Showing papers by "Chittagong University of Engineering & Technology published in 2020"

Cybersecurity data science: an overview from machine learning perspective

Abstract: In a computing context, cybersecurity is undergoing massive shifts in technology and its operations in recent days, and data science is driving the change. Extracting security incident patterns or insights from cybersecurity data and building corresponding data-driven model, is the key to make a security system automated and intelligent. To understand and analyze the actual phenomena with data, various scientific methods, machine learning techniques, processes, and systems are used, which is commonly known as data science. In this paper, we focus and briefly discuss on cybersecurity data science, where the data is being gathered from relevant cybersecurity sources, and the analytics complement the latest data-driven patterns for providing more effective security solutions. The concept of cybersecurity data science allows making the computing process more actionable and intelligent as compared to traditional ones in the domain of cybersecurity. We then discuss and summarize a number of associated research issues and future directions. Furthermore, we provide a machine learning based multi-layered framework for the purpose of cybersecurity modeling. Overall, our goal is not only to discuss cybersecurity data science and relevant methods but also to focus the applicability towards data-driven intelligent decision making for protecting the systems from cyber-attacks.

A review on nanofluid: preparation, stability, thermophysical properties, heat transfer characteristics and application

Abstract: Nanofluid is a suspension of nanoparticles which is promising heat transfer fluid in the heat transfer enhancement having a plethora of applications because of its superior thermal conductivity and rheological properties. This paper points out the previous studies and recent progress in the improvement of heat transfer using nanofluid. The recent progresses on preparation and enhancement of stability were reviewed. Thermophysical, heat transfer characteristics of nanofluid and different factors such as particle size, shape, surfactant, temperature, etc. on thermal conductivity were presented. The present study reveals potential applications by utilizing nanofluid such as heat exchanger, transportation cooling, refrigeration, electronic equipment cooling, transformer oil, industrial cooling, nuclear system, machining operation, solar energy and desalination, defense, etc. Few barriers and challenges were also addressed. Finally, the challenges and further research opportunities were presented.

BehavDT: A Behavioral Decision Tree Learning to Build User-Centric Context-Aware Predictive Model

Abstract: This paper formulates the problem of building a context-aware predictive model based on user diverse behavioral activities with smartphones. In the area of machine learning and data science, a tree-like model as that of decision tree is considered as one of the most popular classification techniques, which can be used to build a data-driven predictive model. The traditional decision tree model typically creates a number of leaf nodes as decision nodes that represent context-specific rigid decisions, and consequently may cause overfitting problem in behavior modeling. However, in many practical scenarios within the context-aware environment, the generalized outcomes could play an important role to effectively capture user behavior. In this paper, we propose a behavioral decision tree, “BehavDT” context-aware model that takes into account user behavior-oriented generalization according to individual preference level. The BehavDT model outputs not only the generalized decisions but also the context-specific decisions in relevant exceptional cases. The effectiveness of our BehavDT model is studied by conducting experiments on individual user real smartphone datasets. Our experimental results show that the proposed BehavDT context-aware model is more effective when compared with the traditional machine learning approaches, in predicting user diverse behaviors considering multi-dimensional contexts.

Fabry Perot laser properties with high pump lasers for upgrading fiber optic transceiver systems

Abstract: The work has outlined the Fabry Perot laser properties with high pump lasers for upgrading fiber optic transceiver systems. The physical structures of the light source are taken into account such as active layer length and active layer width. High pump laser is used for providing strength to the signal through the transmission/reception stages. Peak and minimum signal power levels are measured in the spectral frequency domain and time domain. Signal amplitude level margin is also measured with the optimum physical parameters of the light source. The optimum operation system performance efficiency is achieved with an active layer length of 0.06 cm, and active layer width suitable is 1.5 × 10−4 cm.

