Infinity Laplacian on graphs with gradient terms for image and data clustering

The process of merging data of different data sources is referred to as record linkage. A medical environment with increased preconditions on privacy protection demands the transformation of clear-text attributes like first name or date of birth into one-way encrypted pseudonyms. When performing an automated or privacy preserving record linkage there might be the need of a binary classification deciding whether two records should be classified as the same entity. The classification is the final of the four main phases of the record linkage process: Preprocessing, indexing, matching and classification. The choice of binary classification techniques in dependence of project specifications in particular data quality has not extensively been studied yet. The aim of this work is the introduction and evaluation of an automatable semi-supervised binary classification system applied within the field of record linkage capable of competing or even surpassing advanced automated techniques of the domain of unsupervised classification. This work describes the rationale leading to the model and the final implementation of an automatable semi-supervised binary classification system and the comparison of its classification performance to an advanced active learning approach out of the domain of unsupervised learning. The performance of both systems has been measured on a broad variety of artificial test sets (n = 400), based on real patient data, with distinct and unique characteristics. While the classification performance for both methods measured as F-measure was relatively close on test sets with maximum defined data quality, 0.996 for semi-supervised classification, 0.993 for unsupervised classification, it incrementally diverged for test sets of worse data quality dropping to 0.964 for semi-supervised classification and 0.803 for unsupervised classification. Aside from supplying a viable model for semi-supervised classification for automated probabilistic record linkage, the tests conducted on a large amount of test sets suggest that semi-supervised techniques might generally be capable of outperforming unsupervised techniques especially on data with lower levels of data quality.

Sample-weighted fused graph-based semi-supervised learning on multi-view data

Clustering-based segmentation algorithm is one of the common methods in image segmentation. However, how to avoid the influence of the initialization and the singular value is a problem we have to face. For overcoming this problem, we proposed a new image segmentation technology which is based on the AIC criterion and not affected by the singular values. The simulation results show that the method has good performance on stability and accuracy. Image segmentation is the key step in image analysis of computer vision and image processing field. It has been widely applied in various areas such as in the process of automating production, sensing images and biomedical images. The main goal of image segmentation is to simplify an image into segments that have a strong correlation with objects in the real world. The existing segmentation algorithm can be generally divided into three classes: edge detection, region- based segmentation and data clustering. Edge detection method relies on the edges of the object with background and the other objects. For the images with noise, it is difficult to separate the boundary and the target background (1). Region- based segmentation method segments an image into several meaningful sub-domains that are non-overlapping and same- nature. But it may lead to the over-segmentation, and has to combine with other methods to use (2). Data clustering method is based on the whole image and considers the distance between each data. Advantages of data clustering method are low complexity and easy to implement. Disadvantages are sensitive to the noise and selection of the initial centroids (3). Data clustering is one of the common techniques in image segmentation. Its purpose is to cluster pixels into several parts (regions) according to the feature of image. Based on the properties of clustering algorithm, researchers have proposed various image segmentation algorithm, such as based on hierarchical clustering, based on K-means clustering, based on fuzzy clustering, based on mean shift, and so on (3-5). Clustering analysis is an unsupervised process of partitioning a data set into subsets of similar data objects. The elementary principle is as far as possible to increase the difference between categories and reduce the difference within the category. In the process of image segmentation based on data clustering, thus the objective is that the similar pixels are divided into the same regions and the non-similar pixels are divided into different regions. However, Two difficulties we must face are that how to determine the cluster centers and avoid influence of noise. In this paper, a new image segmentation algorithm is proposed based on the AIC criterion. The new segmentation technology can achieve an ideal performance and not affected by the initialization of centroids and the singular values. The simulation results verify the efficiency of proposed algorithm.

Over the years, research on fuzzy clustering algorithms has attracted the attention of many researchers, and they have been applied to various areas, such as image segmentation and data clustering. Various fuzzy clustering algorithms have been put forward based on the initial Fuzzy C-Means clustering (FCM) with Euclidean distance. However, the existing fuzzy clustering approaches ignore two problems. Firstly, clustering algorithms based on Euclidean distance have a high error rate, and are more sensitive to noise and outliers. Secondly, the parameters of the fuzzy clustering algorithms are hard to determine. In practice, they are often determined by the user’s experience, which results in poor performance of the clustering algorithm. Therefore, considering the above deficiencies, this paper proposes a novel fuzzy clustering algorithm by combining the Gaussian kernel function and Grey Wolf Optimizer (GWO), called Kernel-based Picture Fuzzy C-Means clustering with Grey Wolf Optimizer (KPFCM-GWO). In KPFCM-GWO, the Gaussian kernel function is used as a symmetrical measure of distance between data points and cluster centers, and the GWO is utilized to determine the parameter values of PFCM. To verify the validity of KPFCM-GWO, a comparative study was conducted. The experimental results indicate that KPFCM-GWO outperforms other clustering methods, and the improvement of KPFCM-GWO is mainly attributed to the combination of the Gaussian kernel function and the parameter optimization capability of the GWO. What is more, the paper applies KPFCM-GWO to analyzes the value of an airline’s customers, and five levels of customer categories are defined.

