Relational Learning with GPUs: Accelerating Rule Coverage

Relational learning algorithms mine complex databases for interesting patterns. Usually, the search space of patterns grows very quickly with the increase in data size, making it impractical to solve important problems. In this work we present the design of a relational learning system, that takes advantage of graphics processing units (GPUs) to perform the most time consuming function of the learner, rule coverage. To evaluate performance, we use four applications: a widely used relational learning benchmark for predicting carcinogenesis in rodents, an application in chemo-informatics, an application in opinion mining, and an application in mining health record data. We compare results using a single and multiple CPUs in a multicore host and using the GPU version. Results show that the GPU version of the learner is up to eight times faster than the best CPU version.     

Self-adaptation and secure information flow in multiparty communications

We present a comprehensive model of structured communications in which self-adaptation and security concerns are jointly addressed. More specifically, we propose a model of multiparty, self-adaptive communications with access control and secure information flow guarantees. In our model, multiparty protocols (choreographies) are described as global types; security violations occur when process implementations of protocol participants attempt to read or write messages of inappropriate security levels within directed exchanges. Such violations trigger adaptation mechanisms that prevent the violations to occur and/or to propagate their effect in the choreography. Our model is equipped with local and global adaptation mechanisms for reacting to security violations of different gravity; type soundness results ensure that the overall multiparty protocol is still correctly executed while the system adapts itself to preserve the participants’ security.     

A PC-based architecture for parameter analysis of vector-controlled induction motor drive

In this article a methodology for constructing a user interface for system control and data acquisition of a drive which is suitable for three-phase induction motors (3?IM) is presented. The entire hardware implementation is shown, including power and digital stages. Communication between the computer and the controller is engaged in order to enhance an analysis of power quality, adjust the controller parameters for tuning the flux and speed loops, and performance of the embedded algorithm. USB and Ethernet protocols have been put into operation in the user front-end because a high speed sample frequency is required in order to guarantee the real-time operation of the whole system. A software interface is developed using LabVIEW environment enabling features not only as filtering signals of phase voltage and current but also power spectrum measurements. Vector-controlled drive is programmed on a digital signal processor (DSP) in order to ensure efficient use of energy in the power stage and proper tracking of the reference at low and high speeds.     

Non-photorealistic, depth-based image editing

Recent works in image editing are opening up new possibilities to manipulate and enhance input images. Within this context, we leverage well-known characteristics of human perception along with a simple depth approximation algorithm to generate non-photorealistic renditions that would be difficult to achieve with existing methods. Once a perceptually plausible depth map is obtained from the input image, we show how simple algorithms yield powerful new depictions of such an image. Additionally, we show how artistic manipulation of depth maps can be used to create novel non-photorealistic versions, for which we provide the user with an intuitive interface. Our real-time implementation on graphics hardware allows the user to efficiently explore artistic possibilities for each image. We show results produced with six different styles proving the versatility of our approach, and validate our assumptions and simplifications by means of a user study.     

Semantic-based authorization architecture for Grid

There are a few issues that still need to be covered regarding security in the Grid area. One of them is authorization where there exist good solutions to define, manage and enforce authorization policies in Grid scenarios. However, these solutions usually do not provide Grid administrators with semantic-aware components closer to the particular Grid domain and easing different administration tasks such as conflict detection or resolution. This paper defines a proposal based on Semantic Web to define, manage and enforce security policies in a Grid scenario. These policies are defined by means of semantic-aware rules which help the administrator to create higher-level definitions with more expressiveness. These rules also permit performing added-value tasks such as conflict detection and resolution, which can be of interest in medium and large scale scenarios where different administrators define the authorization rules that should be followed before accessing a resource in the Grid. The proposed solution has been also tested providing some reasonable response times in the authorization decision process.     

