Semi-Supervised Classification Using Sparse Gaussian Process Regression

Gaussian Processes (GPs) are promising Bayesian methods for classification and regression problems. They have also been used for semi-supervised learning tasks. In this paper, we propose a new algorithm for solving semi-supervised binary classification problem using sparse GP regression (GPR) models. It is closely related to semi-supervised learning based on support vector regression (SVR) and maximum margin clustering. The proposed algorithm is simple and easy to implement. It gives a sparse solution directly unlike the SVR based algorithm. Also, the hyperparameters are estimated easily without resorting to expensive cross-validation technique. Use of sparse GPR model helps in making the proposed algorithm scalable. Preliminary results on synthetic and real-world data sets demonstrate the efficacy of the new algorithm.

A new classification and linear sequence of extant gymnosperms

A new classification and linear sequence of the gymnosperms based on previous molecular and morphological phylogenetic and other studies is presented. Currently accepted genera are listed for each family and arranged according to their (probable) phylogenetic position. A full synonymy is provided, and types are listed for accepted genera. An index to genera assists in easy access to synonymy and family placement of genera.

Functional outcome and health-related quality of life after traumatic brain injury in the framework of the International Classification of Functioning, Disability and Health (ICF)

Traumatic brain injury (TBI) affects people of all ages and is a cause of long-term disability. In recent years, the epidemiological patterns of TBI have been changing. TBI is a heterogeneous disorder with different forms of presentation and highly individual outcome regarding functioning and health-related quality of life (HRQoL). The meaning of disability differs from person to person based on the individual s personality, value system, past experience, and the purpose he or she sees in life. Understanding of all these viewpoints is needed in comprehensive rehabilitation. This study examines the epidemiology of TBI in Finland as well as functioning and HRQoL after TBI, and compares the subjective and objective assessments of outcome. The frame of reference is the International Classification of Functioning, Disability and Health (ICF). The subjects of Study I represent the population of Finnish TBI patients who experienced their first TBI between 1991 and 2005. The 55 Finnish subjects of Studies II and IV participated in the first wave of the international Quality of life after brain injury (QOLIBRI) validation study. The 795 subjects from six language areas of Study III formed the second wave of the QOLIBRI validation study. The average annual incidence of Finnish hospitalised TBI patients during the years 1991-2005 was 101:100 000 in patients who had TBI as the primary diagnosis and did not have a previous TBI in their medical history. Males (59.2%) were at considerably higher risk of getting a TBI than females. The most common external cause of the injury was falls in all age groups. The number of TBI patients ≥ 70 years of age increased by 59.4% while the number of inhabitants older than 70 years increased by 30.3% in the population of Finland during the same time period. The functioning of a sample of 55 persons with TBI was assessed by extracting information from the patients medical documents using the ICF checklist. The most common problems were found in the ICF components of Body Functions (b) and Activities and Participation (d). HRQoL was assessed with the QOLIBRI which showed the highest level of satisfaction on the Emotions, Physical Problems and Daily Life and Autonomy scales. The highest scores were obtained by the youngest participants and participants living independently without the help of other people, and by people who were working. The relationship between the functional outcome and HRQoL was not straightforward. The procedure of linking the QOLIBRI and the GOSE to the ICF showed that these two outcome measures cover the relevant domains of TBI patients functioning. The QOLIBRI provides the patients subjective view, while the GOSE summarises the objective elements of functioning. Our study indicates that there are certain domains of functioning that are not traditionally sufficiently documented but are important for the HRQoL of persons with TBI. This was the finding especially in the domains of interpersonal relationships, social and leisure activities, self, and the environment. Rehabilitation aims to optimize functioning and to minimize the experience of disability among people with health conditions, and it needs to be based on a comprehensive understanding of human functioning. As an integrative model, the ICF may serve as a frame of reference in achieving such an understanding.