Support vector machines (SVMs) have been dominant learning techniques for almost ten years, and mostly applied to supervised learning problems. Recently nice results are obtained by two-class unsupervised and semi-supervised classification algorithms where the optimization problems based on bounded C-SVMs, bounded v-SVMs and Lagrangian SVMs respectively are relaxed to semi-definite programming (SDP). These support vector methods implicitly assume that training data in the optimization problems are known exactly. But in practice, the training data are usually subjected to measurement noise. Zhao et al proposed robust version to unsupervised and semi-supervised classification problems based on Bounded C-SVMs, which need to find the dual problem twice. In this paper we propose unsupervised classification algorithm based on primal problem of standard SVMs with perturbations, which directly relaxes it with label variables to a semi-definite programming. Numerical results confirm the robustness of the proposed method.

In this paper, we propose a unified model called flexible affinity matrix learning (FAML) for unsupervised and semisupervised classification by exploiting both the relationship among data and the clustering structure simultaneously. To capture the relationship among data, we exploit the self-expressiveness property of data to learn a structured matrix in which the structures are induced by different norms. A rank constraint is imposed on the Laplacian matrix of the desired affinity matrix, so that the connected components of data are exactly equal to the cluster number. Thus, the clustering structure is explicit in the learned affinity matrix. By making the estimated affinity matrix approximate the structured matrix during the learning procedure, FAML allows the affinity matrix itself to be adaptively adjusted such that the learned affinity matrix can well capture both the relationship among data and the clustering structure. Thus, FAML has the potential to perform better than other related methods. We derive optimization algorithms to solve the corresponding problems. Extensive unsupervised and semisupervised classification experiments on both synthetic data and real-world benchmark data sets show that the proposed FAML consistently outperforms the state-of-the-art methods.

We consider the problem of deep semi-supervised classification, where label information is obtained in the form of pairwise constraints. Existing approaches to this problem begin with a clustering network and utilize custom loss functions to encourage the learned representations to conform to the obtained constraints. We present a novel framework that seamlessly integrates pairwise constrained clustering, semi-supervised classification, and supervised classification. This approach leverages advances in unsupervised learning by jointly training a Siamese network and autoencoder to learn a representation that is amenable for both clustering and classification. The resulting framework outperforms existing approaches on common image recognition datasets.

CURL: Image Classification using co-training and Unsupervised Representation Learning

Machine learning approaches are valuable methods in hyperspectral remote sensing, especially for the classification of land cover or for the regression of physical parameters. While the recording of hyperspectral data has become affordable with innovative technologies, the acquisition of reference data (ground truth) has remained expensive and time-consuming. There is a need for methodological approaches that can handle datasets with significantly more hyperspectral input data than reference data. We introduce the Supervised Self-organizing Maps (SuSi) framework, which can perform unsupervised, supervised and semi-supervised classification as well as regression on high-dimensional data. The methodology of the SuSi framework is presented and compared to other frameworks. Its different parts are evaluated on two hyperspectral datasets. The results of the evaluations can be summarized in four major findings: (1) The supervised and semi-Supervised Self-organizing Maps (SOM) outperform random forest in the regression of soil moisture. (2) In the classification of land cover, the supervised and semi-supervised SOM reveal great potential. (3) The unsupervised SOM is a valuable tool to understand the data. (4) The SuSi framework is versatile, flexible, and easy to use. The SuSi framework is provided as an open-source Python package on GitHub.

This paper presents local spline regression for semi-supervised classification. The core idea in our approach is to introduce splines developed in Sobolev space to map the data points directly to be class labels. The spline is composed of polynomials and Green's functions. It is smooth, nonlinear, and able to interpolate the scattered data points with high accuracy. Specifically, in each neighborhood, an optimal spline is estimated via regularized least squares regression. With this spline, each of the neighboring data points is mapped to be a class label. Then, the regularized loss is evaluated and further formulated in terms of class label vector. Finally, all of the losses evaluated in local neighborhoods are accumulated together to measure the global consistency on the labeled and unlabeled data. To achieve the goal of semi-supervised classification, an objective function is constructed by combining together the global loss of the local spline regressions and the squared errors of the class labels of the labeled data. In this way, a transductive classification algorithm is developed in which a globally optimal classification can be finally obtained. In the semi-supervised learning setting, the proposed algorithm is analyzed and addressed into the Laplacian regularization framework. Comparative classification experiments on many public data sets and applications to interactive image segmentation and image matting illustrate the validity of our method.

