Predicting potential deadlocks in multithreaded programs

In a multithreaded program, competition of threads for shared resources raises the deadlock possibility, which narrows the system liveness. Because such errors appear in specific schedules of concurrent executions of threads, runtime verification of threads behavior is a significant concern. In this study, we extended our previous approach for prediction of runtime behavior of threads may lead to an impasse. Such a prediction is of importance because of the nondeterministic manner of competing threads. The prediction process tries to forecast future behavior of threads based on their observed behavior. To this end, we map observed behavior of threads into time‐series data sets and use statistical and artificial intelligence methods for forecasting subsequent members of the sets as future behavior of the threads. The deadlock prediction is carried out based on probing the allocation graph obtained from actual and predicted allocation of resources to threads. In our approach, we use an artificial neural network (ANN) because ANNs enjoy the applicable performance and flexibility in predicting complex behavior. Using three case studies, we contrasted results of the current and our previous approaches to demonstrate results. Copyright © 2015 John Wiley & Sons, Ltd.     

Accelerating video encoding using cluster computing

The great advance and variety of multimedia applications such as video streaming, TV broadcasting, and video conferencing stimulated research to enhance video encoding, where a video is reduced in size and possibly transformed to numerous formats for portability. This paper is concerned with solving the problem of the huge processing time taken by the serial video encoding approaches by proposing a hybrid-parallel video encoding technique to speed up the process. In this work, the Joint Scalable Video Model (JSVM 9.19.14) is chosen as the basic serial video encoding algorithm for building different parallel video encoding architectures. The proposed technique exploits the triple-step nature of JSVM and intelligently determines the best task organization to achieve speedup and increase the efficiency on a cluster computing platform. Moreover, a dynamic load sharing scheme is proposed to redistribute load among different machines for additional parallelism. The remarkable feature of our approach is that, both the granularity of load partitioning among the cluster machines and all the associated overheads are considered. The experimental results are applied on a compact library of 160 mp4 encoded videos and two other bench mark datasets. The results proves a significant improvement in performance in comparison to the sequential version; which ranges from 64.2% to 95.3%, for a cluster with a number of machines ranging from 2 to 20 respectively.

     

Robust multiscale stereo matching from fundus images with radiometric differences

A robust multiscale stereo matching algorithm is proposed to find reliable correspondences between low contrast and weakly textured retinal image pairs with radiometric differences. Existing algorithms designed to deal with piecewise planar surfaces with distinct features and Lambertian reflectance do not apply in applications such as 3D reconstruction of medical images including stereo retinal images. In this paper, robust pixel feature vectors are formulated to extract discriminative features in the presence of noise in scale space, through which the response of low-frequency mechanisms alter and interact with the response of high-frequency mechanisms. The deep structures of the scene are represented with the evolution of disparity estimates in scale space, which distributes the matching ambiguity along the scale dimension to obtain globally coherent reconstructions. The performance is verified both qualitatively by face validity and quantitatively on our collection of stereo fundus image sets with ground truth, which have been made publicly available as an extension of standard test images for performance evaluation.     

Assessing information quality of e-learning systems: a web mining approach

E-learning systems provide a promising solution as an information exchanging channel. Improved technologies could mean faster and easier access to information but do not necessarily ensure the quality of this information; for this reason it is essential to develop valid and reliable methods of quality measurement and carry out careful information quality evaluations. This paper proposes an assessment model for information quality in e-learning systems based on the quality framework we proposed previously: the proposed framework consists of 14 quality dimensions grouped in three quality factors: intrinsic, contextual representation and accessibility. We use the relative importance as a parameter in a linear equation for the measurement scheme. Formerly, we implemented a goal-question-metrics approach to develop a set of quality metrics for the identified quality attributes within the proposed framework. In this paper, the proposed metrics were computed to produce a numerical rating indicating the overall information quality published in a particular e-learning system. The data collection and evaluation processes were automated using a web data extraction technique and results on a case study are discussed. This assessment model could be useful to e-learning systems designers, providers and users as it provides a comprehensive indication of the quality of information in such systems.     

Incremental one-class classifier based on convex–concave hull

Pattern Analysis and Applications - One subject that has been considered less is a binary classification on data streams with concept drifting in which only information of one class (target class)...     

