A Unified Framework for Providing Recommendations in Social Tagging Systems Based on Ternary Semantic Analysis

Social Tagging is the process by which many users add metadata in the form of keywords, to annotate and categorize items (songs, pictures, Web links, products, etc.). Social tagging systems (STSs) can provide three different types of recommendations: They can recommend 1) tags to users, based on what tags other users have used for the same items, 2) items to users, based on tags they have in common with other similar users, and 3) users with common social interest, based on common tags on similar items. However, users may have different interests for an item, and items may have multiple facets. In contrast to the current recommendation algorithms, our approach develops a unified framework to model the three types of entities that exist in a social tagging system: users, items, and tags. These data are modeled by a 3-order tensor, on which multiway latent semantic analysis and dimensionality reduction is performed using both the Higher Order Singular Value Decomposition (HOSVD) method and the Kernel-SVD smoothing technique. We perform experimental comparison of the proposed method against state-of-the-art recommendation algorithms with two real data sets (Last.fm and BibSonomy). Our results show significant improvements in terms of effectiveness measured through recall/precision.     

Phone duration modeling: overview of techniques and performance optimization via feature selection in the context of emotional speech

Accurate modeling of prosody is prerequisite for the production of synthetic speech of high quality. Phone duration, as one of the key prosodic parameters, plays an important role for the generation of emotional synthetic speech with natural sounding. In the present work we offer an overview of various phone duration modeling techniques, and consequently evaluate ten models, based on decision trees, linear regression, lazy-learning algorithms and meta-learning algorithms, which over the past decades have been successfully used in various modeling tasks. Furthermore, we study the opportunity for performance optimization by applying two feature selection techniques, the RReliefF and the Correlation-based Feature Selection, on a large set of numerical and nominal linguistic features extracted from text, such as: phonetic, phonologic and morphosyntactic ones, which have been reported successful on the phone and syllable duration modeling task. We investigate the practical usefulness of these phone duration modeling techniques on a Modern Greek emotional speech database, which consists of five categories of emotional speech: anger, fear, joy, neutral, sadness. The experimental results demonstrated that feature selection significantly improves the accuracy of phone duration prediction regardless of the type of machine learning algorithm used for phone duration modeling. Specifically, in four out of the five categories of emotional speech, feature selection contributed to the improvement of the phone duration modeling, when compared to the case without feature selection. The M5p trees based phone duration model was observed to achieve the best phone duration prediction accuracy in terms of RMSE and MAE.     

Modeling and optimization of a 1 kWe HT-PEMFC-based micro-CHP residential system

A high temperature-proton exchange membrane (HT-PEMFC)-based micro-combined-heat-and-power (CHP) residential system is designed and optimized, using a genetic algorithm (GA) optimization strategy. The proposed system consists of a fuel cell stack, steam methane reformer (SMR) reactor, water gas shift (WGS) reactor, heat exchangers, and other balance-of-plant (BOP) components. The objective function of the single-objective optimization strategy is the net electrical efficiency of the micro-CHP system. The implemented optimization procedure attempts to maximize the objective function by variation of nine decision variables. The value of the objective function for the optimum design configuration is significantly higher than the initial one, with a 20.7% increase.     

Modeling of NC-AFM Imaging of Alkanethiols on the Au (111) Surface

We investigate the mechanisms of contrast formation in NC-AFM imaging of self-assembled monolayers of alkanethiols on the Au (111) surface. Comparing the potential models with implicit and explicit electrostatics, we demonstrate that, similar to the imaging of polar solid surfaces, the electrostatic interaction plays the central role in contrast formation. Careful comparison of several tip models showed that the model of a clean SiO₂ tip gives the closest agreement with the experimental data.     

A discrete methodology for controlling the sign of curvature and torsion for NURBS

This paper develops a discrete methodology for approximating the so-called convex domain of a NURBS curve, namely the domain in the ambient space, where a user-specified control point is free to move so that the curvature and torsion retains its sign along the NURBS parametric domain of definition. The methodology provides a monotonic sequence of convex polyhedra, converging from the interior to the convex domain. If the latter is non-empty, a simple algorithm is proposed, that yields a sequence of polytopes converging uniformly to the restriction of the convex domain to any user-specified bounding box. The algorithm is illustrated for a pair of planar and a spatial Bézier configuration.     