Cyber intrusion detection using machine learning classification techniques

Abstract: As the alarming growth of connectivity of computers and the significant number of computer-related applications increase in recent years, the challenge of fulfilling cyber-security is increasing consistently. It also needs a proper protection system for numerous cyberattacks. Thus, detecting inconsistency and attacks in a computer network and developing intrusion detection system (IDS) that performs a potential role for cyber-security. Artificial intelligence, particularly machine learning techniques, has been used to develop a useful data-driven intrusion detection system. In this paper, we employ various popular machine learning classification algorithms, namely Bayesian Network, Naive Bayes classifier, Decision Tree, Random Decision Forest, Random Tree, Decision Table, and Artificial Neural Network, to detect intrusions due to provide intelligent services in the domain of cyber-security. Finally, we test the effectiveness of various experiments on cyber-security datasets having several categories of cyber-attacks and evaluate the effectiveness of the performance metrics, precision, recall, f1-score, and accuracy.

AquaVision: Automating the detection of waste in water bodies using deep transfer learning

Abstract: Water pollution is one of the serious threats in the society. More than 8 million tons of plastic are dumped in the oceans each year. In addition to that beaches are littered by tourists and residents all around the world. It is no secret that the aquatic life ecosystem is at a risk and soon the ratio of plastic/waste to the marine life particulary fish will be 1:1. Hence, in this paper, we have proposed a dataset known as AquaTrash which is based on TACO data set. Further, we have applied proposed state-of-the-art deep learning-based object detection model known as AquaVision over AquaTrash dataset. Proposed model detects and classifies the different pollutants and harmful waste items floating in the oceans and on the seashores with mean Average Precision (mAP) of 0.8148. The propose method localizes the waste object that help in cleaning the water bodies and contributes to environment by maintaining the aquatic ecosystem.

Recently synthesized (Ti1−xMox)2AlC (0 ≤ x ≤ 0.20) solid solutions: deciphering the structural, electronic, mechanical and thermodynamic properties via ab initio simulations

Abstract: The structural, electronic, mechanical and thermodynamic properties of (Ti1−xMox)2AlC (0 ≤ x ≤ 0.20) were explored using density functional theory. The obtained lattice constants agree well with the experimental values. The electronic band structure confirms the metallic nature. Strengthening of covalent bonds due to Mo substitution is confirmed from the study of band structure, electronic density of states and charge density mapping. The elastic constants satisfy the mechanical stability criteria. Strengthening of covalent bonds leads to enhanced mechanical properties. (Ti1−xMox)2AlC compounds are found to exhibit brittle behavior. The anisotropic nature of (Ti1−xMox)2AlC is revealed from the direction dependent Young's modulus, compressibility, shear modulus and Poisson's ratio as well as the shear anisotropic constants and the universal anisotropic factor. The Debye temperature, minimum thermal conductivity, Gruneisen parameter and melting temperature of (Ti1−xMox)2AlC have been calculated for different Mo contents. Our calculated values are compared with reported values, where available.

A Systematic Review on the Use of AI and ML for Fighting the COVID-19 Pandemic

Abstract: Artificial intelligence (AI) and machine learning (ML) have caused a paradigm shift in healthcare that can be used for decision support and forecasting by exploring medical data. Recent studies have shown that AI and ML can be used to fight COVID-19. The objective of this article is to summarize the recent AI- and ML-based studies that have addressed the pandemic. From an initial set of 634 articles, a total of 49 articles were finally selected through an inclusion-exclusion process. In this article, we have explored the objectives of the existing studies (i.e., the role of AI/ML in fighting the COVID-19 pandemic); the context of the studies (i.e., whether it was focused on a specific country-context or with a global perspective; the type and volume of the dataset; and the methodology, algorithms, and techniques adopted in the prediction or diagnosis processes). We have mapped the algorithms and techniques with the data type by highlighting their prediction/classification accuracy. From our analysis, we categorized the objectives of the studies into four groups: disease detection, epidemic forecasting, sustainable development, and disease diagnosis. We observed that most of these studies used deep learning algorithms on image-data, more specifically on chest X-rays and CT scans. We have identified six future research opportunities that we have summarized in this paper. Impact Statement: Artificial intelligence (AI) and machine learning(ML) methods have been widely used to assist in the fight against COVID-19 pandemic. A very few in-depth literature reviews have been conducted to synthesize the knowledge and identify future research agenda including a previously published review on data science for COVID-19 in this article. In this article, we synthesized reviewed recent literature that focuses on the usages and applications of AI and ML to fight against COVID-19. We have identified seven future research directions that would guide researchers to conduct future research. The most significant of these are: develop new treatment options, explore the contextual effect and variation in research outcomes, support the health care workforce, and explore the effect and variation in research outcomes based on different types of data.