Sediment Volume Tests for Expansive Soil Identification and Classification

Equilibrium sediment volume tests are conducted on field soils to classify them based on their degree of expansivity and/or to predict the liquid limit of soils. The present technical paper examines different equilibrium sediment volume tests, critically evaluating each of them. It discusses the settling behavior of fine-grained soils during the soil sediment formation to evolve a rationale for conducting the latest version of equilibrium sediment volume test. Probable limitations of equilibrium sediment volume test and the possible solution to overcome the same have also been indicated.

A classification of homogeneous operators in the Cowen-Douglas class

An explicit construction of all the homogeneous holomorphic Hermitian vector bundles over the unit disc D is given. It is shown that every such vector bundle is a direct sum of irreducible ones. Among these irreducible homogeneous holomorphic Hermitian vector bundles over D, the ones corresponding to operators in the Cowen-Douglas class B-n(D) are identified. The classification of homogeneous operators in B-n(D) is completed using an explicit realization of these operators. We also show how the homogeneous operators in B-n(D) split into similarity classes. (C) 2011 Elsevier Inc. All rights reserved.

Chance constrained uncertain classification via robust optimization

This paper studies the problem of constructing robust classifiers when the training is plagued with uncertainty. The problem is posed as a Chance-Constrained Program (CCP) which ensures that the uncertain data points are classified correctly with high probability. Unfortunately such a CCP turns out to be intractable. The key novelty is in employing Bernstein bounding schemes to relax the CCP as a convex second order cone program whose solution is guaranteed to satisfy the probabilistic constraint. Prior to this work, only the Chebyshev based relaxations were exploited in learning algorithms. Bernstein bounds employ richer partial information and hence can be far less conservative than Chebyshev bounds. Due to this efficient modeling of uncertainty, the resulting classifiers achieve higher classification margins and hence better generalization. Methodologies for classifying uncertain test data points and error measures for evaluating classifiers robust to uncertain data are discussed. Experimental results on synthetic and real-world datasets show that the proposed classifiers are better equipped to handle data uncertainty and outperform state-of-the-art in many cases.

Combining Bayes Classification and Point Group Symmetry under Boolean Framework for Enhanced Protein Quaternary Structure Inference

Our ability to infer the protein quaternary structure automatically from atom and lattice information is inadequate, especially for weak complexes, and heteromeric quaternary structures. Several approaches exist, but they have limited performance. Here, we present a new scheme to infer protein quaternary structure from lattice and protein information, with all-around coverage for strong, weak and very weak affinity homomeric and heteromeric complexes. The scheme combines naive Bayes classifier and point group symmetry under Boolean framework to detect quaternary structures in crystal lattice. It consistently produces >= 90% coverage across diverse benchmarking data sets, including a notably superior 95% coverage for recognition heteromeric complexes, compared with 53% on the same data set by current state-of-the-art method. The detailed study of a limited number of prediction-failed cases offers interesting insights into the intriguing nature of protein contacts in lattice. The findings have implications for accurate inference of quaternary states of proteins, especially weak affinity complexes.

Two-mode quantum systems: Invariant classification of squeezing transformations and squeezed states

A general analysis of squeezing transformations for two-mode systems is given based on the four-dimensional real symplectic group Sp(4, R). Within the framework of the unitary (metaplectic) representation of this group, a distinction between compact photon-number-conserving and noncompact photon-number-nonconserving squeezing transformations is made. We exploit the U(2) invariant squeezing criterion to divide the set of all squeezing transformations into a two-parameter family of distinct equivalence classes with representative elements chosen for each class. Familiar two-mode squeezing transformations in the literature are recognized in our framework and seen to form a set of measure zero. Examples of squeezed coherent and thermal states are worked out. The need to extend the heterodyne detection scheme to encompass all of U(2) is emphasized, and known experimental situations where all U(2) elements can be reproduced are briefly described.

