Global Journal of Computer Science and Technology (GJCST)
Not a member yet
1830 research outputs found
Sort by
Development of a Portable IP-Based Remote Controlled System for Mobile Robot
Get PDFThe use of Mobile Robots to interact with objects in remote locations has proved to be useful in areas not easily accessible or too dangerous for humans. Various means have been used to remotely operate or control Mobile Robots. These range from wired connection to Wireless connection like radio frequency signal and more recently internet controlled Mobile Robot using the TCP/IP protocol stack. However, the problem of remote control dependence on the Mobile Robot Platform or configuration has made it difficult to switch controllers between Mobile Robots. In this work, a portable IPbased remote control system has been designed and implemented to remove the constraint imposed by the Mobile Robot's platform in choosing the control interface. The system developed was built on three loosely coupled components working together to ensure a high degree of Control interface portability. The Mobile Robot Gateway component was used to receive and send data from the Mobile Robo
A Review of Contact Tracing Approaches for Controlling COVID-19 Pandemic
Get PDFThe year 2020 will always be in the history of mankind due to the deadly outbreak of COVID-19. Many people are already infected around the world due to the spreading of this novel coronavirus. The virus mainly replicates through close contacts, so there are no other alternatives than to keep social distance, use proper safety gear, and maintain self-quarantine. As a result, the growth of the virus has changed the lifestyle of every individual to a great extent. It is also compelling the Governments to dictate strict lock-downs of the highly affected areas, impose work-from-home approaches where applicable, enforce strict social distancing standards, and so on. Some of the countries are also using smartphonebased applications for contact tracing to track the possibly infected individuals. However, there is a lot of discussion around the world about these contact tracing applications and also about their architecture, attribute, data privacy, and so on. In this paper, we have provided a comprehensive review of these contact tracing approaches in terms of their system architecture, key attributes, and data privacy. We have also outlined a list of potential research directions that can improvise the tracing performance while maintaining the privacy of the user to a great extent
AROGYA Intelligent Health Care Application
Get PDFPeople of today pay less attention to their daily diet due to their busy lifestyles. Therefore, there is a great tendency to contract chronic non-communicable diseases. Furthermore, the lack of nutritious meals and daily exercise causes chronic non-communicable diseases in people with no age difference. In our first part, we developed to predict specialization in cardiology using symptoms. However, when we refer to a doctor, we must at least know what specialist should know based on their symptoms. In addition, there is a problem with the recipe. If the pharmacist has misread the prescription given by the doctor, patients can receive bad medications, leading to terrible side effects and even death due to careless writing by the doctor. As a solution, an application function can be proposed that will be developed in the project and the function should be able to improve the readability and intelligibility of the patient with prescription drugs. Therefore, the patient always knows the prescribed medications through the application to avoid the problem mentioned above. According to the 2016 pharmaceutical magazine, there are cholesterol and diabetic patients suffering mainly from chronic non-communicable diseases in Sri Lanka
Machine Learning Model Optimization with Hyper Parameter Tuning Approach
Get PDFHyper-parameters tuning is a key step to find the optimal machine learning parameters. Determining the best hyper-parameters takes a good deal of time, especially when the objective functions are costly to determine, or a large number of parameters are required to be tuned. In contrast to the conventional machine learning algorithms, Neural Network requires tuning hyperparameters more because it has to process a lot of parameters together, and depending on the fine tuning, the accuracy of the model can be varied in between 25%-90%. A few of the most effective techniques for tuning hyper-parameters in the Deep learning methods are: Grid search, Random forest, Bayesian optimization, etc. Every method has some advantages and disadvantages over others. For example: Grid search has proven to be an effective technique to tune hyper-parameters, along with drawbacks like trying too many combinations, and performing poorly when it is required to tune many parameters at a time. In our work, we will determine, show and analyze the efficiencies of a real-world synthetic polymer dataset for different parameters and tuning methods
Unusual Features of the SARS-CoV-2 Genome Suggesting Sophisticated Laboratory Modification as a Biological Robot Rather than Natural Evolution and Delineation of its Probable Synthetic Route
Get PDFThe COVID-19 pandemic caused by the novel coronavirus SARS-CoV-2 has led to over 4.24 million deaths worldwide and unprecedented decimation of the global economy. Despite its tremendous impact, the origin of SARS-CoV-2 has remained mysterious and controversial. The natural origin theory, although widely accepted, lacks substantial support. The alternative theory that the virus may have come from a research laboratory is, however, strictly censored on peerreviewed scientific journals. Nonetheless, SARS-CoV-2 shows biological characteristics that are inconsistent with a naturally occurring, zoonotic virus. In this report, we describe the genomic, structural, medical, and literature evidence, which, when considered together, strongly contradicts the natural origin theory. The evidence shows that SARS-CoV2 should be a laboratory product created by using bat coronaviruses ZC45 and/or ZXC21 as a template and/or backbone