Periodic and rogue waves for Heisenberg models of ferromagnetic spin chains with fractional beta derivative evolution and obliqueness

Abstract: The nonlinear Schrodinger equation (NLSE) in (2 + 1) dimensions with beta derivative evolution is considered here to study nonlinear coherent structures for Heisenberg models of ferromagnetic spin ...

Shoreline change assessment using geospatial tools: a study on the Ganges deltaic coast of Bangladesh

Abstract: The Ganges deltaic coast of Bangladesh experiences an incessant movement over the time. Understanding the shoreline movement of this alluvial delta and a suitable method to calculate the rate of change poses a challenge for this highly dynamic coast having erosion and accretion. Using GIS and multi temporal LANDSAT images, the study investigated the positional change of the Ganges deltaic shoreline for the period of 1977–2017. LANDSAT images were radiometrically corrected and a spectral index i.e., normalized difference water index (NDWI) was applied to differentiate water and land features. A histogram based Otsu’s Binary thresholding method along with image based visual interpretation was used to extract the shorelines. Net changes of shoreline position were statistically calculated using three different techniques, namely; End Point Rate (EPR), Linear Regression Rate (LRR) and Weighted Linear Regression (WLR). A comparison between the techniques was also made to choose and evaluate the suitable statistical technique to estimate the rate of shoreline change for this alluvial delta. Analyses showed that LRR technique had less positional uncertainty in compare to EPR and WLR, although at a particular transect the techniques were closely correlated. The EPR, WLR and LRR technique showed that the shoreline is experiencing landward movement (erosion) with an average rate of 0.62 m/yr, 0.96 m/yr and 0.27 m/yr respectively. Moreover, a high erosion rate of 5 m/yr at the mangrove forest area of the GDC is a great concern for the existence of the mangrove forest. During 1977–2017, an overall 6.29 sq. km land area has been lost although significant land depositions were observed at the river estuaries. This study revealed the spatiotemporal trend of shoreline position of the Ganges deltaic coast and that would be beneficial for the coastal management and planning of the region.

Analysis of a gold coated plasmonic sensor based on a duplex core photonic crystal fiber

Abstract: An exploration of a gold-coated dual-core with hexagonally arranged circular air holes SPR-PCF sensor is presented in this paper. Chemically inactive and stable material gold is provided as plasmonic material in this design. Placement of the gold layer outside of the PCF structure simplifies the detection process of this sensor and also assures fabrication feasibility. The performance of this sensor is investigated in terms of wavelength sensitivity, amplitude sensitivity, resolution, sensor length and linearity response by using the Finite Element Method (FEM) based COMSOL Multiphysics software. Maximum wavelength sensitivity is found to be 10,700 nm/RIU for analyte sensing range between 1.39 to 1.40 by using wavelength interrogation method as well as maximum amplitude sensitivity reaches about 1770 RIU-1 for analyte RI 1.39 by using amplitude interrogation method. Besides, the sensor's resolution is found to be 9.34×10-6. The outcome of alternating structural parameters such as gold layer thickness, pitch, air hole diameter, different shapes of air hole in the core is also discussed as well. The reported sensor might be a fruitful aspirant in the field of biological sample detection, organic chemical sensing and biomolecule recognition.

