Applying reinforcement learning for web pages ranking algorithms

Ranking web pages for presenting the most relevant web pages to user's queries is one of the main issues in any search engine. In this paper, two new ranking algorithms are offered, using Reinforcement Learning (RL) concepts. RL is a powerful technique of modern artificial intelligence that tunes agent's parameters, interactively. In the first step, with formulation of ranking as an RL problem, a new connectivity-based ranking algorithm, called RL_Rank, is proposed. In RL_Rank, agent is considered as a surfer who travels between web pages by clicking randomly on a link in the current page. Each web page is considered as a state and value function of state is used to determine the score of that state (page). Reward is corresponded to number of out links from the current page. Rank scores in RL_Rank are computed in a recursive way. Convergence of these scores is proved. In the next step, we introduce a new hybrid approach using combination of BM25 as a content-based algorithm and RL_Rank. Both proposed algorithms are evaluated by well known benchmark datasets and analyzed according to concerning criteria. Experimental results show using RL concepts leads significant improvements in raking algorithms.     

Integration of distributed energy resources into low voltage grid: A market-based multiperiod optimization model

This paper develops a multiperiod optimization model for an interconnected micro grid with hierarchical control that participates in wholesale energy market to maximize its benefit (i.e. revenues-costs). In addition to the operational constraints of distributed energy resources (DER) including both inter-temporal and non-inter-temporal types, the adequacy and steady-state security constraints of micro grid and its power losses are incorporated in the optimization model. In the presented model, DER are integrated into low voltage grid considering both technical and economical aspects. This integration as a micro grid can participate in wholesale energy market as an entity with dual role including producer and consumer based on the direction of exchanged power. The developed model is evaluated by testing on a micro grid considering different cases and the results are analyzed.     

Influence of fading on the Gaussian approximation for BPSK and QPSK with asynchronous cochannel interference

We investigate the performance of BPSK and QPSK with coherent detection and matched filtering in the presence of both time and phase asynchronous cochannel interfering signals. More precisely, we analyze the role played by different channel statistics on the distribution of the decision variable at the output of the matched filter. The results show that the Gaussian approximation is accurate not only in the (obvious) case of a large number of interferers, but also when the desired signal is subject to fading, whatever the number of interferers is. For example, when the desired signal is subject to Rayleigh fading, even in the presence of only one unfaded interferer the Kullback-Leibler distance between the exact distribution of the decision variable and that obtained with the Gaussian approximation on the interference is lower than 0.01 [nats] for all cases of practical interest.     

Electrospun core-shell nanofibers based on polyethylene oxide reinforced by multiwalled carbon nanotube and silicon dioxide nanofillers: A novel and effective solvent-free electrolyte for lithium ion batteries

Insertion of conductive fillers into solvent-free polymer electrolytes enhances electrochemical behavior of the electrolyte membranes leading to higher ionic conductivity, lower capacity fading, and so on. Although, the presence of the conductive fillers in the polymer matrixes increases the risk of electrical shorting, herein, polyethylene oxide (PEO)-based core-shell nanofibers were prepared via a simple electrospinning method. In the core-shell electrospun fibers, ethylene carbonate (EC) and lithium perchlorate (LiClO₄) were used as a plasticizer and as a lithium salt, respectively. The core component was enwrapped by the PEO/EC/LiClO₄ shell part incorporated with SiO₂ nanoparticles. Various properties of the fabricated membranes were evaluated by changing the ratio of multiwall carbon nanotubes (MWCNTs) in the core part of the nanofibers. The morphology and core-shell structure of the electrospun fibers were studied by FESEM and TEM images. According to FTIR and XRD results, addition of the EC plasticizer and the fillers into the as-spun fibers increased the fraction of free ions and the amorphous regions. From electrochemical impedance spectroscopy studies, the ionic conductivity enhanced by insertion of the plasticizer molecules and the filler particles into the core-shell structures. The highest ionic conductivities of 0.09 and 0.21 mS.cm⁻¹ were obtained for the free-filler and the filler-loaded nanofibrous membranes, respectively. The prepared mats obeyed the Arrhenius behavior ( R²~1 ). Dielectric studies confirmed the obtained data from the ionic conductivities. Furthermore, the capacity residual was enhanced from 69% to 85% by incorporation of the MWCNTs filler into the core component of the electrospun nanofibers. The presented results may facilitate development of versatile nanofibrous membranes embedded with the conductive fillers as solvent-free electrolytes applicable in lithium-ion batteries.     

