LFW and our self-built Driver face databases with moderate illumination variations. The experimental results indicate that the proposed GLDM can efficiently improve the performance of logarithm difference edge-map against illumination variations, especially for severe illumination variations. [ABSTRACT FROM AUTHOR]

Heterogeneous defect prediction (HDP) refers to predicting defect-prone software modules in one project (target) using heterogeneous data collected from other projects (source). Recently, several HDP methods have been proposed. However, these methods do not sufficiently incorporate the two characteristics of the defect data: (1) data could be linear inseparable, and (2) data could be highly imbalanced. These two data characteristics make it challenging to build an effective HDP model. In this paper, we propose a novel Two-Stage Ensemble Learning (TSEL) approach to HDP, which contains two stages: ensemble multi-kernel domain adaptation (EMDA) stage and ensemble data sampling (EDS) stage. In the EMDA stage, we develop an Ensemble Multiple Kernel Correlation Alignment (EMKCA) predictor, which combines the advantage of multiple kernel learning and domain adaptation techniques. In the EDS stage, we employ RESample with replacement (RES) technique to learn multiple different EMKCA predictors and use average ensemble to combine them together. These two stages create an ensemble of defect predictors. Extensive experiments on 30 public projects show that the proposed TSEL approach outperforms a range of competing methods. The improvement is 20.14–33.92% in AUC, 36.05–54.78% in f-measure, and 5.48–19.93% in balance, respectively. [ABSTRACT FROM AUTHOR]

In practice, huge amount of samples is readily available, while labeled samples are often very limited and too expensive to be easily obtained. Multi-view features usually reveal different types of traits of labeled and unlabeled samples. Semi-supervised multi-view learning is a learning paradigm designed to meet the requirement of learning from complementary information of multiple views of labeled and unlabeled samples. In this paper, we propose a semi-supervised multiple kernel intact discriminant space learning (SMKIDSL) method to discover latent intact feature representations for those samples. SMKIDSL employs correlation discriminant analysis and label regression to fully use class label information for enhancing the discriminant power of latent intact feature representations. In SMKIDSL, multi-view collaboration learning mechanism is utilized to efficiently integrate complementary information of multiple views, which enables optimal view being dominant in learning process. Besides, kernel technique is used to tackle nonlinear issue of original multi-view features for exploiting more discriminant information. Comprehensive experiments are conducted on Caltech 101, LFW, MNIST and RGB-D datasets. And the experimental results demonstrate the effectiveness and efficiency of our proposed method. The robustness of our method is also confirmed by those results. [ABSTRACT FROM AUTHOR]

Low illumination is a common problem for recognition and tracking. Low illumination video-based person re identification (re-id) is an important application in practice. Low illumination usually results in severe loss of visual appearance and space-time information contained in pedestrian image or video, which brings large difficulty to re-identification. However, the problem of low illumination video-based person re-id (LIVPR) has not been well studied. In this paper, we propose a novel triplet-based manifold discriminative distance learning (TMD2L) approach for LIVPR. By regarding each video as an image set, TMD2L aims to learn a manifold-based distance metric, under which the intrinsic structure of image sets can be preserved, and the distance between truly matching sets is smaller than that between wrong matching sets. Experiment results on the new collected low illumination person sequence (LIPS) dataset, as well as two simulated datasets LI-PRID 2011 and LI-iLIDS-VID show that our proposed approach TMD2L outperforms existing representative person re-id methods. [ABSTRACT FROM AUTHOR]

Cross-project defect prediction (CPDP) refers to predicting defects in a target project using prediction models trained from historical data of other source projects. And CPDP in the scenario where source and target projects have different metric sets is called heterogeneous defect prediction (HDP). Recently, HDP has received much research interest. Existing HDP methods only consider the linear correlation relationship among the features (metrics) of the source and target projects, and such models are insufficient to evaluate nonlinear correlation relationship among the features. So these methods may suffer from the linearly inseparable problem in the linear feature space. Furthermore, existing HDP methods do not take the class imbalance problem into consideration. Unfortunately, the imbalanced nature of software defect datasets increases the learning difficulty for the predictors. In this paper, we propose a new cost-sensitive transfer kernel canonical correlation analysis (CTKCCA) approach for HDP. CTKCCA can not only make the data distributions of source and target projects much more similar in the nonlinear feature space, where the learned features have favorable separability, but also utilize the different misclassification costs for defective and defect-free classes to alleviate the class imbalance problem. We perform the Friedman test with Nemenyi’s post-hoc statistical test and the Cliff’s delta effect size test for the evaluation. Extensive experiments on 28 public projects from five data sources indicate that: (1) CTKCCA significantly performs better than the related CPDP methods; (2) CTKCCA performs better than the related state-of-the-art HDP methods. [ABSTRACT FROM AUTHOR]