Network politics in an educational organization

The electronic mail (EM) network in a large, multi-campus community college district was used by some employees to gain support for positions contrary to those of the leadership. The case study offered an opportunity to look at technology within an organizational setting. It raised questions about the loose coupling of the educational organization's technical and authority systems, about the strength of coupling among employee groups during the incident, about the boundaries of the EM political activists, and about the power manifested within educational organization's technical and authority systems. A variety of research methods (stages of event progression, fantasy types associated with consciousness-building, and evidence of user technical and rhetorical skills) were used to answer questions about the organization's loosely coupled systems during the EM political incident. Notes were taken of conversations and more formal interviews. From this the technical and authority systems of the institution were described and compared. Results indicated that (1) the loose coupling of the technical and the authority systems made the EM political incident possible; (2) employees were more tightly coupled on organizational goals and more loosely coupled on organizational means; (3) political activists did not make full use of the EM's political medium potential; and (4) when the college district's administration refused to limit anyone's use of the EM network, they reinforced the integrity of both the authority and the technical systems.

Implications included: (1) the potential of some of the research methods for EM study, especially fantasy theme analysis; (2) a political interpretation of EM language, especially flaming; (3) the importance of technical and rhetorical skills for mature EM users; and (4) the role of the authority and technical systems in the debate about appropriate EM network use within an organization.     

Nanocrystalline Fe‐Ni alloy/polyamide 6 composites of high mechanical performance made by ultrasound‐assisted master batch technique

The focus of this study is to improve the dispersion state of nanocrystalline (nc) Fe‐Ni particles in polyamide 6 (PA6) matrix and the filler‐matrix interfacial interactions to provide Fe‐Ni alloy/PA6 nanocomposites of remarkable mechanical performance for engineering applications. nc Fe₄₀Ni₆₀ particles were chemically synthesized. Then Fe₄₀Ni₆₀/PA6 nanocomposites of various nanofiller loading were prepared by compounding via a newly modified master batch technique called ultrasound assisted master batch (UMB), followed by injection molding (IM). Their mechanical properties, morphology and structural parameters were characterized and compared with the corresponding properties of Fe₄₀Ni₆₀/PA6 nanocomposites made by solution mixing (SM) and IM. The study reveals that the UMB process is more cost effective and time efficient, simpler and easier to scale up compared with the SM process. In addition, UMB nanocomposites exhibit superior mechanical properties and distinctive morphology compared with the corresponding SM ones. Moreover, structural analyses indicate that physical structural changes occurred in PA6 due to presence of alloy particles are affected differently by the different compounding methods, profound understanding of such phenomenon is focused throughout the article. These distinctive advantages recommend that UMB technique can be of great potential in commercial production of polymer nanocomposites (PNCs). It is concluded that the sonication of nc Fe₄₀Ni₆₀ particles in dilute polymer solution during UMB compounding, a new step that is incorporated for the first time in the master batch process, is mainly responsible for the good wetting between nanoparticles and polymer chains, strong filler‐matrix interactions and consequently the remarkable mechanical performance of UMB PNCs. POLYM. COMPOS., 35:2343–2352, 2014. © 2014 Society of Plastics Engineers     

Analysis and characterization of comparison shopping behavior in the mobile handset domain

In this work we characterize the session-level behavior of users on an Indian mobile phone comparison shopping website. We also correlate the popularity of handset on various news sources to its popularity on the shopping website. There are three aspects to our study: data analysis, correlation between news sources of product information and popularity of a handset, and behavior prediction. We have used KL divergence to show that a time-homogeneous Markov chain is observed when the number of clicks varies from 5 to 30. Our results depict that Markov chain model does not hold in entirety for comparison shopping setting but tells us how far the Markov chain model holds for this setting. Our analysis corroborates intuition that increasing price leads to decrease in popularity. After the strong correlation between various variables and user behavior was found, we predict the users macro (the overall sales of handset) and micro behavior (whether a user will convert or exit the site) using Markov logic networks. Our predictive model validates the intuition that past browsing behavior is an important predictor for future behavior. Methodology of combining data analysis with machine learning is, in our opinion, a new approach to the empirical study of such data sets.