Statistical evaluation of selected air quality parameters influenced by COVID-19 lockdown

Abstract: Air pollution has become a serious concern for its potential health hazard, however, often got less attention in developing countries, like Bangladesh It is expected that worldwide lockdown due to COVID-19 widespread cause reduction in environmental pollution in particularly the air pollution: however, such changes have been different in different places In Chittagong, a city scale lockdown came in force on 26 March 2020, a week after when first three cases of COVID-19 have been reported in Bangladesh This study aims to statistically evaluate the effects of COVID-19 lockdown (26 March to 26 April 2020) on selected air quality pollutants and air quality index s The daily average concentrations of air pollutants PMsub10/sub, PMsub2 5/sub, NOsub2/sub, SOsub2/sub and CO of Chittagong city during COVID-19 lockdown were statistically evaluated and were compared with dry season data averaging over previous 8 years (2012 to 2019) During lockdown, except NOsub2/sub, all other pollutants studied showed statistically significant decreasing trend During the COVID-19 shutdown notable reduction of 40%, 32% and 13% compared to the daily mean concentrations of these previous dry season were seen for PMsub2 5/sub, PMsub10/sub and NOsub2/sub, respectively The improvement in air quality index value was found as 26% in comparison to the previous dry season due to less human activities in COVID-19 shutdown The factor analysis showed that AQI in Chittagong city is largely influenced by PMsub10/sub and PMsub2 5/sub during COVID-19 shutdown The lesson learnt in this forced measure of lockdown is not surprising and unexpected It is rather thought provoking for the decision makers to tradeoff the tangible air quality benefits with ongoing development strategies' that was often overlooked directly or indirectly

Impact of LULC Changes on LST in Rajshahi District of Bangladesh: A Remote Sensing Approach

Abstract: Urbanization has been contributing more in global climate warming, with more than 50% of the population living in cities. Rapid population growth and change in land use / land cover (LULC) are closely linked. The transformation of LULC due to rapid urban expansion significantly affects the functions of biodiversity and ecosystems, as well as local and regional climates. Improper planning and uncontrolled management of LULC changes profoundly contribute to the rise of urban land surface temperature (LST). This study evaluates the impact of LULC changes on LST for 1997, 2007 and 2017 in the Rajshahi district (Bangladesh) using multi-temporal and multi-spectral Landsat 8 OLI and Landsat 5 TM satellite data sets. The analysis of LULC changes exposed a remarkable increase in the built-up areas and a significant decrease in the vegetation and agricultural land. The built-up area was increased almost double in last 20 years in the study area. The distribution of changes in LST shows that built-up areas recorded the highest temperature followed by bare land, vegetation and agricultural land and water bodies. The LULC-LST profiles also revealed the highest temperature in built-up areas and the lowest temperature in water bodies. In the last 20 years, LST was increased about 13oC. The study demonstrates decrease in vegetation cover and increase in non-evaporating surfaces with significantly increases the surface temperature in the study area. Remote-sensing techniques were found one of the suitable techniques for rapid analysis of urban expansions and to identify the impact of urbanization on LST. Article history Received: 08 August 2019 Revised: 08 December 2019 Accepted: 15 December 2019

Design of a liquid sensing photonic crystal fiber with high sensitivity, bireferingence & low confinement loss

Abstract: This paper represents a Photonic Crystal Fiber (PCF) based sensor structure with concurrently high sensitivity, high birefringence and low confinement loss for liquid sensing applications. We explored the efficiency of the constructed PCFs for Water to be sensed as a liquid sample. The numerical analysis of the proposed structure is performed using the full Finite Element Method (FEM). To minimize the fabrication complexity, circular air holes have been chosen instead of elliptical holes in the core region. The substantial analysis is described at a broad spectrum of wavelengths (1.3 μm–2 μm) and the effect of different design parameters of proposed structures has been studied very sincerely. According to FEM numerical results, the designed PCF sensor offers considerable performance in terms of sensitivity is 49.13% as well as birefringence is 0.008. The suggested framework can be used extremely in the area of bio-sensing studies and commercial applications.

The topology and robustness of two Dirac cones in S-graphene: A tight binding approach.