Classification of microwatersheds based on morphological characteristics

Three classification techniques, namely, K-means Cluster Analysis (KCA), Fuzzy Cluster Analysis (FCA), and Kohonen Neural Networks (KNN) were employed to group 25 microwatersheds of Kherthal watershed, Rajasthan into homogeneous groups for formulating the basis for suitable conservation and management practices. Ten parameters, mainly, morphological, namely, drainage density (D-d), bifurcation ratio (R-b), stream frequency (F-u), length of overland flow (L-o), form factor (R-f), shape factor (B-s), elongation ratio (R-e), circulatory ratio (R-c), compactness coefficient (C-c) and texture ratio (T) are used for the classification. Optimal number of groups is chosen, based on two cluster validation indices Davies-Bouldin and Dunn's. Comparative analysis of various clustering techniques revealed that 13 microwatersheds out of 25 are commonly suggested by KCA, FCA and KNN i.e., 52%; 17 microwatersheds out of 25 i.e., 68% are commonly suggested by KCA and FCA whereas these are 16 out of 25 in FCA and KNN (64%) and 15 out of 25 in KNN and CA (60%). It is observed from KNN sensitivity analysis that effect of various number of epochs (1000, 3000, 5000) and learning rates (0.01, 0.1-0.9) on total squared error values is significant even though no fixed trend is observed. Sensitivity analysis studies revealed that microwatershecls have occupied all the groups even though their number in each group is different in case of further increase in the number of groups from 5 to 6, 7 and 8. (C) 2010 International Association of Hydro-environment Engineering and Research, Asia Pacific Division. Published by Elsevier B.V. All rights reserved.

Video coding mode decision as a classification problem

In this paper, we show that it is possible to reduce the complexity of Intra MB coding in H.264/AVC based on a novel chance constrained classifier. Using the pairs of simple mean-variances values, our technique is able to reduce the complexity of Intra MB coding process with a negligible loss in PSNR. We present an alternate approach to address the classification problem which is equivalent to machine learning. Implementation results show that the proposed method reduces encoding time to about 20% of the reference implementation with average loss of 0.05 dB in PSNR.

Solving a part classification problem using simulated annealing-like hybrid

Part classification and coding is still considered as laborious and time-consuming exercise. Keeping in view, the crucial role, which it plays, in developing automated CAPP systems, the attempts have been made in this article to automate a few elements of this exercise using a shape analysis model. In this study, a 24-vector directional template is contemplated to represent the feature elements of the parts (candidate and prototype). Various transformation processes such as deformation, straightening, bypassing, insertion and deletion are embedded in the proposed simulated annealing (SA)-like hybrid algorithm to match the candidate part with their prototype. For a candidate part, searching its matching prototype from the information data is computationally expensive and requires large search space. However, the proposed SA-like hybrid algorithm for solving the part classification problem considerably minimizes the search space and ensures early convergence of the solution. The application of the proposed approach is illustrated by an example part. The proposed approach is applied for the classification of 100 candidate parts and their prototypes to demonstrate the effectiveness of the algorithm. (C) 2003 Elsevier Science Ltd. All rights reserved.

Soft classification based Sub-pixel allocation model

Land cover (LC) refers to what is actually present on the ground and provide insights into the underlying solution for improving the conditions of many issues, from water pollution to sustainable economic development. One of the greatest challenges of modeling LC changes using remotely sensed (RS) data is of scale-resolution mismatch: that the spatial resolution of detail is less than what is required, and that this sub-pixel level heterogeneity is important but not readily knowable. However, many pixels consist of a mixture of multiple classes. The solution to mixed pixel problem typically centers on soft classification techniques that are used to estimate the proportion of a certain class within each pixel. However, the spatial distribution of these class components within the pixel remains unknown. This study investigates Orthogonal Subspace Projection - an unmixing technique and uses pixel-swapping algorithm for predicting the spatial distribution of LC at sub-pixel resolution. Both the algorithms are applied on many simulated and actual satellite images for validation. The accuracy on the simulated images is ~100%, while IRS LISS-III and MODIS data show accuracy of 76.6% and 73.02% respectively. This demonstrates the relevance of these techniques for applications such as urban-nonurban, forest-nonforest classification studies etc.