Recently, multi-view dictionary learning technique has attracted lots of research interest. Although some multi-view dictionary learning methods have been addressed, there exists much room for improvement. How to explore and utilize both the diversity and the useful correlation information of different views with dictionaries has not been well studied. In this paper, we propose a novel multi-view dictionary learning approach named multi-view discriminant dictionary learning via learning view-specific and shared structured dictionaries (MDVSD), which aims to learn a structured dictionary shared by all views and multiple view-specific structured dictionaries with each corresponding to a specific view. The shared dictionary is combined with each view-specific dictionary to represent data of the specific view. MDVSD makes the view-specific dictionaries corresponding to different views uncorrelated for effectively exploring the diversity of different views. Furthermore, we introduce structural uncorrelation into shared dictionary learning procedure, such that the useful correlation information of different views can be effectively exploited. Dictionary-atoms in shared and view-specific dictionaries have correspondence to class labels so that the learned dictionaries have favorable discriminant ability and the obtained reconstruction error is discriminative. Three widely used datasets are employed as test data. Experimental results demonstrate the effectiveness of the proposed approach. [ABSTRACT FROM AUTHOR]

Kernel discriminant subspace learning technique is effective to exploit the structure of image dataset in the high-dimensional nonlinear space. However, for large-scale image recognition applications, this technique usually suffers from large computational burden. Although some kernel accelerating methods have been presented, how to greatly reduce computing time and simultaneously keep favorable recognition accuracy is still challenging. In this paper, we introduce the idea of parallel computing into kernel subspace learning and build a parallel kernel discriminant subspace learning framework. In this framework, we firstly design a random non-overlapping equal data division strategy to divide the whole training set into several subsets and assign each computational node a subset. Then, we separately learn kernel discriminant subspaces from these subsets without mutual communications and finally select the most appropriate subspace to classify test samples. Under the built framework, we propose two novel kernel subspace learning approaches, i.e., parallel kernel discriminant analysis (PKDA) and parallel kernel semi-supervised discriminant analysis (PKSDA). We show the superiority of the proposed approaches in terms of time complexity as compared with related methods, and provide the fundamental supports for our framework. For experiment, we establish a parallel computing environment and employ three public large-scale image databases as experiment data. Experimental results demonstrate the efficiency and effectiveness of the proposed approaches. [ABSTRACT FROM AUTHOR]

Software defect prediction can automatically predict defect-prone software modules for efficient software test in software engineering. When the previous defect labels of modules are limited, predicting the defect-prone modules becomes a challenging problem. In static software defect prediction, there exist the similarity among software modules, a software module can be approximated by a sparse representation of the other part of the software modules, and class-imbalance problem, the number of defect-free modules is much larger than that of defective ones. In this paper, we propose to use graph based semi-supervised learning technique to predict software defect. By using Laplacian score sampling strategy for the labeled defect-free modules, we construct a class-balance labeled training dataset firstly. And then, we use a nonnegative sparse algorithm to compute the nonnegative sparse weights of a relationship graph which serve as clustering indicators. Lastly, on the nonnegative sparse graph, we use a label propagation algorithm to iteratively predict the labels of unlabeled software modules. We thus propose a nonnegative sparse graph based label propagation approach for software defect classification and prediction, which uses not only few labeled data but also abundant unlabeled ones to improve the generalization capability. We vary the size of labeled software modules from 10 to 30 % of all the datasets in the widely used NASA projects. Experimental results show that the NSGLP outperforms several representative state-of-the-art semi-supervised software defect prediction methods, and it can fully exploit the characteristics of static code metrics and improve the generalization capability of the software defect prediction model. [ABSTRACT FROM AUTHOR]

In this paper, we present an active contour model for image segmentation based on a nonparametric distribution metric without any intensity a priori of the image. A novel nonparametric distance metric, which is called joint probability classification, is established to drive the active contour avoiding the instability induced by multimodal intensity distribution. Considering an image as a Riemannian manifold with spatial and intensity information, the contour evolution is performed on the image manifold by embedding geometric image feature into the active contour model. The experimental results on medical and texture images demonstrate the advantages of the proposed method. [ABSTRACT FROM AUTHOR]

Discriminant subspace learning is a widely used feature extraction technique for image recognition, since it can extract effective discriminant features by employing the class information and Fisher criterion. A crucial research topic on this technique is how to rapidly extract sufficient and effective features. Recently, recursive discriminant subspace learning technique has attracted lots of research interest because it can acquire sufficient discriminant features. Generally, it recursively decomposes image samples and extracts features from a number of decomposed sample sets. The major drawback of most recursive discriminant subspace learning methods is that they calculate the projective vectors one by one, such that they suffer from big computational costs. The recursive modified linear discriminant method and the incremental recursive Fisher linear discriminant method employ a simple solution for this problem, which calculates the class number minus one projective vectors in each recursion. However, this solution produces the unfavorable projective vectors with poor discriminant capabilities, and it cannot provide the terminating criterion for recursive computation and make the projective vectors orthogonal. In this paper, we propose a novel recursive learning approach that is group recursive discriminant subspace learning, which can rapidly learn multiple orthogonal subspaces with each spanned by a group of projective vectors. And we present a rule to select favorable projective vectors per recursion and provide a matrix-form-based terminating criterion to determine the number of recursions. Experiments on three widely used databases demonstrate the effectiveness and efficiency of the proposed approach. [ABSTRACT FROM AUTHOR]