Image colourisation using graph-based semi-supervised learning

A novel colourisation algorithm using graph-based semi-supervised learning (SSL) is presented. We show that the assumption of the colourisation problem is consistent with the fundamental of graph-based SSL methods. Satisfactory results are obtained in the experiments that validate the proposed algorithm. To reduce the time and memory requirements when dealing with large scale images, we further propose a two-stage speedup approach. Comparative results show that the computation complexity is dramatically reduced.     

A graph-based semi-supervised approach to classification learning in digital geographies

As the distinction between online and physical spaces rapidly degrades, social media have now become an integral component of how many people's everyday experiences are mediated. As such, increasing interest has emerged in exploring how the content shared through those online platforms comes to contribute to the collaborative creation of places in physical space at the urban scale. Exploring digital geographies of social media data using methods such as qualitative coding (i.e., content labelling) is a flexible but complex task, commonly limited to small samples due to its impracticality over large datasets. In this paper, we propose a new tool for studies in digital geographies, bridging qualitative and quantitative approaches, able to learn a set of arbitrary labels (qualitative codes) on a small, manually-created sample and apply the same labels on a larger set. We introduce a semi-supervised, deep neural network approach to classify geo-located social media posts based on their textual and image content, as well as geographical and temporal aspects. Our innovative approach is rooted in our understanding of social media posts as augmentations of the time-space configurations that places are, and it comprises a stacked multi-modal autoencoder neural network to create joint representations of text and images, and a spatio-temporal graph convolution neural network for semi-supervised classification. The results presented in this paper show that our approach performs the classification of social media content with higher accuracy than traditional machine learning models as well as two state-of-art deep learning frameworks.     

Instance selection method for improving graph-based semi-supervised learning

Graph-based semi-supervised learning is an important semi-supervised learning paradigm. Although graph-based semi-supervised learning methods have been shown to be helpful in various situations, they may adversely affect performance when using unlabeled data. In this paper, we propose a new graph-based semi-supervised learning method based on instance selection in order to reduce the chances of performance degeneration. Our basic idea is that given a set of unlabeled instances, it is not the best approach to exploit all the unlabeled instances; instead, we should exploit the unlabeled instances that are highly likely to help improve the performance, while not taking into account the ones with high risk. We develop both transductive and inductive variants of our method. Experiments on a broad range of data sets show that the chances of performance degeneration of our proposed method are much smaller than those of many state-of-the-art graph-based semi-supervised learning methods.     

A discriminant sparse representation graph-based semi-supervised learning for hyperspectral image classification

Multimedia Tools and Applications - The classification of hyperspectral image with a paucity of labeled samples is a challenging task. In this paper, we present a discriminant sparse representation...     

A novel semi-supervised learning framework with simultaneous text representing

Text representation has received extensive attention in text mining tasks. There are various text representation models. Among them, vector space model is the most commonly used one. For vector space model, the core technique is term weighting. To date, a great deal of different term-weighting methods have been proposed, which can be divided into supervised group and unsupervised group. However, it is not advisable to use these two groups of methods directly in semi-supervised applications. In semi-supervised applications, the majority of the supervised term-weighting methods are not applicable as the label information is insufficient; meanwhile, the unsupervised term-weighting methods cannot make use of the provided category labels. Thus, a semi-supervised learning framework for iteratively revising the text representation by an EM-like strategy is proposed in this paper. Furthermore, a new supervised term-weighting method t f.sd f is proposed. T f.sd f has the ability to emphasize the importance of terms that are unevenly distributed among all the classes and weaken the importance of terms that are uniformly distributed. Experimental results on real text data show that the proposed semi-supervised learning framework with the aid of t f.sd f performs well. Also, t f.sd f is shown to be efficient for supervised learning.     

