Kevin S. Xu

Kevin S. Xu
Are you Kevin S. Xu?

Claim your profile, edit publications, add additional information:

Contact Details

Name
Kevin S. Xu
Affiliation
Location

Pubs By Year

Pub Categories

 
Computer Science - Learning (10)
 
Physics - Physics and Society (7)
 
Statistics - Machine Learning (6)
 
Statistics - Methodology (5)
 
Computer Science - Computer Vision and Pattern Recognition (2)
 
Cosmology and Nongalactic Astrophysics (2)
 
Astrophysics (2)
 
Instrumentation and Methods for Astrophysics (2)
 
Astrophysics of Galaxies (2)
 
Statistics - Applications (2)
 
Computer Science - Data Structures and Algorithms (1)
 
Computer Science - Distributed; Parallel; and Cluster Computing (1)
 
Statistics - Computation (1)
 
Computer Science - Databases (1)
 
Computer Science - Computers and Society (1)
 
Mathematics - Statistics (1)
 
Statistics - Theory (1)
 
Computer Science - Discrete Mathematics (1)

Publications Authored By Kevin S. Xu

A common problem in large-scale data analysis is to approximate a matrix using a combination of specifically sampled rows and columns, known as CUR decomposition. Unfortunately, in many real-world environments, the ability to sample specific individual rows or columns of the matrix is limited by either system constraints or cost. In this paper, we consider matrix approximation by sampling predefined blocks of columns (or rows) from the matrix. Read More

The task of predicting future relationships in a social network, known as link prediction, has been studied extensively in the literature. Many link prediction methods have been proposed, ranging from common neighbors to probabilistic models. Recent work by Yang et al. Read More

The measurement and analysis of Electrodermal Activity (EDA) offers applications in diverse areas ranging from market research, to seizure detection, to human stress analysis. Unfortunately, the analysis of EDA signals is made difficult by the superposition of numerous components which can obscure the signal information related to a user's response to a stimulus. We show how simple pre-processing followed by a novel compressed sensing based decomposition can mitigate the effects of the undesired noise components and help reveal the underlying physiological signal. Read More

The Herschel Space Observatory was the fourth cornerstone mission in the European Space Agency (ESA) science programme with excellent broad band imaging capabilities in the sub-mm and far-infrared part of the spectrum. Although the spacecraft finished its observations in 2013, it left a large legacy dataset that is far from having been fully scrutinised and still has a large potential for new scientific discoveries. This is specifically true for the photometric observations of the PACS and SPIRE instruments. Read More

We consider the problem of identifying patterns in a data set that exhibit anomalous behavior, often referred to as anomaly detection. Similarity-based anomaly detection algorithms detect abnormally large amounts of similarity or dissimilarity, e.g. Read More

There has been great interest in recent years on statistical models for dynamic networks. In this paper, I propose a stochastic block transition model (SBTM) for dynamic networks that is inspired by the well-known stochastic block model (SBM) for static networks and previous dynamic extensions of the SBM. Unlike most existing dynamic network models, it does not make a hidden Markov assumption on the edge-level dynamics, allowing the presence or absence of edges to directly influence future edge probabilities while retaining the interpretability of the SBM. Read More

Significant progress has been made recently on theoretical analysis of estimators for the stochastic block model (SBM). In this paper, we consider the multi-graph SBM, which serves as a foundation for many application settings including dynamic and multi-layer networks. We explore the asymptotic properties of two estimators for the multi-graph SBM, namely spectral clustering and the maximum-likelihood estimate (MLE), as the number of layers of the multi-graph increases. Read More

Significant efforts have gone into the development of statistical models for analyzing data in the form of networks, such as social networks. Most existing work has focused on modeling static networks, which represent either a single time snapshot or an aggregate view over time. There has been recent interest in statistical modeling of dynamic networks, which are observed at multiple points in time and offer a richer representation of many complex phenomena. Read More

