Webspam demotion: Low complexity node aggregation methods

Search engines result pages (SERPs) for a specific query are constructed according to several mechanisms. One of them consists in ranking Web pages regarding their importance, regardless of their semantic. Indeed, relevance to a query is not enough to provide high quality results, and popularity is used to arbitrate between equally relevant Web pages. The most well-known algorithm that ranks Web pages according to their popularity is the PageRank.The term Webspam was coined to denotes Web pages created with the only purpose of fooling ranking algorithms such as the PageRank. Indeed, the goal of Webspam is to promote a target page by increasing its rank. It is an important issue for Web search engines to spot and discard Webspam to provide their users with a nonbiased list of results. Webspam techniques are evolving constantly to remain efficient but most of the time they still consist in creating a specific linking architecture around the target page to increase its rank.In this paper we propose to study the effects of node aggregation on the well-known ranking algorithm of Google (the PageRank) in the presence of Webspam. Our node aggregation methods have the purpose to construct clusters of nodes that are considered as a sole node in the PageRank computation. Since the Web graph is way to big to apply classic clustering techniques, we present four lightweight aggregation techniques suitable for its size. Experimental results on the WEBSPAM-UK2007 dataset show the interest of the approach, which is moreover confirmed by statistical evidence.     

Investigation of some niobate-based dielectrics in view of base metal co-sintering

Bearing in mind the excellent dielectric properties at high frequency of some niobates like ZnNb₂O₆, Zn₃Nb₂O₈, BaNb₂O₆, Ba₅Nb₄O₁₅ (_r 20–45, tan δ < 10 × 10⁻⁴ and ρ_i > 10¹⁰ Ω cm), synthesis, sintering and properties of these oxides are reported. The lowering of their sintering temperature has been investigated for these four ceramics using sintering aids. Using appropriate additive, it is possible to densify these ceramics at a temperature for which base metal electrodes, e.g. Cu and Ag can be employed. Two formulations were found to be sinterable at 1000 °C (lower than the copper melting point) whereas a third formulation is co-sinterable with silver electrodes. For this later, the dielectric properties are globally maintained in comparison with the pure compound sintered at 1200 °C. This result makes this formulation suitable for silver based passive components devices manufacturing.     

Identification of chemical species created during γ-irradiation of antioxidant used in polyethylene and polyethylene-co-vinyl acetate multilayer film

With the increasing use of γ-irradiated containers made of multilayer polymeric flexible films for food and biopharmaceutical applications, the possible migration of degradation products of the polymers and their additives is becoming a topic of concern. This article aims at highly reliably identifying the degradation products generated after gamma irradiation and their origin to later on assess their potential harmfulness in single-use containers. In this study, GC–MS is used to identify by-products created by γ-irradiation of primary and secondary antioxidants usually present in polyolefin-based biotechnological single-use materials and to confirm identification relevancy based on the literature survey or standard when available. Degradation pathways are proposed to account for the formation of by-products identified during the study and to list intermediates and other by-products present in too small amounts to be detected and identified accurately in all extractable studies.     

Neural models of solar collectors for prediction of daily performance

This paper presents the application of an online identification neural technique to the prediction of the in-situ daily performance of solar collectors. First, it is shown that the use of the Laplace transform helps to find the order of an approximated model; the input of the studied system being the solar radiation. Then it is shown that an NNOE model can be more accurate using the right size of the regression vector; the learning database consisting of the data obtained during half a day. Finally, it is shown that a multiple inputs single output (MISO) NNOE model can be accurate; the inputs being the solar radiation and the thermal heat loss conductance that depends on the wind velocity. In any case the differential between the actual value of the daily energy and the value computed by a neural model is less than 0.5%.     

