Search

We address the problem of efficiently estimating the influence degree for all the nodes simultaneously in the network under the SIR setting. The proposed approach is a further improvement over the existing work of the bond percolation process which was demonstrated to be very effective, i.e., three orders of magnitude faster than direct Monte Carlo simulation, in approximately solving the influence maximization problem. We introduce two pruning techniques which improve computational efficiency by an order of magnitude. This approach is generic and can be instantiated to any specific diffusion model. It does not require any approximations or assumptions to the model that were needed in the existing approaches. We demonstrate its effectiveness by extensive experiments on two large real social networks. Main finding includes that different network structures have different epidemic thresholds and the node influence can identify influential nodes that the existing centrality measures cannot. We analyze how the performance changes when the network structure is systematically changed using synthetically generated networks and identify important factors that affect the performance.

The detrimental impact of a radiotherapy waiting list can in part be compensated by patient prioritisation. Such prioritisation is phrased as an optimisation problem where the probability of local control for the overall population is the objective to be maximised and a simple analytical solution derived. This solution is compared with a simulation of a waiting list for the same population of patients. It is found that the analytical solution can provide an optimal ordering of patients though cannot explicitly constrain optimal waiting times. The simulation-based solution was undertaken using both the analytical solution and a numerical optimisation routine for daily patient ordering. Both solutions provided very similar results with the analytical approach reducing the calculation time of the numerical solution by several orders of magnitude. It is suggested that treatment delays due to resource limitations and resulting waiting lists be incorporated into treatment optimisation and that the derived analytical solution provides a mechanism for this to occur.

Moore’s Law technology scaling has improved VLSI performance by five orders of magnitude in the last four decades. As advanced technologies continue the pursuit of Moore’s Law, a variety of challenges will need to be overcome. One of these challenges is management of process variation. This paper discusses the importance of process variation in modern CMOS transistor technology, reviews front-end variation sources, presents device and circuit variation measurement techniques (including circuit and SRAM data from the 32 nm node), and compares recent intrinsic transistor variation performance from the literature.

A number of models have recently been, or are currently being, developed to enable the assessment of radiation doses from ionising radiation to non-human species. A key component of these models is the ability to predict whole-organism activity concentrations in a wide range of wildlife. In this paper, we compare the whole-organism activity concentrations predicted by eight models participating within the IAEA Environmental Modelling for Radiation Safety programme for a range of radionuclides to terrestrial and freshwater organisms. In many instances, there was considerable variation, ranging over orders of magnitude, between the predictions of the different models. Reasons for this variability (including methodology, data source and data availability) are identified and discussed. The active participation of groups responsible for the development of key models within this exercise is a useful step forward in providing the transparency in methodology and data provenance required for models which are either currently being used for regulatory purposes or which may be used in the future. The work reported in this paper, and supported by other findings, demonstrates that the largest contribution to variability between model predictions is the parameterisation of their transfer components. There is a clear need to focus efforts and provide authoritative compilations of those data which are available.

Automation has increased the speed of DNA sequencing by established methods by several orders of magnitude. Now, commercial and academic efforts to develop alternative sequencing technologies are trying to push the envelope even further. Laura Bonetta reports.

New methods are considered for correcting additive and multiplicative systematic errors of ratiometric measurement transducers utilizing sensors whose output signal u(x) is proportional to their power supply voltage or current. It is shown that synthesized algorithms for processing the intermediate results of measurements provide a reduction in the additive errors by several orders of magnitude and almost completely eliminate multiplicative errors.

An international project is underway to design and build a radiotelescope with an effective collecting area two orders of magnitude greaterthan the largest existing instruments. One of the many scientific goals of thisinstrument will be the investigation of the extragalactic radio sourcepopulation at flux densities two to three orders of magnitude fainter thanthe limits of existing observations. We present simulations of the radiosky at 1.4 GHz down to a flux density limit of 0.1 μJy usingextrapolations of known radio luminosity functions for two differentpopulation scenarios. The resulting simulations confirm that a resolutionof 0'' .1 is necessary to avoid formal confusion, but sourceblending may still dominate if the intrinsic size of such faint sourcesis larger than a few kiloparsecs.

Orders of magnitude reasoning in artificial intelligence (AI) and qualitative algebra are discussed and applied in selecting shield in this paper. It includes basic quantitative calculation, qualitative constraint and principle of orders of magitude reasoning. The method has potential prospects in dealing with the engineering problems.

The effects of applied voltage and the duration of application upon human epidermal membrane (HEM) alterations and recovery were investigated. All experiments were conducted using a two-chamber diffusion cell with constant DC voltage (250-4000 mV) applied over a predetermined period, and HEM changes were monitored by measuring the electrical resistance before and after voltage termination. The key findings were that the rate of decrease in resistance was strongly dependent upon the applied voltage, the reversible recovery times were dependent upon both the magnitude and the duration of the applied field (frequently were several orders of magnitude greater than times for attaining significant resistance reduction), and reversible recovery times were much longer when lower voltages were applied for longer times to attain the same decrease in electrical resistance than for higher voltages at short times. These findings closely parallel those obtained on electrical breakdown/recovery of bilayer membranes (electroporation). The second part of this work examined the hypothesis that decreases in HEM electrical resistance induced by the applied voltage are accompanied by proportional increases in HEM permeability. A study was designed to test this hypothesis involving a four-stage protocol with HEM: passive transport, 250-mV iontophoresis, 2000-mV iontophoresis for 10 min, then back to 250-mV iontophoresis. The data obtained strongly support the view that the HEM alterations induced by the electric field result in pore formation and in the expected changes in HEM permeability.

A review of technology assessment studies as well as original research papers is presented in two diagrams with respect to estimating (1) the performance of fusion reactors in containing tritium, and (2) the radiological consequences of normal operational losses as well as accidental releases of tritium. Predictions are shown to vary over several orders of magnitude.

The electrochemical corrosion behaviour of sixteen currently marketed amalgams for dental fillings was studiedin vitro under experimental conditions closely resembling those of clinical application. Depending upon the type of amalgam and the treatment of the filling the electrochemical corrosion currents varied over several orders of magnitude; their change as a function of time was also very different. Suggestions are made for a stability rating of the products based on the results obtained.