Encrypted Network Traffic Classification and Resource Allocation with Deep Learning in Software Defined Network

Wireless Personal Communications - The climate has changed absolutely in every area in just a few years as digitized, making high-speed internet service a significant need in the future. Future...     

A new efficient SIF-based FCIL (SIF–FCIL) mining algorithm in predicting the crime locations

In our innovative crime location forecast method, at the outset, the crime features are mined from the crime database and used for performing the adaptive mutation-based artificial bee colony (AMABC) algorithm, in which the database attributes and crime values are bunched together. Subsequently, the frequent closed itemsets lattice (FCIL) is built by the rules support factor values, and from this the frequent rules are extracted. In the course of the FCIL creation, the clustered attributes values are processed like a sliding window. In accordance with the frequent rules, the related crime locations are created. Thus, our proposed sliding with itemsets factor-based FCIL proposed technique is endowed with the superb skill of fruitfully forecasting the locations by means of AMABC and FCIL methods. In our innovative approach, we apply an UCI Machine Learning Repository-Communities and Crime Data Set for the offence investigation. The novel method is analysed and contrasted with the modern mining algorithms such as Apriori, Eclat and conservative FCIL.     

Evolutionary multi objective optimization for rule mining: a review

Evolutionary multi objective optimization (EMOO) systems are evolutionary systems which are used for optimizing various measures of the evolving system. Rule mining has gained attention in the knowledge discovery literature. The problem of discovering rules with specific properties is treated as a multi objective optimization problem. The objectives to be optimized being the metrics like accuracy, comprehensibility, surprisingness, novelty to name a few. There are a variety of EMOO algorithms in the literature. The performance of these EMOO algorithms is influenced by various characteristics including evolutionary technique used, chromosome representation, parameters like population size, number of generations, crossover rate, mutation rate, stopping criteria, Reproduction operators used, objectives taken for optimization, the fitness function used, optimization strategy, the type of data, number of class attributes and the area of application. This study reviews EMOO systems taking the above criteria into consideration. There are other hybridization strategies like use of intelligent agents, fuzzification, meta data and meta heuristics, parallelization, interactiveness with the user, visualization, etc., which further enhance the performance and usability of the system. Genetic Algorithms (GAs) and Genetic Programming (GPs) are two widely used evolutionary strategies for rule knowledge discovery in Data mining. Thus the proposed study aims at studying the various characteristics of the EMOO systems taking into consideration the two evolutionary strategies of Genetic Algorithm and Genetic programming.     

All fiber based multispeckle modality endoscopic system for imaging medical cavities

Disease detection in body cavities, such as the detection of abnormal growths in the colon path, has been illustrated here using an image fiber guided catheter based multispeckle modality endoscopic system. An all fiber-optic approach for the illumination and imaging of the inner cavity walls is adopted here. An endoscope probe to carry the illumination fibers as well as the imaging lens-image fiber unit is designed and custom fabricated in order to operate the probe in its various direction sensitive configurations. This is facilitated by the selection of suitable optical elements such as beam combiner and biprism at the probe proximal end. Experimental investigations were carried out using the endoscope system employing phantom model of colon as the test specimen that has normal and abnormal (representing growth) regions and the obtained results indicated the system effectiveness in identifying the abnormal growths at an early stage.     

Optimized ensemble decision-based multi-focus imagefusion using binary genetic Grey-Wolf optimizer in camera sensor networks

Modern developments in image technology enabled easy access to an innovative type of sensor-based networks, Camera or Visual Sensor Networks (VSN). Nevertheless, more sensor data sources bring about the problem of overload information. To solve this problem, some researchers have been carried out on the techniques to counteract the data overload caused by sensors without losing useful data. The aim of fusion in each application is to combine images from several sensors, which leads to the decreased amount of input image data, producing an image with more accurate data. This paper proposes a noisy feature removal scheme for multi-focus image fusion combining the decision information of optimized individual features. The proposed scheme is developed in two main steps. In the first step, the diverse types of features are extracted from each block of input blurred images. The useful information of these individual features indicates which image block is more focused among corresponding blocks in source images. After that, noisy features are removed using binary Genetic Grey wolf optimizer (GGWO) algorithm. The ensemble decision based on individual features is employed to fuse blurred images in the second step. The experimentation is evaluated on different multi-focus images and it reveals that GGWO based proposed method performs better visual quality than other methods.     