Detection of icing on wind turbine blades by means of vibration and power curve analysis

Ice accretion on wind turbines' blades is one of the main challenges of systems installed in cold climate locations, resulting in power performance deterioration and excessive nacelle oscillation. In this work, consistent detection of icing events is achieved utilizing indications from the nacelle accelerometers and power performance analysis. Features extracted from these two techniques serve as inputs in a decision‐making scheme, allowing early activation of de‐icing systems or shut down of the wind turbine. An additional parameter is the month of operation, assuring consistent outcomes in both winter and summer seasons. The amplitude of lateral nacelle vibration at rotor speed is the used condition indicator from vibration standpoint, which is verified by the presence of sinusoidal shape in high‐resolution time waveforms. Employment of k‐nearest neighbour on wind speed ‐ power production data sets leads to successful recognition of power performance deterioration. Results from one wind park consisting of 13 turbines operating under icing are presented, where similar patterns on both vibration and power curve data validate the effectiveness of the proposed approach on the reliable detection of icing formation. Copyright © 2015 John Wiley & Sons, Ltd.     

Microalgal Lipid Extracts Have Potential to Modulate the Inflammatory Response: A Critical Review

Noncommunicable diseases (NCD) and age-associated diseases (AAD) are some of the gravest health concerns worldwide, accounting for up to 70% of total deaths globally. NCD and AAD, such as diabetes, obesity, cardiovascular disease, and cancer, are associated with low-grade chronic inflammation and poor dietary habits. Modulation of the inflammatory status through dietary components is a very appellative approach to fight these diseases and is supported by increasing evidence of natural and dietary components with strong anti-inflammatory activities. The consumption of bioactive lipids has a positive impact on preventing chronic inflammation and consequently NCD and AAD. Thus, new sources of bioactive lipids have been sought out. Microalgae are rich sources of bioactive lipids such as omega-6 and -3 polyunsaturated fatty acids (PUFA) and polar lipids with associated anti-inflammatory activity. PUFAs are enzymatically and non-enzymatically catalyzed to oxylipins and have a significant role in anti and pro-resolving inflammatory responses. Therefore, a large and rapidly growing body of research has been conducted in vivo and in vitro, investigating the potential anti-inflammatory activities of microalgae lipids. This review sought to summarize and critically analyze recent evidence of the anti-inflammatory potential of microalgae lipids and their possible use to prevent or mitigate chronic inflammation.     

Economic evaluation of energy saving measures in a common type of Greek building

This paper deals with the economic analysis and evaluation of various energy saving measures in the building sector, focusing on a domestic detached house in Greece, i.e. in a typical Mediterranean climate. In order to detect the energy saving measures that, in addition to energy benefits, can also provide economic profits, the study examines the following measures: all kinds of insulation; upgrading of the heating system; use of thermal solar systems; upgrading of lighting; upgrading of electric appliances; upgrading of the cooling system. The economic evaluation methods used for ranking the energy saving measures are the Net Present Value, the Internal Rate of Return, the Savings to Investment Ratio and the Depreciated Payback Period. It has been found that amongst the most effective energy saving methods are the upgrading of lighting, the insulation of the roof of the building and the installation of an automatic temperature control system.     

An inter-comparison exercise on the capabilities of CFD models to predict the short and long term distribution and mixing of hydrogen in a garage

The paper presents the results of the CFD inter-comparison exercise SBEP-V3, performed within the activity InsHyde, internal project of the HySafe network of excellence, in the framework of evaluating the capability of various CFD tools and modelling approaches in predicting the short and long term mixing and distribution of hydrogen releases in confined spaces. The experiment simulated was INERIS-TEST-6C, performed within the InsHyde project by INERIS, consisting of a 1 g/s vertical hydrogen release for 240 s from an orifice of 20 mm diameter into a rectangular room (garage) of dimensions 3.78 × 7.2 × 2.88 m in width, length and height respectively. Two small openings at the bottom of the front side of the room assured constant pressure conditions. During the test hydrogen concentration time histories were measured at 12 positions in the room, for a period up to 5160 s after the end of release, covering both the release and the subsequent diffusion phases. The benchmark was organized in two phases. The first phase consisted of blind simulations performed prior to the execution of the tests. The second phase consisted of post-calculations performed after the tests were concluded and the experimental results made available. The participation in the benchmark was high: 12 different organizations (2 non-HySafe partners), 10 different CFD codes and 8 different turbulence models. Large variation in predicted results was found in the first phase of the benchmark, between the various modelling approaches. This was attributed mainly to differences in turbulence models and numerical accuracy options (time/space resolution and discretization schemes). During the second phase of the benchmark the variation between predicted results was reduced.