The different algorithms appropriate for point source photometry on data from the SPIRE instrument on-board the Herschel Space Observatory, within the Herschel Interactive Processing Environment (HIPE) are compared. Point source photometry of a large ensemble of standard calibration stars and dark sky observations is carried out using the 4 major methods within HIPE: SUSSEXtractor, DAOphot, the SPIRE Timeline Fitter and simple Aperture Photometry. Colour corrections and effective beam areas as a function of the assumed source spectral index are also included to produce a large number of photometric measurements per individual target, in each of the 3 SPIRE bands (250, 350, 500um), to examine both the accuracy and repeatability of each of the 4 algorithms. Read More

The ability to predict social interactions between people has profound applications including targeted marketing and prediction of information diffusion and disease propagation. Previous work has shown that the location of an individual at any given time is highly predictable. This study examines the predictability of social interactions between people to determine whether interaction patterns are similarly predictable. Read More

To date, most studies on spam have focused only on the spamming phase of the spam cycle and have ignored the harvesting phase, which consists of the mass acquisition of email addresses. It has been observed that spammers conceal their identity to a lesser degree in the harvesting phase, so it may be possible to gain new insights into spammers' behavior by studying the behavior of harvesters, which are individuals or bots that collect email addresses. In this paper, we reveal social networks of spammers by identifying communities of harvesters with high behavioral similarity using spectral clustering. Read More

Significant efforts have gone into the development of statistical models for analyzing data in the form of networks, such as social networks. Most existing work has focused on modeling static networks, which represent either a single time snapshot or an aggregate view over time. There has been recent interest in statistical modeling of dynamic networks, which are observed at multiple points in time and offer a richer representation of many complex phenomena. Read More

Many real-world networks, including social and information networks, are dynamic structures that evolve over time. Such dynamic networks are typically visualized using a sequence of static graph layouts. In addition to providing a visual representation of the network structure at each time step, the sequence should preserve the mental map between layouts of consecutive time steps to allow a human to interpret the temporal evolution of the network. Read More

We report on single-dish radio CO observations towards the inter-galactic medium (IGM) of the Stephan's Quintet (SQ) group of galaxies. Extremely bright mid-IR H2 rotational line emission from warm molecular gas has been detected by Spitzer in the kpc-scale shock created by a galaxy collision. We detect in the IGM CO(1-0), (2-1) and (3-2) line emission with complex profiles, spanning a velocity range of 1000 km/s. Read More

We consider the problem of identifying patterns in a data set that exhibit anomalous behavior, often referred to as anomaly detection. In most anomaly detection algorithms, the dissimilarity between data samples is calculated by a single criterion, such as Euclidean distance. However, in many cases there may not exist a single dissimilarity measure that captures all possible anomalous patterns. Read More

In many practical applications of clustering, the objects to be clustered evolve over time, and a clustering result is desired at each time step. In such applications, evolutionary clustering typically outperforms traditional static clustering by producing clustering results that reflect long-term trends while being robust to short-term variations. Several evolutionary clustering algorithms have recently been proposed, often by adding a temporal smoothness penalty to the cost function of a static clustering method. Read More

The Great Observatories All-sky LIRG Survey (GOALS) consists of a complete sample of 202 Luminous Infrared Galaxies (LIRGs) selected from the IRAS Revised Bright Galaxy Sample (RBGS). The galaxies span the full range of interaction stages, from isolated galaxies to interacting pairs to late stage mergers. We present a comparison of the UV and infrared properties of 135 galaxies in GOALS observed by GALEX and Spitzer. Read More

We present a study of the nearby Seyfert galaxy NGC 1068 using mid- and far- infrared data acquired with the IRAC, IRS, and MIPS instruments aboard the Spitzer Space Telescope. The images show extensive 8 um and 24 um emission coinciding with star formation in the inner spiral approximately 15" (1 kpc) from the nucleus, and a bright complex of star formation 47" (3 kpc) SW of the nucleus. The brightest 8 um PAH emission regions coincide remarkably well with knots observed in an Halpha image. Read More

We discuss spectral energy distributions, photometric redshifts, redshift distributions, luminosity functions, source-counts and the far infrared to optical luminosity ratio for sources in the SWIRE Legacy Survey. The spectral energy distributions of selected SWIRE sources are modelled in terms of a simple set of galaxy and quasar templates in the optical and near infrared, and with a set of dust emission templates (cirrus, M82 starburst, Arp 220 starburst, and AGN dust torus) in the mid infrared. The optical data, together with the IRAC 3. Read More