Construction of Near-Optimum Burst Erasure Correcting Low-Density Parity-Check Codes

In this paper, a simple and effective tool for the design of low-density parity-check (LDPC) codes for iterative correction of bursts of erasures is presented. The design method consists of starting from the parity-check matrix of an LDPC code and developing an optimized parity-check matrix, with the same performance over the memoryless erasure channel, and suitable also for the iterative correction of single erasure bursts. The parity-check matrix optimization is performed by an algorithm called pivot searching and swapping (PSS) algorithm. It executes permutations of carefully chosen columns of the parity-check matrix, after a local analysis of particular variable nodes called stopping set pivots. This algorithm can be in principle applied to any LDPC code. If the input parity-check matrix is designed to achieve a good performance over the memoryless erasure channel, then the code obtained after the application of the algorithm provides a good joint correction of independent erasures and single erasure bursts. Numerical results are provided in order to show the algorithm effectiveness when applied to different categories of LDPC codes.     

Error Probability and SINR Analysis of Optimum Combining in Rician Fading

This paper considers the analysis of optimum combining systems in the presence of both co-channel interference and thermal noise. We address the cases where either the desired-user or the interferers undergo Rician fading. Exact expressions are derived for the moment generating function of the SINR which apply for arbitrary numbers of antennas and interferers. Based on these, we obtain expressions for the symbol error probability with M-PSK. For the case where the desireduser undergoes Rician fading, we also derive exact closed-form expressions for the moments of the SINR. We show that these moments are directly related to the corresponding moments of a Rayleigh system via a simple scaling parameter, which is investigated in detail. Numerical results are presented to validate the analysis and to examine the impact of Rician fading on performance.     

Absolute measurements of radiation damage in nanometer-thick films

The problem of absolute measurements of radiation damage in films of nanometer thicknesses is addressed. Thin films of DNA (～2-160 nm) are deposited onto glass substrates and irradiated with varying doses of 1.5-keV X-rays under dry N(2) at atmospheric pressure and room temperature. For each different thickness, the damage is assessed by measuring the loss of the supercoiled configuration as a function of incident photon fluence. From the exposure curves, the G-values are deduced, assuming that X-ray photons interacting with DNA deposit all of their energy in the film. The results show that the G-value (i.e. damage per unit of deposited energy) increases with film thickness and reaches a plateau at 30±5 nm. This thickness dependence provides a correction factor to estimate the actual G-value for films with thicknesses <30 nm thickness. Thus, the absolute values of the damage can be compared with that of films of any thickness under different experimental conditions.     

On optimum combining of M-PSK signals with unequal-power interferers and noise

In this letter, we derive a closed-form symbol-error probability expression for adaptive antenna array with optimum (or, equivalently, linear minimum mean-square error) combining. We consider coherent detection of M-ary phase-shift keying signals in the presence of unequal-power interferers and thermal noise. The analysis is based on our new results on the eigenvalues distribution of central Wishart matrices with correlation.     

A Stochastic Geometry Approach to Coexistence in Heterogeneous Wireless Networks

With the increasing proliferation of different communication devices sharing the same spectrum, it is critical to understand the impact of interference in heterogeneous wireless networks. In this paper, we put forth a mathematical model for coexistence in networks composed of both narrowband (NB) and ultrawideband (UWB) wireless nodes, based on fundamental tools from stochastic geometry. Our model considers that the interferers are spatially scattered according to a Poisson field, and are operating asynchronously in a wireless environment. We first determine the statistical distribution of the aggregate interference for both cases of NB and UWB emitters. We then provide error probability expressions for two dual configurations: 1) a NB victim link subject to the aggregate UWB interference, and 2) a UWB victim link subject to the aggregate NB interference. The results show that while the impact of a single interferer on a link is often negligible due to restrictions on the transmitted power, the aggregate effect of multiple interferers may cause significant degradation. Therefore, aggregate interference must be considered to ensure coexistence in heterogeneous networks. The proposed analytical framework shows good agreement with physical-level simulations of the system.     

Pervaporation study of ethylene glycol dehydration through synthesized (PVA–4A)/polypropylene mixed matrix composite membranes

Poly(vinyl alcohol) (PVA)/zeolite 4A mixed matrix composite membranes supported on polypropylene microfiltration membranes were prepared by solution casting method and crosslinked with glutaraldehyde to investigate their pervaporation (PV) separation properties of water–ethylene glycol mixtures. Scanning electron microscopy images showed homogeneous distribution of zeolite nanoparticles within the polymer matrix without any visible macroscopic voids at the zeolite–polymer interface. The PV experiments were accomplished to investigate the effects of water concentration (10–40 wt%) and temperature (60–80°C) on separation performance of the membranes. It was found out that 5 wt% loading of the 4A nanoparticles into the PVA matrix is optimal to obtain the best separation performance. The experimental results revealed that loading of zeolite 4A enhanced the membrane performance [both permeation flux (5%) and separation factor (32%) at 5 wt% zeolite loading]. POLYM. ENG. SCI., 2013. © 2012 Society of Plastics Engineers