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.

Bile acid based semi-rigid molecular tweezers

The easily constructed bile acid-based semi-rigid molecular tweezer 2 binds guest 8 in chloroform with an association constant of 83 dm(3) mol(-1).

Robust Semi-Blind Estimation for Beamforming Based MIMO Wireless Communication

In this paper, we present robust semi-blind (SB) algorithms for the estimation of beamforming vectors for multiple-input multiple-output wireless communication. The transmitted symbol block is assumed to comprise of a known sequence of training (pilot) symbols followed by information bearing blind (unknown) data symbols. Analytical expressions are derived for the robust SB estimators of the MIMO receive and transmit beamforming vectors. These robust SB estimators employ a preliminary estimate obtained from the pilot symbol sequence and leverage the second-order statistical information from the blind data symbols. We employ the theory of Lagrangian duality to derive the robust estimate of the receive beamforming vector by maximizing an inner product, while constraining the channel estimate to lie in a confidence sphere centered at the initial pilot estimate. Two different schemes are then proposed for computing the robust estimate of the MIMO transmit beamforming vector. Simulation results presented in the end illustrate the superior performance of the robust SB estimators.

Unsteady flow and heat transfer on a semi-infinite flat plate with an aligned magnetic field

The unsteady laminar boundary layer flow of an electrically conducting fluid past a semi-infinite flat plate with an aligned magnetic field has been studied when at time t > 0 the plate is impulsively moved with a constant velocity which is in the same or opposite direction to that of free stream velocity. The effect of the induced magnetic field has been included in the analysis. The non-linear partial differential equations have been solved numerically using an implicit finite-difference method. The effect of the impulsive motion of the surface is found to be more pronounced on the skin friction but its effect on the x-component of the induced magnetic field and heat transfer is small. Velocity defect occurs near the surface when the plate is impulsively moved in the same direction as that of the free stream velocity. The surface shear stress, x-component of the induced magnetic field on the surface and the surface heat transfer decrease with an increasing magnetic field, but they increase with the reciprocal of the magnetic Prandtl number. However, the effect of the reciprocal of the magnetic Prandtl number is more pronounced on the x-component of the induced magnetic field. (C) 1999 Elsevier Science Ltd. All rights reserved.

An inverse problem for the Helmholtz equation involving two semi-infinite fluids

An analytical method is developed for solving an inverse problem for Helmholtz's equation associated with two semi-infinite incompressible fluids of different variable refractive indices, separated by a plane interface. The unknowns of the inverse problem are: (i) the refractive indices of the two fluids, (ii) the ratio of the densities of the two fluids, and (iii) the strength of an acoustic source assumed to be situated at the interface of the two fluids. These are determined from the pressure on the interface produced by the acoustic source. The effect of the surface tension force at the interface is taken into account in this paper. The application of the proposed analytical method to solve the inverse problem is also illustrated with several examples. In particular, exact solutions of two direct problems are first derived using standard classical methods which are then used in our proposed inverse method to recover the unknowns of the corresponding inverse problems. The results are found to be in excellent agreement.

A fast quasi-Newton method for semi-supervised SVM

Due to its wide applicability, semi-supervised learning is an attractive method for using unlabeled data in classification. In this work, we present a semi-supervised support vector classifier that is designed using quasi-Newton method for nonsmooth convex functions. The proposed algorithm is suitable in dealing with very large number of examples and features. Numerical experiments on various benchmark datasets showed that the proposed algorithm is fast and gives improved generalization performance over the existing methods. Further, a non-linear semi-supervised SVM has been proposed based on a multiple label switching scheme. This non-linear semi-supervised SVM is found to converge faster and it is found to improve generalization performance on several benchmark datasets. (C) 2010 Elsevier Ltd. All rights reserved.

Stream-Close: Fast mining of Closed Frequent Itemsets in high speed data streams

With the emergence of large-volume and high-speed streaming data, the recent techniques for stream mining of CFIpsilas (closed frequent itemsets) will become inefficient. When concept drift occurs at a slow rate in high speed data streams, the rate of change of information across different sliding windows will be negligible. So, the user wonpsilat be devoid of change in information if we slide window by multiple transactions at a time. Therefore, we propose a novel approach for mining CFIpsilas cumulatively by making sliding width(ges1) over high speed data streams. However, it is nontrivial to mine CFIpsilas cumulatively over stream, because such growth may lead to the generation of exponential number of candidates for closure checking. In this study, we develop an efficient algorithm, stream-close, for mining CFIpsilas over stream by exploring some interesting properties. Our performance study reveals that stream-close achieves good scalability and has promising results.

Odonataas Indicators of Riparian Ecosystem Health A Case Study From South Western Karnakata, India

The influence of riparian land use on the diversity an~ distribution were investigated by sampling 113 localities covering 4 districts in south-western Karnataka. A total of 55 species in 12 families were recorded. Streams, rivers and lakes had higher diversity than marshes and sea coast. However, lakes had low endemism than streams and rivers. Streams flowing through evergreen forests had higher diversity and endemism. Human impacted riparian zones such as paddy fields had relatively lower species richness. However, streams flowing through forestry plantations had higher diversity than other natural riparian zones such as dry deciduous, moist deciduous and semi evergreen forests. Myristica swamps-a relict evergreen forest marsh had low diversity and high endemism. Odonate communities of lentic ecosystems, and human impacted streams and rivers were characterized by widespread generalist species. Endemics and habitat specialists were. restricted to streams and rivers with undisturbed riparian zone. The study documents possible odonate community change due to human impact: The influence of riparian 'Ianduse change on odonate community is also discussed.

A fast algorithm for finding frequent episodes in event streams

Frequent episode discovery is a popular framework for mining data available as a long sequence of events. An episode is essentially a short ordered sequence of event types and the frequency of an episode is some suitable measure of how often the episode occurs in the data sequence. Recently,we proposed a new frequency measure for episodes based on the notion of non-overlapped occurrences of episodes in the event sequence, and showed that, such a definition, in addition to yielding computationally efficient algorithms, has some important theoretical properties in connecting frequent episode discovery with HMM learning. This paper presents some new algorithms for frequent episode discovery under this non-overlapped occurrences-based frequency definition. The algorithms presented here are better (by a factor of N, where N denotes the size of episodes being discovered) in terms of both time and space complexities when compared to existing methods for frequent episode discovery. We show through some simulation experiments, that our algorithms are very efficient. The new algorithms presented here have arguably the least possible orders of spaceand time complexities for the task of frequent episode discovery.