Non-uniform Euler-Bernoulli beams under a single moving oscillator: An approximate analytical solution in time domain

Dynamic responses of simply supported non-uniform beams traversed by a moving oscillator are analysed in this paper. An approximate analytical method based on Rayleigh-Ritz (R-R) formulation is developed. The fundamental approximate mode obtained from R-R method is used in the present formulation to determine the responses of the beam and the oscillator. Effects of surface irregularities on the displacement and acceleration responses of the beam and the vehicle are also analysed. The results are compared with those obtained using Finite element method (FEM). A numerical example is provided to illustrate the validity of the present method which shows that the proposed method is simple, computationally more efficient compared to FEM and gives fairly good results. Though the single-mode approach used in the present paper is a classical one and numerous studies on the responses of uniform beams under moving loads have been reported in the past, its application to non-uniform beams (for which there does not exist any closed form expression for mode shapes) under a moving load, especially a moving oscillator, is presented for the first time.     

New approach for prediction of influence of vehicle dynamics parameters on instability of unmanned track vehicle using robotic approach

Journal of Mechanical Science and Technology - Achieving stability of unmanned, heavy tracked vehicles is challenging, especially under conditions of teleoperation because the remote operator...     

Magnetic Properties of Mn‐doped SnO2 Thin Films Prepared by the Sol–Gel Dip Coating Method for Dilute Magnetic Semiconductors

Manganese‐doped tin oxide (SnO₂:Mn) thin films were deposited on glass substrates by the sol–gel dip coating technique. The effect on structural, morphological, magnetic, electrical, and optical properties in the films with different Mn concentrations (0–5 mol%) were investigated. X‐ray diffraction patterns (XRD) showed the deterioration of crystallinity with increase in Mn‐doping concentration. Scanning electron microscopy (SEM) studies showed an inhibition of grain growth with an increase in Mn concentration. X ray photoelectron spectroscopy (XPS) revealed the presence of Sn⁴⁺ and Mn³⁺ in SnO₂: Mn films. SnO₂: Mn films show ferromagnetic and paramagnetic behavior. These SnO₂:Mn films acquire n‐type conductivity for 0–3 mol% (SnO₂ ‐ Sn_0.97Mn_0.03O_{2) ‐}doping concentration and p type for 5 mol% Mn‐doping concentration(Sn_0.95Mn_0.05O₂) in SnO₂ films. An average transmittance of > 75% (in UV‐Vis region) was observed for all the SnO₂:Mn films. Optical band gap energy of SnO₂: Mn films were found to vary in the range 3.55 to 3.71 eV with the increase in Mn‐doping concentration. Photoluminescence (PL) spectra of the films exhibited an increase in the emission intensity with increase in Mn‐doping concentration which may be due to structural defects or luminescent centers, such as nanocrystals and defects in the SnO₂. Such SnO₂:Mn films with structural, magnetic and optical properties can be used as dilute magnetic semiconductors.     

Decreasing the Linoleic Acid to α-Linolenic Acid Diet Ratio Increases Eicosapentaenoic Acid in Erythrocytes in Adults

The n-6/n-3 fatty acid (FA) ratio has increased in the Western-style diet to ~10–15:1 during the last century, which may have contributed to the rise in cardiovascular disease (CVD). Prior studies have evaluated the effects on CVD risk factors of manipulating the levels of n-6 and n-3 FA using food and supplements or investigated the metabolic fate of linoleic acid (LNA) and α-linolenic acid (ALA) by varying the n-6/n-3 ratios. However, no previous studies have investigated the potential interaction between diet ratios and supplementation with eicosapentaenoic acid (EPA, 20:5n-3) and docosahexaenoic acid (DHA, 22:6n-3). We used a factorial design approach with adults (n = 24) in a controlled feeding trial to compare the accretion of EPA and DHA into red blood cell membranes (RBC) by adding a direct source (algal oil supplement) of EPA and DHA in a diet with a 10:1 versus 2:1 ratio of n-6/n-3 FA. Subjects were randomized into 8-week crossover diet sequences and each subject consumed three of four diets [10:1, 10:1 plus supplement (10:1 + S), 2:1 and 2:1 + S]. LNA and ALA intakes were 9.4 and 7.7%, and 1.0 and 3.0% during the low and high ALA diets, respectively. Compared to the Western-style 10:1 diet, the 2:1 diet increased EPA by 60% (P < 0.0001) in RBC membranes without the direct EPA source and a 34% increase (P = 0.027) was observed with the 10:1 + S diet; however, DHA levels increased in both diet ratios only with a direct DHA source. Shifting towards a 2:1 diet is a valid alternative to taking EPA-containing supplements.