Browsing by Author "Hossain, Arman"

Now showing 1 - 5 of 5

Diabetics Prediction System Using Machine Learning Algorithms “A Case Study Among Female Patient
(DAFFODIL INTERNATIONAL UNIVERSITY, 2024-01-29) Hossain, Arman
Millions of individuals worldwide suffer with diabetes mellitus, a chronic metabolic ailment, and early diagnosis is essential to controlling the condition and avoiding complications. The development of precise prediction systems for a variety of medical illnesses has shown encouraging outcomes in recent years thanks to machine learning techniques. The goal of this study is to employ machine learning to create a diabetic prediction system that is specifically created for females. The planned study will examine a dataset that includes detailed health records for females, together with data on their lifestyle choices, medical histories, and demographics. The dataset will be processed using a suitable machine learning technique to produce a predictive model. The accuracy, recall, f-measure and precision of various algorithms will be compared in order to determine which is the most efficient. I'm using the Pima Indian Diabetes Database data set from Kaggle.com for this research. In this work I have used some machine learning algorithms which is logistic regression, Naive-bayes, support vector machine (SVM), decision tree and Random Forest. After preprocessing the data set and applying this algorithms the prediction performance of diabetics disease analysis shows that random forest obtained the uppermost performance with the accuracy of 85% and decision tree has achieved the second highest accuracy which is 81%”
OTA Management System of Tripey
(Independent University, Bangladesh, 2022-08-14) Hossain, Arman
This project includes a thorough analysis of the tour and travel management system. The main goals of this mobile app are to provide users with information about travel packages that have the best amenities and the most recent deals. It will be incredibly simple to search because you may get the information you need with only one tap. A consumer illustrates how difficult and time-consuming it is to look for a variety of packages across important websites, in contact with travel agents, and through other means, all of which involve passive methods. This mobile app will help travelers choose the best vacation package out of all the available options by providing them with useful details about the destinations they want to visit, including an image, a hotel's amenities, a Google map, a transportation option, and a description. This Online Travel Agency management system for tours and travel will be useful for the tourism industry
ShielDroid: a hybrid ML and DL approach for real-time malware detection system in Android
(BRAC University, 2021-09) Ahmed, Md Faisal; Biash, Zarin Tasnim; Shakil, Abu Raihan; Ryen, Ahmed Ann Noor; Hossain, Arman; Hossain, Muhammad Iqbal; Ashraf, Faisal Bin
Due to the rapid development of the advanced world of technology, there is a high increase in devices such as smartphones and tablets, which increase the number of applications used. Though an application has to pass the malware detection test before appearing in the play store, many applications successfully get trusted and accepted even though they contain malicious software variants that are challenging to detect. The application requires physical execution to see these malicious contents, which get undetected during the rst screening test. Due to the physical implementation of the application, it may be too late to undo the malware's damage. In this work, the usage of real-time Android malware detection analyzing Android applications to detect and swiftly distinguish complex malware has been discussed. This work focuses on the use of dynamic algorithms implemented by hybrid detection techniques of Android malware. After ltrating the collected dataset, the process of separation between harmful and benign apps is discussed. Then summarization and evaluation of the various techniques and classi cation algorithms employed have been discussed, identifying the best-suited method that gives the most accurate result in a minimum amount of time. The best way to reach the target is a hybrid Random Forest, and Multilayer perceptron network, where the overall accuracy achieved was 97.5% with an execution time of 22.945 seconds. An Android application, namely,\Shield: Malware Scanner", was developed using Java in determining if malware is present in an application. If there is any malware, it detects the type of malware and advises the user on securing their data and privacy and recovering from it.
Study of Energy Detection Based On Spectrum Sensing in Cognitive Radio
(East West University, 4/19/2016) Hossain, Arman; Hasan, Mirza Hasibul
To cope up with the demand of user’s traffic and availability of spectrum of wireless network, cognitive radio can pave the way to optimize them. Cognitive radio networks (CRN) deals with two types of users: one is primary users (PUs) or licensed users who have the higher priority to use a specific part of the spectrum and the other one is secondary users (SUs) or unlicensed users who have less priority of using spectrum. SUs use the spectrum in such a way that they do not cause interference with PUs. In cognitive radio network, different techniques (i.e. Energy detection, Cyclostationary Feature Detection, Matched filter, Cooperative Sensing, Waveform Based Sensing etc.) have been used to improve the performance of channel sensing. However, some of these techniques are vulnerable to malicious users and may not correctly follow sensing order of saving energy and/or time or to lunch denial of service attacks against the network. The performance of such a network solely depends on the accuracy of spectrum sensing, i.e. minimization of false alarm and maximization of detection. These two parameters are heavily affected by the presence of malicious users inside the network. In this paper we have studied cooperative sensing energy detection method of cognitive radio network and show the relation of bit error and signal to noise ratio (SNR) in Rayleigh fading channel. Also analyze the probability of miss alarm and probability of detection for different number of users.
Waste to Energy : Potential of using activated carbon and rice husk ash in primary cells
(Department of Mechanical and Production Engineering (MPE),Islamic University of Technology(IUT), Board Bazar, Gazipur, Bangladesh, 2021-03-30) Hossain, Arman; Maroof, Irteza; Mahmud, Md. Fahim
Rapid growth of population has brought up some serious concerns, such as waste management and excessive use of fossil fuels causing pollution and negative impact on climate change. A large portion of the waste is organic waste, which can be turned into Activated Carbon (AC) and AC can be used to generate electricity by using it as a cathode catalyst in metal-air cell. Aluminum-air cell has the potential of becoming vital in energy-storage application in future because of its high theoretical energy density, even higher than commonly used lithium-ion batteries. However, it is not used widely because the cost of air cathode catalyst and metal anode is high. But if we replace the catalyst by activated carbon obtained from waste or a mixture of activated carbon with other waste (rice husk ash) and use recycled aluminum foil as anode then the production cost might be feasible for vast use of this type of cells. Our objective is to utilize some common and organic waste in the process of fabricating Al-air battery suitable for small and day-to-day usage reducing production cost and limitations. In this paper, we have focused on the feasibility of using activated carbon and rice husk mixture as air cathode catalyst of Al-air cell; and the observations were interesting. We prepared 11 samples of mixture of rice husk and AC in different ratios and initially found the best results to be from 0.68-0.72V, which increases by 8-20% measuring each sample after 3 days. In this study, graphite cathode of a dry cell is replaced too by mixture of AC and RHA. Voltage drop is quite negligible for the mixture of 10% RHA. And, voltage is similar to the new battery in case of 100% activated carbon as cathode. If considered the environmental effect, using recycled activated carbon and rice husk ash will decrease the pollution