The efforts in the forearm during the use of anti-vibration gloves in simulated work tasks

The objective of this article is to analyze the levels of forearm muscular contraction associated with the use of anti-vibration gloves, in which the contraction levels with gloves and without gloves are compared. Two different vibrating tools - a Bosch Compact Duty Multi-Cutter and a Bosch Rotary Hammer carrying an Ø20 mm bit - were used in a simulated work environment. Standard operations were performed by each subject (n = 14) in an 80 × 40 mm pine bar (cross-sectional cuts with the Multi-Cutter) and on a concrete slab 2 × 2 m, 70 mm thick (20 mm holes with the Rotary Hammer). The forearm muscular efforts were measured by surface electromyography (EMG) in four different muscles: the flexor digitorum superficialis (FDS), flexor carpi ulnaris (FCU), extensor carpi radialis longus (ECRL) and extensor carpi ulnaris (ECU). For the flexor muscles (FDS and FCU), a decrease tendency (p > 0.05) in the measured EMG was observed when the operations are performed with gloves relative to the bare hand (a reduction of 5-23% in the percentage of maximum voluntary exertion (%MVE)). For the extensor muscles (ECU), a tendency toward increased (p > 0.05) muscular contraction was observed when the operations were performed with gloves (an increase of 3-20% in the %MVE). No such tendency was found in the ECRL muscle. The ECU was the muscle with the highest %MVE for 10 and 11 operators (n = 14), during the operations, respectively, with the Multi-Cutter and the Rotary Hammer. As a final conclusion from the study, anti-vibration gloves may increase forearm fatigue in the posterior forearm (ECU muscle) and decrease forearm fatigue in the FDS muscle during operations with the above-mentioned tools.

Relevance to industry

Anti-vibration gloves have been applied in industry to reduce the vibration transmitted into the hand and arms through the palms and fingers. This study analyzed forearm muscular efforts during simulated work tasks performed with two different vibrating tools, operated with the bare hand and with three different gloves, based on the analysis of EMG data.     

Segmentation of ultrasound images of the carotid using RANSAC and cubic splines

A new algorithm is proposed for the semi-automatic segmentation of the near-end and the far-end adventitia boundary of the common carotid artery in ultrasound images. It uses the random sample consensus method to estimate the most significant cubic splines fitting the edge map of a longitudinal section. The consensus of the geometric model (a spline) is evaluated through a new gain function, which integrates the responses to different discriminating features of the carotid boundary: the proximity of the geometric model to any edge or to valley shaped edges; the consistency between the orientation of the normal to the geometric model and the intensity gradient; and the distance to a rough estimate of the lumen boundary.A set of 50 longitudinal B-mode images of the common carotid and their manual segmentations performed by two medical experts were used to assess the performance of the method. The image set was taken from 25 different subjects, most of them having plaques of different classes (class II to class IV), sizes and shapes.The quantitative evaluation showed promising results, having detection errors similar to the ones observed in manual segmentations for 95% of the far-end boundaries and 73% of the near-end boundaries.     

Frequency domain regularization of d-dimensional structure tensor-based directional fields

The present work is intended to address two of the major difficulties that can be found when tackling the estimation of the local orientation of the data in a scene, a task which is usually accomplished by means of the computation of the structure tensor-based directional field. On one hand, the orientation information only exists in the non-homogeneous regions of the dataset, while it is zero in the areas where the gradient (i.e. the first-order intensity variation) remains constant. Due to this lack of information, there are many cases in which the overall shape of the represented objects cannot be precisely inferred from the directional field. On the other hand, the orientation estimation is highly dependent on the particular choice of the averaging window used for its computation (since a collection of neighboring gradient vectors is needed to obtain a dominant orientation), typically resulting in vector fields which vary from very irregular (thus yielding a noisy estimation) to very uniform (but at the expense of a loss of angular resolution). The proposed solution to both drawbacks is the regularization of the directional field; this process extends smoothly the previously computed vectors to the whole dataset while preserving the angular information of relevant structures. With this purpose, the paper introduces a suitable mathematical framework and deals with the d-dimensional variational formulation which is derived from it. The proposed formulation is finally translated into the frequency domain in order to obtain an increase of insight on the regularization problem, which can be understood as a low-pass filtering of the directional field. The frequency domain point of view also allows for an efficient implementation of the resulting iterative algorithm. Simulation experiments involving datasets of different dimensionality prove the validity of the theoretical approach.