Uncovering the effect of dominant attributes on community topology: A case of facebook networks

Community structure points to structural patterns and reflects organizational or functional associations of networks. In real networks, each node usually contains multiple attributes representing the node’s characteristics. It is difficult to identify the dominant attributes, which have definitive effects on community formation. In this paper, we obtain the overlapping communities using game-theoretic clustering and focus on identifying the dominant attributes in terms of each community. We uncover the association of attributes to the community topology by defining dominance ratio and applying Pearson correlation. We test our method on Facebook data of 100 universities and colleges in the U.S. The study enables an integrating observation on how the offline lives infer online consequences. The results showed that people in class year 2010 and people studying in the same major tend to form denser and smaller groups on Facebook. Such information helps e-marketing campaigns target right customers based on demographic information and without the knowledge of underlying social networks.     

A NEW EYENET MODEL FOR DIAGNOSIS OF DIABETIC RETINOPATHY

Diabetic retinopathy (DR) is an eye disease caused by complications of diabetes and it should be detected early for effective treatment. As diabetes progresses, the vision of a patient may start to deteriorate and lead to diabetic retinopathy. Two types were identified: nonproliferative diabetic retinopathy (NPDR) and proliferative diabetic retinopathy (PDR). In this study, to diagnose diabetic retinopathy, we have proposed a new EYENET model that was obtained by combining the modified probabilistic neural network (PNN) and a modified radial basis function neural network (RBFNN), and hence, it possesses the advantages of both models. The features such as blood vessels and hemorrhages of the NPDR image and exudates of the PDR image are extracted from the raw images using image-processing techniques and are fed to the classifier for classification. A total of 600 fundus images were used, out of which 400 were used for training, and 200 images were used for testing. Experimental results show that PNN has an accuracy of 96%, modified PNN has an accuracy of 97.5%, RBFNN has an accuracy of 93.5%, modified RBFNN has an accuracy of 95.5%, and the proposed EYENET model has an accuracy of 98.5%. This infers that our proposed model outperforms all other models.     

Stub Column Tests of Cold-Formed Steel Built-Up Square Sections with Intermediate Stiffeners

Strength of Materials - This paper describes the experimental and theoretical investigation on ultimate strength of cold-formed steel built-up square stub columns under axial compression. In total,...     

Sporadic model building for efficiency enhancement of the hierarchical BOA

Efficiency enhancement techniques—such as parallelization and hybridization—are among the most important ingredients of practical applications of genetic and evolutionary algorithms and that is why this research area represents an important niche of evolutionary computation. This paper describes and analyzes sporadic model building, which can be used to enhance the efficiency of the hierarchical Bayesian optimization algorithm (hBOA) and other estimation of distribution algorithms (EDAs) that use complex multivariate probabilistic models. With sporadic model building, the structure of the probabilistic model is updated once in every few iterations (generations), whereas in the remaining iterations, only model parameters (conditional and marginal probabilities) are updated. Since the time complexity of updating model parameters is much lower than the time complexity of learning the model structure, sporadic model building decreases the overall time complexity of model building. The paper shows that for boundedly difficult nearly decomposable and hierarchical optimization problems, sporadic model building leads to a significant model-building speedup, which decreases the asymptotic time complexity of model building in hBOA by a factor of to where n is the problem size. On the other hand, sporadic model building also increases the number of evaluations until convergence; nonetheless, if model building is the bottleneck, the evaluation slowdown is insignificant compared to the gains in the asymptotic complexity of model building. The paper also presents a dimensional model to provide a heuristic for scaling the structure-building period, which is the only parameter of the proposed sporadic model-building approach. The paper then tests the proposed method and the rule for setting the structure-building period on the problem of finding ground states of 2D and 3D Ising spin glasses.     