The impact of Stereotype Threat on the simulated driving performance of older drivers

Older drivers are perceived as being dangerous and overly cautious by other drivers. We tested the hypothesis that this negative stereotype has a direct influence on the performance of older drivers. Based on the Stereotype Threat literature, we predicted that older driving performance would be altered after exposure to a Stereotype Threat. Sixty-one older drivers aged 65 and above completed a simulated driving assessment course. Prior to testing, half of the participants were told that the objective of the study was to investigate why older adults aged 65 and above were more implicated in on-road accidents (Stereotype Threat condition) and half were showed a neutral statement. Results confirmed that exposure to the threat significantly altered driving performance. Older adults in the Stereotype Threat condition made more driving mistakes than those in the control group. Interestingly, under a Stereotype Threat condition, older adults tended to commit more speeding infractions. We also observed that domain identification (whether driving is deemed important or not) moderated the impact of the threat. Taken together, these results support recent older drivers’ performance models suggesting that the interaction between individual and social factors need to be considered when examining older drivers’ performance.     

Online Fouling Detection of Domestic Hot Water Heat Exchangers

Due to hardness of cold water supply in many countries, there is a risk of fouling in domestic hot water (DHW) counterflow plate heat exchangers. The scaling will result in increased resistance to heat transfer, which has negative effects on the economics of the district heating network. A common approach is to clean or change the heat exchanger periodically, which can be expensive if only limited fouling has occurred (unnecessary) or if a higher than expected scaling layer has formed (inefficiency). A better approach is to monitor the state of the heat exchangers and clean them when actually required. This would result in more energy-efficient operation and provide an optimum schedule for heat exchanger cleaning. This can be simple if the heat exchangers are operating under steady-state conditions; however, if large variations in the inlets are experienced, as is the case with the mass flows in DHW heat exchangers, it quickly becomes impossible with standard methods. In this paper it is proposed to monitor the state of the heat exchanger online by using measurements that are easily obtainable under normal operation and applying fast mathematical models to estimate the overall heat transfer coefficient of the heat exchanger. The results show that the methods proposed can be used to detect fouling in DHW heat exchangers.     

Accurate measurement of anhydride fluorhydric acid concentration in controlled gaseous flow with STW sensors

This paper presents theoretical and experimental developments for the implementation of surface acoustic waves (SAW) sensors able to detect small concentrations of anhydride fluorhydric (HF) acid in air. Solutions based on the use of surface transverse waves (STW) on quartz (YXlt)/36 degrees/90 degrees have been analyzed to evaluate their sensitivity to HF. Devices have been tested first in a NH4F solution to evaluate the kinetics of the reaction. Measurements then were performed under various gaseous conditions to characterize the sensors when they are submitted to different controlled dilutions of HF in air. STW resonators have been successfully tested in different conditions, with capabilities to detect HF concentration much smaller than 1 ppm.     

Investigation of the biopharmaceutical behavior of theophylline hydrophilic matrix tablets using USP methods and an artificial digestive system

This work aimed to investigate the biopharmaceutical behavior of hydrophilic matrix tablets of theophylline using different in vitro methods: USP II, USP IV, and a novel in vitro system simulating the gastrointestinal tract in man called the artificial digestive system (ADS). The potentiality of each method was evaluated by establishing in vitro/in vivo correlation. Using USP methods, the drug release was pH-independent and dependent on agitation intensity. Level A IVIVCs could be established using the different in vitro methods but one to one correlation was established only when the ADS method was used. For the prediction of in vivo drug dosage form behavior based on in vitro methods, the ADS showed a high predictability when compared to USP in vitro methods.     

Detection and high-precision positioning of liquid droplets using SAW systems

The capability to accurately handle liquids in small volumes is a key point for the development of lab-on-chip devices. In this paper, we investigate an application of surface acoustic waves (SAW) for positioning micro-droplets. A SAW device based on a 2 x 2 matrix of inter-digital transducers (IDTs) has been fabricated on a (YXI)/128 degrees LiNbO3 substrate, which implies displacement and detection in two dimensions of droplets atop a flat surface. Each IDT operates at a given frequency, allowing for an easy addressing of the active channel. Furthermore, very low cross-talk effects were observed as no frequency mixing arose in our device. Continuous as well as pulsed excitations of the IDTs have been studied, yielding, respectively, continuous and step-by-step droplet displacement modes. In addition, we also have used these two excitation types to control the velocity and the position of the droplets. We also have developed a theoretical analysis of the detection mode, which has been validated by experimental assessment.