We address the problem of feature weight learning for image clustering. In practice, before clustering data, we generally normalize all data features between 0 and 1, because we cannot determine which features are more important. In this paper, we provide a feature weight learning framework for clustering which can obtain the feature weights and cluster labels simultaneously. An alternative optimization algorithm is adopted to solve this problem. Empirical studies on the toy data and real image data demonstrate our algorithm's effectiveness in improving the clustering performance.

Data clustering is a discovery process that partitions a data set into groups (clusters) such that data points within the same group have high similarity while being very dissimilar to points in other groups (Han & Kamber, 2001). The ultimate goal of data clustering is to discover natural groupings in a set of patterns, points, or objects without prior knowledge of any class labels. In fact, in the machine-learning literature, data clustering is typically regarded as a form of unsupervised learning as opposed to supervised learning. In unsupervised learning or clustering, there is no training function as in supervised learning. There are many applications for data clustering including, but not limited to, pattern recognition, data analysis, data compression, image processing, understanding genomic data, and market-basket research.

For the task of image recognition, it is unpractical to process an image as a whole, since it is highly inefficient and most of the times challenging for the users. Image segmentation techniques are used to divide one image into multiple regions or clusters, it consists in assigning regions to each of the image pixels using different metrics such as pixel colour value, gray-scale intensity, edge finding algorithms, among others. The most diverse image segmentation algorithms have already been proposed and are currently used. However, there are three main categories of segmentation to which the existing algorithms can be assigned and we are going to discuss in this survey: region-based segmentation, data clustering, and edge-based segmentation. With this document, we study diverse commonly used segmentation algorithms, analyse their behaviour as well as their benefits and disadvantages, and present comparative experimental results when applied to a test image.

This paper proposes a novel k-means algorithm for clustering analysis for the cases that the true number of clusters in a data or points set is not known in advance. That is, assuming that the number of seed-points in the algorithm is set to be larger than the true number k of clusters in the data set, the proposed algorithm can assign the k seed-points to the actual clusters, respectively, with the extra seed-points corresponding to the empty clusters, i.e., having no winning points according to a newly defined distance. Via using the Mahalanobis distance, the proposed algorithm can be further extended to elliptical clustering analysis. It is demonstrated well by the experiments on simulated data set and the wine data that the proposed k- means algorithm can find the correct number of clusters in the sample data with a good correct classification rate. Moreover, the algorithm is successfully applied to unsupervised color image segmentation.

Environmental screening of gamma radiation consists of detecting weak nuisance and anomaly signal in the presence of strong and highly varying background. In a typical scenario, a mobile detector-spectrometer continuously measures gamma radiation spectra in short, e.g., one-second, signal acquisition intervals. The measurement data is a 2D matrix, where one dimension is gamma ray energy, and the other dimension is the number of measurements or total time. In principle, gamma radiation sources can be detected and identified from the measured data by their unique spectral lines. Detecting sources from data measured in a search scenario is difficult due to the highly varying background because of naturally occurring radioactive material (NORM), and low signal-to-noise ratio (S/N) of spectral signal measured during one-second acquisition intervals. The objective of this work is to explore unsupervised machine learning (ML) algorithms for development of a digital twin of gamma radiation background, and for detection and identification of weak nuisances and anomalies events in the presence of highly fluctuating background. In one segment of work, we developed a gamma background estimation model using a Longshort term memory (LSTM) network for one-step CPS time series prediction. The LSTM model was validated with two data sets of measurements from two independent NaI detectors positioned on a mobile platform. The data sets contained background radiation only and no orphan isotope sources. The LSTM model was constructed and tested using data from one of the detectors. Performance of the LSTM model was validate through one-step prediction of CPS time series of another NaI detector without re-training. This approach allows to create a digital twin for nuclear background estimation. Using LSTM, it could be possible to detect a source through subtraction of the estimated counts from the measured background. In another segment of work, we investigated detection of gamma emitting sources in the presence of complex background using unsupervised machine learning. Spectral lines of isotopes are difficult to observe in one-second measurements. Averaging over the entire measurement campaign data set reveals spectral lines of most common background isotopes. Spectral lines of orphan sources, which might appear only in a few measurements during the campaign, will be washed out if averaging is performed over the entire measurement data set. The approach we have explored consists of extracting one-second measurements containing weak spectral features through data clustering. Averaging one-second spectra in a cluster should reveal the presence of anomaly sources. We created two ML models using K-means clustering and Neural Network Self-organizing Map (SOM). Performance of these ML models was benchmarked using search data. One data set contained 137Cs source, and another dataset contained 131I source.