A Study on Finding the Factors, Hindering the use of digital wallets among youth in Developing Countries
Get PDFA digital wallet or eWallet is a mobile base software application that securely persists consumer s payment information and passwords for numerous payment methods By using an eWallet users can complete their transactions easily and quickly with technologies such as Quick Response QR and near-field communications NFC In the modern world people use digital wallets to engage in financial and non-financial activities using the internet It stores physical financial instruments digitally and provides high availability to convenience to the user Typically to facilitate high security to those financial assets in the digital wallet of the user and to maintain reliability and availability mobile application developing organizations follow different authorization mechanisms Problem prevailing in a developing country similar to Sri Lanka is the less use of digital wallets in day to day transactions The final outcome of the research will be to find the factors which affect the use of digital wallets in Sri Lankan students who study in western province universitie
Assessing the Price Relationship and Weather Impact on Selected Pairs of Closely Related Commodities Assessing the Price Relationship and Weather Impact on Selected Pairs of Closely Related Commodities
No full textAs indicated by various works of literature, climate change has a significant impact on agricultural commodities resulting in variation between demand and supply. The research study adopted quantitative analysis for comparative analysis of price relationships for three pairs of agricultural commodities against closely related products and how weather impacts them. As an interesting comparison, we also selected a pair of non-agricultural commodities for analysis. Downloaded data for the analysis were daily historical price data for the commodities, and daily summary of weather data for precipitation and temperature for the regions were the selected commodities are most produced. Using programming languages like Python and R, we carried out exploratory data analysis using the following statistics, such as graphs, scatter plots of returns, QQ plots for normality, time series diagnostics (AC, PAC) ARIMA, correlation. An exciting part of our work is our model selection, where we used SARIMAX for regressing endogenous data, i.e., commodity prices and exogenous data weather data
Multi-Target Detection Capability of Linear Fusion Approach Under Different Swerling Models of Target Fluctuation
Get PDFIn evolving radar systems, detection is regarded as a fundamental stage in their receiving end. Consequently, detection performance enhancement of a CFAR variant represents the basic requirement of these systems, since the CFAR strategy plays a key role in automatic detection process. Most existing CFAR variants need to estimate the background level before constructing the detection threshold. In a multi-target state, the existence of spurious targets could cause inaccurate estimation of background level. The occurrence of this effect will result in severely degrading the performance of the CFAR algorithm. Lots of research in the CFAR design have been achieved. However, the gap in the previous works is that there is no CFAR technique that can operate in all or most environmental varieties. To overcome this challenge, the linear fusion (LF) architecture, which can operate with the most environmental and target situations, has been presented
Comparative Study of OpenCV Inpainting Algorithms
Get PDFDigital image processing has been a significant and important part in the realm of computing science since its inception. It entails the methods and techniques that are used to manipulate a digital image using a digital computer. It is a type of signal processing in which the input and output maybe image or features/characteristics associated with that image. In this age of advanced technology, digital image processing has its uses manifold, some major fields being image restoration, medical field, computer vision, color processing, pattern recognition and video processing. Image inpainting is one such important domain of image processing. It is a form of image restoration and conservation. This paper presents a comparative study of the various digital inpainting algorithms provided by Open CV (a popular image processing library) and also identifies the most effective inpainting algorithm on the basis of Peak Signal to Noise Ratio (PSNR), Structural Similarity Index (SSIM) and runtime metrics
P vs NP: P is Equal to NP: Desired Proof
Get PDFComputations and computational complexity are fundamental for mathematics and all computer science, including web load time, cryptography (cryptocurrency mining), cybersecurity, artificial intelligence, game theory, multimedia processing, computational physics, biology (for instance, in protein structure prediction), chemistry, and the P vs. NP problem that has been singled out as one of the most challenging open problems in computer science and has great importance as this would essentially solve all the algorithmic problems that we have today if the problem is solved, but the existing complexity is deprecated and does not solve complex computations of tasks that appear in the new digital age as efficiently as it needs. Therefore, we need to realize a new complexity to solve these tasks more rapidly and easily. This paper presents proof of the equality of P and NP complexity classes when the NP problem is not harder to compute than to verify in polynomial time if we forget recursion that takes exponential running time and goes to regress only (every problem in NP can be solved in exponential time, and so it is recursive, this is a key concept that exists, but recursion does not solve the NP problems efficiently). The papers goal is to prove the existence of an algorithm solving the NP task in polynomial running time. We get the desired reduction of the exponential problem to the polynomial problem that takes O(log n) complexity