Survey of data-selection methods in statistical machine translation

Statistical machine translation has seen significant improvements in quality over the past several years. The single biggest factor in this improvement has been the accumulation of ever larger stores of data. We now find ourselves, however, the victims of our own success, in that it has become increasingly difficult to train on such large sets of data, due to limitations in memory, processing power, and ultimately, speed (i.e. data-to-models takes an inordinate amount of time). Moreover, the training data has a wide quality spectrum. A variety of methods for data cleaning and data selection have been developed to address these issues. Each of these methods employs a search or filtering algorithm to select a subset of the data, given a defined set of feature functions. In this paper we provide a comparative overview of research in this area based on application scenario, feature functions and search method.     

Hydrodynamic behavior of inverse fluidized bed biofilm reactor for phenol biodegradation using Pseudomonas fluorescens

The hydrodynamic characteristic performance of an internal draft tube inverse fluidized bed biofilm reactor was studied for the aerobic biodegradation of phenol (1,200 mg/l) using Pseudomonas fluorescens for various ratios of settled bed volume to reactor working volume (V _b /V _r ) under batchwise condition with respect to liquid phase. The operating parameters, such as superficial gas velocity, phase hold ups, aspect ratio and bed height, were analyzed for different ratios of (V _b /V _r ). The effect of biodegradation on synthetic phenolic effluent was determined from the reduction in chemical oxygen demand and phenol removal efficiency. The optimum value of (V _b /V _r ) _m was 0.20 for the optimal superficial gas velocity, U _gm =0.220 m/s with the COD reduction efficiency of 98.5% in 48 hours. The biomass and biofilm characteristics of P. fluorescens were determined for optimal hydrodynamic operating parameters by evaluating its biofilm dry density and thickness, bioparticle density, suspended and attached biomass concentration.     

Design of a Compressive Sensing Based Fall detection System for Elderly Using WSN

Wireless Personal Communications - Recent researches have pointed out that one third persons are aged 65 and above requires special health care. As the number of elderly person is increasing, home...     

A Distortion-Sensitive Seam Carving Algorithm for Content-Aware Image Resizing

As displays become less expensive and are incorporated into more and more devices, there has been an increased focus on image resizing techniques to fill an image to an arbitrary screen size. Traditional methods such as cropping or resampling can introduce undesirable losses in information or distortion in perception. Recently, content-aware image retargeting methods have been proposed (Avidan and Shamir, ACM Trans Graphics 26(3), 2007; Guo et al., IEEE Trans Multimedia 11(5):856–867, 2009; Shamir and Avidan, Commun ACM 52(1), 2009; Simakov et al. 2008; Wolf et al. 2007), which produce exceptional results. In particular, seam carving, proposed by Avidan and Shamir, has gained attention as an effective solution. However, there are many cases where it can fail. In this paper we propose a distortion-sensitive seam carving algorithm for content-aware image resizing that improves edge preservation and decreases aliasing artifacts. In the proposed approach, we use local gradient information along with a thresholding technique to guide the seam selection process and provide a mechanism to halt seam carving when further processing would introduce unacceptable visual distortion in the resized image. Furthermore, anti-aliasing filter is used to reduce the aliasing artifacts caused by seam removal. Experiments have demonstrated superior performance over the current seam carving methods.     

Dynamic connectivity and path formation time in Poisson networks

The connectivity of wireless networks is commonly analyzed using static geometric graphs. However, with half-duplex radios and due to interference, static or instantaneous connectivity cannot be achieved. It is not necessary, either, since packets take multiple time slots to propagate through the network. For example, if a packet traverses a link in one time slot, it is irrelevant if the next link is available in that time slot also, but it is relevant if the next hop exists in the next time slot. To account for half-duplex constraints and the dynamic changes in the transmitting set of nodes due to MAC scheduling and traffic loads, we introduce a random multi-digraph that captures the evolution of the network connectivity in a dynamic fashion. To obtain concrete results, we focus on Poisson networks, where transmitters form a Poisson point process on the plane at all time instants. We first provide analytical results for the degree distribution of the graph and derive the distributional properties of the end-to-end connection delay using techniques from first-passage percolation and epidemic processes. Next, we prove that under some assumptions, the delay scales linearly with the source–destination distance even in the presence of interference. We also provide simulation results in support of the theoretical results.