A new graph-based semi-supervised method for surface defect classification

Vision-based defect classification is an important technology to control the quality of product in manufacturing system. As it is very hard to obtain enough labeled samples for model training in the real-world production, the semi-supervised learning which learns from both labeled and unlabeled samples is more suitable for this task. However, the intra-class variations and the inter-class similarities of surface defect, named as the poor class separation, may cause the semi-supervised methods to perform poorly with small labeled samples. While graph-based methods, such as graph convolution network (GCN), can solve the problem well. Therefore, this paper proposes a new graph-based semi-supervised method, named as multiple micrographs graph convolutional network (MMGCN), for surface defect classification. Firstly, MMGCN performs graph convolution by constructing multiple micrographs instead of a large graph, and labels unlabeled samples by propagating label information from labeled samples to unlabeled samples in the micrographs to obtain multiple labels. Weighting the labels can obtain the final label, which can solve the limitations of computation complexity and practicality of original GCN. Secondly, MMGCN divides unlabeled dataset into multiple batches and sets an accuracy threshold. When the model accuracy reaches the threshold, the unlabeled datasets are labeled in batches. A famous case has been used to evaluate the performance of the proposed method. The experimental results demonstrate that the proposed MMGCN can achieve better computation complexity and practicality than GCN. And for accuracy, MMGCN can also obtain the best performance and the best class separation in the comparison with other semi-supervised surface defect classification methods.     

Graph-based semi-supervised learning

Recent years have witnessed a surge of interest in graph-based semi-supervised learning. However, two of the major problems in graph-based semi-supervised learning are: (1) how to set the hyperparameter in the Gaussian similarity; and (2) how to make the algorithm scalable. In this article, we introduce a general framework for graphbased learning. First, we propose a method called linear neighborhood propagation, which can automatically construct the optimal graph. Then we introduce a novel multilevel scheme to make our algorithm scalable for large data sets. The applications of our algorithm to various real-world problems are also demonstrated.     

A unified framework for semi-supervised PU learning

Traditional supervised classifiers use only labeled data (features/label pairs) as the training set, while the unlabeled data is used as the testing set. In practice, it is often the case that the labeled data is hard to obtain and the unlabeled data contains the instances that belong to the predefined class but not the labeled data categories. This problem has been widely studied in recent years and the semi-supervised PU learning is an efficient solution to learn from positive and unlabeled examples. Among all the semi-supervised PU learning methods, it is hard to choose just one approach to fit all unlabeled data distribution. In this paper, a new framework is designed to integrate different semi-supervised PU learning algorithms in order to take advantage of existing methods. In essence, we propose an automatic KL-divergence learning method by utilizing the knowledge of unlabeled data distribution. Meanwhile, the experimental results show that (1) data distribution information is very helpful for the semi-supervised PU learning method; (2) the proposed framework can achieve higher precision when compared with the state-of-the-art method.     

Robust self-tuning semi-supervised learning

We investigate the issue of graph-based semi-supervised learning (SSL). The labeled and unlabeled data points are represented as vertices in an undirected weighted neighborhood graph, with the edge weights encoding the pairwise similarities between data objects in the same neighborhood. The SSL problem can be then formulated as a regularization problem on this graph. In this paper we propose a robust self-tuning graph-based SSL method, which (1) can determine the similarities between pairwise data points automatically; (2) is not sensitive to outliers. Promising experimental results are given for both synthetic and real data sets.     

A novel graph-based clustering method using noise cutting

Recently, many methods have appeared in the field of cluster analysis. Most existing clustering algorithms have considerable limitations in dealing with local and nonlinear data patterns. Algorithms based on graphs provide good results for this problem. However, some widely used graph-based clustering methods, such as spectral clustering algorithms, are sensitive to noise and outliers. In this paper, a cut-point clustering algorithm (CutPC) based on a natural neighbor graph is proposed. The CutPC method performs noise cutting when a cut-point value is above the critical value. Normally, the method can automatically identify clusters with arbitrary shapes and detect outliers without any prior knowledge or preparatory parameter settings. The user can also adjust a coefficient to adapt clustering solutions for particular problems better. Experimental results on various synthetic and real-world datasets demonstrate the obvious superiority of CutPC compared with k-means, DBSCAN, DPC, SC, and DCore.     

HORNLOG: A graph-based interpreter for general Horn clauses

This paper presents hornlog, a general Horn-clause proof procedure that can be used to interpret logic programs. The system is based on a form of graph rewriting, and on the linear-time algorithm for testing the unsatisfiability of propositional Horn formulae given by Dowling and Gallier [8]. hornlog applies to a class of logic programs which is a proper superset of the class of logic programs handled by PROLOG systems. In particular, negative Horn clauses used as assertions and queries consisting of disjunctions of negations of Horn clauses are allowed. This class of logic programs admits answers which are indefinite, in the sense that an answer can consist of a disjunction of substitutions. The method does not use the negation-by- failure semantics [6] in handling these extensions and appears to have an immediate parallel interpretation.     

A rotation based regularization method for semi-supervised learning

Pattern Analysis and Applications - In manifold learning, the intrinsic geometry of the manifold is explored and preserved by identifying the optimal local neighborhood around each observation. It...     

