No abstract

Multi-view clustering aims to achieve more promising clustering results than single-view clustering by exploring the multi-view information. Since statistic properties of different views are diverse, even incompatible, few approaches implement multi-view clustering based on the concatenated features directly. However, feature concatenation is a natural way to combine multiple views. To this end, this paper proposes a novel multi-view subspace clustering approach dubbed Feature Concatenation Multi-view Subspace Clustering (FCMSC). Specifically, by exploring the consensus information, multi-view data are concatenated into a joint representation firstly, then, l2,1-norm is integrated into the objective function to deal with the sample-specific and cluster-specific corruptions of multiple views for benefiting the clustering performance. Furthermore, by introducing graph Laplacians of multiple views, a graph regularized FCMSC is also introduced to explore both the consensus information and complementary information for clustering. It is noteworthy that the obtained coefficient matrix is not derived by directly applying the Low-Rank Representation (LRR) to the joint view representation simply. Finally, an effective algorithm based on the Augmented Lagrangian Multiplier (ALM) is designed to optimized the objective functions. Comprehensive experiments on six real world datasets illustrate the superiority of the proposed methods over several state-of-the-art approaches for multi-view clustering.

Abstract:The problem of linear systems subject to actuator faults (outage, loss of effectiveness and stuck), parameter uncertainties and external disturbances is considered. An active fault compensation control law is designed which utilizes compensation in such a way that uncertainties, disturbances and the occurrence of actuator faults are account for. The main idea is designing a robust adaptive output feedback controller by automatically compensating the fault dynamics to render the close-loop stability. According to the information from the adaptive mechanism, the updating control law is derived such that all the parameters of the unknown input signal are bounded. Furthermore, a disturbance decoupled fault reconstruction scheme is presented to evaluate the severity of the fault and to indicate how fault accommodation should be implemented. The advantage of fault compensation is that the dynamics caused by faults can be accommodated online. The proposed design method is illustrated on a rocket fairing structural-acoustic model.

This study simulated the effects of changes in the underlying surface induced by long‐term urbanization on trends in surface air temperature (SAT) over three extensive urban agglomerations (Beijing‐Tianjin‐Hebei, BTH; the Yangtze River Delta, YRD; and the Pearl River Delta, PRD) in China during 1980–2009. To isolate the effects of continuous urban expansion on SAT with the least computation cost, we employed the Community Land Model (CLM4.5) in an off‐line mode for a relatively long period. Based on a high‐quality land use data set dating back to the 1980s, two scenarios were designed to represent the distributions of both nonurban and historically urban land use. By comparing the results of two numerical experiments, urban‐induced warming in daily mean SAT (Tmean) over the three urban agglomerations, BTH, YRD, and PRD, were found to be 0.13°C/30 yrs, 0.12°C/30 yrs, and 0.09°C/30 yrs, contributing about 9.70%, 10.3%, and 9.68% to the mean long‐term SAT trends, respectively. In addition, a higher contribution of urban‐related warming was found in winter for BTH and in summer for the other two regions. However, urban‐related warming had no significant effect on the trends of daily maximum SAT (Tmax) when compared with daily minimum SAT (Tmin). Specifically, at a local scale, the contributions of urban warming to the background warming in three representative cities, Beijing, Shanghai, and Guangzhou, were 12.7%, 29.0%, and 23.6%, respectively.