Reduced stiffness buckling of sandwich cylindrical shells under uniform external pressure

A reduced stiffness lower bound method for the buckling of laterally pressure loaded sandwich cylindrical shell is proposed. Also, an attempt is made to assess the validity of the proposed reduced stiffness lower bound with FEM numerical examples. In addition, the proposed method is compared with classical and Plantema's approaches of the buckling of the laterally pressure loaded sandwich cylindrical shell. Comparison of the proposed reduced stiffness lower bound with that obtained from non-linear FEM analysis verifies that it indeed provides a safe lower bound to the buckling of laterally pressure loaded sandwich cylindrical shells. The attractive feature of the proposed reduced stiffness method is that it can be readily used in designing laterally pressure loaded sandwich cylindrical shells without being concerned about geometrical imperfections.     

Effect of nanoparticle nature on hydrogen concentration profiles and improved switching characteristics in Gd switchable mirrors

A detailed elastic recoil detection analysis using 40 MeV 28Si5+ ions has been carried out to study the changes in the H concentration and concentration profiles during the hydrogenation/dehydrogenation process in polycrystalline and nanoparticle Gd layers formed using vacuum evaporation and inert gas evaporation techniques, respectively. Nanoparticle sample exhibits a larger difference in the [H]/[Gd] values (2.9 and 1.7) in comparison to polycrystalline sample (2.4 and 2.0) in the loaded and deloaded states. Hydrogenation/dehydrogenation activity is restricted to the top portion in case of polycrystalline sample. In contrast to this, size induced structural transformation; enhanced surface area and the presence of large number of inter particle boundaries due to nanoparticle character result in the complete Gd layer becoming active during switching.     

High-Performance Multifunctional Fuel Additives Derived from Renewable Fatty Acids and Phosphazene

The increasing need for high-quality diesel fuel can be fulfilled by using environmental friendly, high-performance additives. We have synthesized novel fatty acid-derived multifunctional additives coupled with phosphazene moiety having the backbone of alternative nitrogen and phosphorous atoms which provided stability, while the additives possessed affinity for the metal surface to form a surface-complex film leading to the hindered metal contact and providing friction-reducing and antiwear properties. In addition, the applicability of the synthesized additives was investigated by testing the anti-oxidant and anticorrosion properties in polyethylene glycol (PEG200) which was taken as the reference base fluid. A rotating bomb oxidation test was used to evaluate the anti-oxidant characteristics, and a four-ball test for the tribological properties. Among the synthesized additives, phosphazene ester of erucic acid exhibited the best antiwear and antifriction properties; however, phosphazene ester of oleic acid was found to be the best anti-oxidant additive.     

Eight Element Low-Cost Microstrip MIMO Antenna for Wi-MAX and 5G Applications

Wireless Personal Communications - In this communication, a low-cost four/eight-element MIMO antenna is proposed for Wi-MAX (IEEE 802.16d) and 5G applications band. A reverse U-shaped slot is...     

SDSN@RT: A middleware environment for single-instance multitenant cloud applications

With the single-instance multitenancy (SIMT) model for composite Software-as-a-Service (SaaS) applications, a single composite application instance can host multiple tenants, yielding the benefits of better service and resource utilization and reduced operational cost for the SaaS provider. An SIMT application needs to share services and their aggregation (the application) among its tenants while supporting variations in the functional and performance requirements of the tenants. The SaaS provider requires a middleware environment that can deploy, enact, and manage a designed SIMT application, to achieve the varied requirements of the different tenants in a controlled manner. This paper presents the SDSN@RT (software-defined service networks at runtime) middleware environment that can meet the aforementioned requirements. SDSN@RT represents an SIMT composite cloud application as a multitenant service network, where the same service network simultaneously hosts a set of virtual service networks, one for each tenant. A service network connects a set of services and coordinates the interactions between them. A virtual service network realizes the requirements for a specific tenant and can be deployed, configured, and logically isolated in the service network at runtime. SDSN@RT also supports the monitoring and runtime changes of the deployed multitenant service networks. We show the feasibility of SDSN@RT with a prototype implementation and demonstrate its capabilities to host SIMT applications and support their changes with a case study. The performance study of the prototype implementation shows that the runtime capabilities of our middleware incur little overhead.     

BILEAT: a highly generalized and robust approach for unified aspect-based sentiment analysis

Applied Intelligence - Aspect-based sentiment analysis (ABSA) includes two subtasks, namely, aspect term extraction and aspect-level sentiment classification. Most existing works address these...