基于k-means和半监督机制的单类中心学习算法

提出了一个基于k means算法框架和半监督机制的single means算法,以解决单类中心学习问题。k means算法实质上是对一种混合高斯模型的期望最大化（EM）算法的近似,对该模型随机生成的多类混合数据集,从目标类中随机标定的初始中心出发,能确定地收敛到该类的实际中心。将single means算法应用到对单类文本中心学习问题中,实验结果表明：在给定目标类中的小标定文本集后,新算法能够有效地改进类的初始中心,且对数据稀疏和方差较大的实际问题具有健壮性。     

Self-paced learning-based multi-graphs semi-supervised learning

Multimedia Tools and Applications - Graph-based semi-supervised learning has received considerable attention in machine learning community. The performance of existing methods highly depends on the...     

半监督学习在链接预测问题中的应用

链接预测是社会网络分析领域的关键问题,研究如何从已知网络中预测可能存在的新链接。现实网络中存在了大量未连接的节点对,从中挖掘潜在信息可以帮助实现链接预测任务。将链接预测视为二类分类问题,使用半监督学习技术,利用网络中的未标记数据帮助学习。使用了两种半监督范式:自我训练和协同训练。在现实数据集Enron和DBLP中的实验结果表明,链接预测任务中采用未标记数据能够有效提高预测的准确率。     

A second order cone programming approach for semi-supervised learning

Semi-supervised learning (SSL) involves the training of a decision rule from both labeled and unlabeled data. In this paper, we propose a novel SSL algorithm based on the multiple clusters per class assumption. The proposed algorithm consists of two stages. In the first stage, we aim to capture the local cluster structure of the training data by using the k-nearest-neighbor (kNN) algorithm to split the data into a number of disjoint subsets. In the second stage, a maximal margin classifier based on the second order cone programming (SOCP) is introduced to learn an inductive decision function from the obtained subsets globally. For linear classification problems, once the kNN algorithm has been performed, the proposed algorithm trains a classifier using only the first and second order moments of the subsets without considering individual data points. Since the number of subsets is usually much smaller than the number of training points, the proposed algorithm is efficient for handling big data sets with a large amount of unlabeled data. Despite its simplicity, the classification performance of the proposed algorithm is guaranteed by the maximal margin classifier. We demonstrate the efficiency and effectiveness of the proposed algorithm on both synthetic and real-world data sets.     

基于半监督学习的多示例多标签改进算法

多示例多标签学习框架是一种针对解决多义性问题而提出的新型机器学习框架,在多示例多标签学习框架中,一个对象是用一组示例集合来表示,并且和一组类别标签相关联。E-MIMLSVM~+算法是多示例多标签学习框架中利用退化思想的经典分类算法,针对其无法利用无标签样本进行学习从而造成泛化能力差等问题,使用半监督支持向量机对该算法进行改进。改进后的算法可以利用少量有标签样本和大量没有标签的样本进行学习,有助于发现样本集内部隐藏的结构信息,了解样本集的真实分布情况。通过对比实验可以看出,改进后的算法有效提高了分类器的泛化性能。     

Interactive genetic algorithms with large population and semi-supervised learning

Interactive genetic algorithms are effective methods of solving optimization problems with implicit (qualitative) criteria by incorporating a user's intelligent evaluation into traditional evolution mechanisms. The heavy evaluation burden of the user, however, is crucial and limits their applications in complex optimization problems. We focus on reducing the evaluation burden by presenting a semi-supervised learning assisted interactive genetic algorithm with large population. In this algorithm, a population with many individuals is adopted to efficiently explore the search space. A surrogate model built with an improved semi-supervised learning method is employed to evaluate a part of individuals instead of the user to alleviate his/her burden in evaluation. Incorporated with the principles of the improved semi-supervised learning, the opportunities of applying and updating the surrogate model are determined by its confidence degree in estimation, and the informative individuals reevaluated by the user are selected according to the concept of learning from mistakes. We quantitatively analyze the performance of the proposed algorithm and apply it to the design of sunglasses lenses, a representative optimization problem with one qualitative criterion. The empirical results demonstrate the strength of our algorithm in searching for satisfactory solutions and easing the evaluation burden of the user.     

Flexible data representation with feature convolution for semi-supervised learning

Applied Intelligence - Data representation plays a crucial role in semi-supervised learning. This paper proposes a framework for semi-supervised data representation. It introduces a flexible...