Subspace clustering (SC) is a promising technology involving clusters that are identified based on their association with subspaces in high-dimensional spaces. SC can be classified into hard subspace clustering (HSC) and soft subspace clustering (SSC). While HSC algorithms have been studied extensively and are well accepted by the scientific community, SSC algorithms are relatively new. However, as they are said to be more adaptable than their HSC counterparts, SSC algorithms have been attracting more attention in recent years. A comprehensive survey of existing SSC algorithms and recent developments in the field are presented in this paper. SSC algorithms have been systematically classified into three main categories: conventional SSC (CSSC), independent SSC (ISSC), and extended SSC (XSSC). The characteristics of these algorithms are highlighted and potential future developments in the area of SSC are discussed. Through a comprehensive review of SSC, this paper aims to provide readers with a clear profile of existing SSC methods and to foster the development of more effective clustering technologies and significant research in this area.As discussed previously, SSC algorithms can be broadly classified into three main categories: CSSC, ISSC, and XSSC. Each of these categories can be further divided into subcategories based on the clustering mechanisms that are adopted, as shown in Table 3. In CSSC, clustering is performed by first identifying the subspace using some strategies, and then carrying out clustering in the subspace that was obtained, in order to partition the data. This is referred to as separated feature weighting, where data partitioning involves two separate processessubspace identification and clustering in subspace. Clustering can also be conducted by performing the two processes simultaneously, an approach known as coupled feature weighting. In ISSC, algorithms are developed based on the K-means model, fuzzy C-means (FCM) model, and probability mixture model, in a process where fuzzy weighting, entropy weighting, or other weighting mechanisms are adopted to implement feature weighting. Finally, XSSC algorithms can be subdivided into eight subcategories, depending on the strategies used to enhance the CSSC and ISSC algorithms. These subcategories are between-class separation, evolutionary learning, the adoption of new metrics, ensemble learning, multi-view learning, imbalanced * Denotes the values achieved by each algorithm when the lowest value of the loss function is obtained within the 10 runs.

The existing, semisupervised, spectral clustering approaches have two major drawbacks, i.e., either they cannot cope with multiple categories of supervision or they sometimes exhibit unstable effectiveness. To address these issues, two normalized affinity and penalty jointly constrained spectral clustering frameworks as well as their corresponding algorithms, referred to as type-I affinity and penalty jointly constrained spectral clustering (TI-APJCSC) and type-II affinity and penalty jointly constrained spectral clustering (TII-APJCSC), respectively, are proposed in this paper. TI refers to type-I and TII to type-II. The significance of this paper is fourfold. First, benefiting from the distinctive affinity and penalty jointly constrained strategies, both TI-APJCSC and TII-APJCSC are substantially more effective than the existing methods. Second, both TI-APJCSC and TII-APJCSC are fully compatible with the three well-known categories of supervision, i.e., class labels, pairwise constraints, and grouping information. Third, owing to the delicate framework normalization, both TI-APJCSC and TII-APJCSC are quite flexible. With a simple tradeoff factor varying in the small fixed interval (0, 1], they can self-adapt to any semisupervised scenario. Finally, both TI-APJCSC and TII-APJCSC demonstrate strong robustness, not only to the number of pairwise constraints but also to the parameter for affinity measurement. As such, the novel TI-APJCSC and TII-APJCSC algorithms are very practical for medium- and small-scale semisupervised data sets. The experimental studies thoroughly evaluated and demonstrated these advantages on both synthetic and real-life semisupervised data sets.

No abstract

Clustering with multiview data is becoming a hot topic in data mining, pattern recognition, and machine learning. In order to realize an effective multiview clustering, two issues must be addressed, namely, how to combine the clustering result from each view and how to identify the importance of each view. In this paper, based on a newly proposed objective function which explicitly incorporates two penalty terms, a basic multiview fuzzy clustering algorithm, called collaborative fuzzy c-means (Co-FCM), is firstly proposed. It is then extended into its weighted view version, called weighted view collaborative fuzzy c-means (WV-Co-FCM), by identifying the importance of each view. The WV-Co-FCM algorithm indeed tackles the above two issues simultaneously. Its relationship with the latest multiview fuzzy clustering algorithm Collaborative Fuzzy K-Means (Co-FKM) is also revealed. Extensive experimental results on various multiview datasets indicate that the proposed WV-Co-FCM algorithm outperforms or is at least comparable to the existing state-of-the-art multitask and multiview clustering algorithms and the importance of different views of the datasets can be effectively identified.