Abstract: Present work reports an elegant method to address the emergence of two Dirac cones in a non-hexagonal graphene allotrope S-graphene (SG). We have availed nearest neighbour tight binding (NNTB) model to validate the existence of two Dirac cones reported from density functional theory (DFT) computations. Besides, the real space renormalization group (RSRG) scheme clearly reveals the key reason behind the emergence of two Dirac cones associated with the given topology. Furthermore, the robustness of these Dirac cones has been explored in terms of hopping parameters. As an important note, the Fermi velocity of the SG system (vF $$\simeq $$ c/80) is almost 3.75 times that of the graphene. It has been observed that the Dirac cones can be easily shifted along the symmetry lines without breaking the degeneracy. We have attained two different conditions based on the sole relations of hopping parameters and on-site energies to break the degeneracy. Further, in order to perceive the topological aspect of the system we have obtained the phase diagram and Chern number of Haldane model. This exact analytical method along with the supported DFT computation will be very effective in studying the intrinsic behaviour of the Dirac materials other than graphene.

Improved gray wolf optimization algorithm for solving placement and sizing of electrical energy storage system in micro-grids.

Abstract: Micro-grids consist of distributed power generation systems (DGs), distributed energy storage devices (DSs), and loads. Micro-grids are small-scale networks at low voltage levels that are use to provide thermal and electrical loads of small locations where there is no access to the main electrical grid. Given the environmental and economic issues for these areas, micro-grids can be a good solution for energy production. In this paper, determining the size and location of optimal electrical energy storage systems is presented. In other side, a new method based on the cost benefit analysis for optimal sizing of an energy storage system in a microgrid (MG) is proposed. The uncertainties associated with renewable energy sources and the occurrence of defects in the grid connection network and the effect of the contribution of load responses in a micro-grid are taken into account. The combined system consists of wind turbines and fuel cells. Basically, wind power is not definitively available. The new proposed method is based on two-stage randomization design (TSRD) for modeling the effect of wind power uncertainty so that the predicted wind energy error is considered as the main random parameter in the model. A standard probability distribution function is used to represent the error variations. Given the continuity of the mentioned function, the probability error function is extracted using the new discrete method and a certain number of scenarios with a certain probability. Finally, the problem has been transformed into an optimization problem, and a gray wolf optimization (GWO) algorithm has been used to solve it. In the proposed developed model based on local and global search, the algorithm tries to reach the final result in the shortest possible time and with the most precision. The results of the simulation show the efficiency of the proposed method in solving the micro-grid problem.

Machine Learning Approach to Model Rock Strength: Prediction and Variable Selection with Aid of Log Data

Abstract: Comprehensive knowledge and analysis of in situ rock strength and geo-mechanical characteristics of rocks are crucial in hydrocarbon and mineral exploration stage to maximize wellbore performance, maintain wellbore stability, and optimize hydraulic fracturing process Due to the high cost of laboratory-based measurements of rock mechanics properties, the log-based prediction is a viable option Nowadays, the machine learning tools are being used for estimation of the in situ rock properties using wireline log data This paper proposes a machine learning approach for rock strength (uniaxial compressive strength) prediction The main objectives are to investigate the performance of data-driven predictive model in determining this vital parameter and to select features of predictor log variables in the model The backpropagation multilayer perception (MLP) artificial neural network (ANN) with Levenberg–Marquardt training algorithm as well as the least squares support vector machine (LS-SVM) with coupled simulated annealing (CSA) optimization technique is employed to develop the dynamic data-driven models Capturing nonlinear, high dimensional, and complex nature of real field log data, the rock strength models’ performances are evaluated using statistical criteria to ensure concerning the model reliability and accuracy The model predictions are compared and validated against the measured values as well as the results obtained from existing log-based correlations Both the MLP-ANN and the CSA-based LS-SVM connectionist strategies are able to predict the rock strength so that there is a very good match between the model results and corresponding measured values The input log parameters are ranked based on their contributions in prediction performance The acoustic travel time and gamma ray are found to have the highest relative significance in estimating rock strength New correlations are also developed to obtain the in situ rock strength of the siliciclastic sedimentary rocks using the most important log parameters such as dynamic sonic slowness, formation electron density, and shalyness effect The developed correlations can be used to obtain quick estimation of dynamic uniaxial compressive strength profile using wireline logging data, instead of static data from the surface measurements or laboratory data It is expected that the proposed models and tools will enable oil and gas engineers to better predict rock strength and thus enhance wellbore performance