Inquiry (E-Journal - Faculty of Business and Administration, International University of Sarajevo)
Not a member yet
209 research outputs found
Sort by
Heart Attack Prediction Model Based on Feature Selection and Decision Tree Approaches
Full text linkThe purpose of this study is creating a machine learning based model is to predict heart attacks is to improve the capacity to anticipate the occurrence of this dangerous medical condition. It is feasible to find significant and linked variables that may cause heart attacks by using the decision tree as a tool for medical data analysis. The system analyzes clinical data using artificial intelligence techniques to find patterns that might suggest the possibility of a heart attack. The advantage is early disease detection and prediction, which allows the medical staff to better plan treatment and take preventative action. This kind of system can aid in enhancing patient care and lowering the likelihood that. Throuought the study, two paths will be examined, the first one is applying machine learning algorithms without applying feature selection, and the second one with feature selection process. Three mainly feature selection algorithms will be examined to find the most correlated features that affect the heart attack. The model will examine six machine learning decision tree algorithms namely (decision stump, hoeffding tree, j48, LMT, random forest, and rep tree) to find the accurate algorithm in prediction. The results show that LMT have the accurate prediction accuracy with 82.5%
Investigating the Impact of Hidden Speed Check Radar on Road Accident in Oman
Full text linkThe road is the place that people use to move in their cars, but it is a double-edged sword, as it is one of the reasons that lead to the loss of lives. so, our research focuses on studying how the hidden speed camera impact on accidents either increases or decreases. The goal is to target a group of people who can drive through the questionnaire that they will answer, which contains several questions related to whether hidden cameras are the biggest cause of accidents or not. The aim of this research is to protect the general population and find out whether a hidden camera may be the cause of accidents or not, and to educate them on the importance of adhering to the speed limit and paying attention when driving, whether the radar is hidden or visible. Determine whether hidden speed cameras have a positive or negative impact on traffic accidents. Study the role of speed cameras in reducing speed. It is also to study the relationship between hidden cameras and reducing accidents and evaluate the safety and effectiveness of speed cameras. The importance of the research lies in educating people to reduce accidents that are caused by speeding and knowing if the speed camera that is hidden so that it can catch reckless drivers may cause accidents or not through the graphs that will show whether the hypothesis was developed, which is that if the hidden speed camera is a reason for deterring reckless drivers, then it may reduce accidents because it forces the drivers. It is possible that this study shows the truth that the hidden speed control camera has a significant impact, as it causes accidents because the camera light may suddenly turn on, causing the driver to deteriorate, resulting in an accident
Brain Stroke Prediction Model Based SMOTE and Machine Learning Algorithms
Full text linkA brain stroke is a critical medical emergency condition that causes disability and death. The pre-diagnosis of this case can reduce the complications and problems that affect the brain as a result of being affected by the complications that occur during the injury. This study lists an analysis process on a brain stroke dataset using the KNIME tool, which provides a set of different machine learning components such as random forest, Decision Tree Learner, Gradient Boosted Trees Learner, and Logistic Regression algorithms. The problem of imbalanced data will be handled as part of data preprocessing. The factors that affect the brain stroke will be explored based on feature selection approaches such as forward feature selection, backward feature elimination, genetic algorithms, and random. The aim is to build a model that helps doctors diagnose the disease accurately based on the results we obtained from the study and analysis. The results showed that logistic regression outperformed the other algorithms after applying the algorithm with forward feature selection and backward feature elimination
Analysis of Priority Queueing System with Working Breakdown,Vacation and Vacation Interruption under Random Environment
Full text linkstudy priority queueing system consisting of working breakdown, repair, single vacation and vacation interruption. At a breakdown instant during the busy period, without stop the servie, the server provides service at slower rate for the current customer. After completion of priority service, depending on the environment, the server can choose type I or type II vacation. Both types of vacation are considered as a single vacation. On completion of type I vacation if the server finds thepriority queue is empty, he has the option to take type II vacation. During the type II vacation, when priority customers arrive the server has the option to interrupt the vacation. We use the established norm which is the corresponding steady state results for time dependent probability generating functions are obtained. Along with that, the expected waiting time for theexpected number of customers for both high and low priority queues are computed. Numerical results along with the graphical representations are shown elaborately
High Frequency Compact Microstrip Low Pass Filter
Full text linkThe purpose of this work is to discuss the design and analysis of a low pass filter. The research of a low pass filter with a cut-off frequency of 11.45GHz and an area of 13x6.3mm2 is proposed in this paper. Initially, low pass prototype filter was designed with ground layer. The proposed low pass filter is miniature and low cost because it is mounted on FR4 substrate. The FR4 substrate has thickness of 1.55mm. By using three main stubs that are placed symmetrically around X and Y axis is obtained. The cutoff frequency could be changed by tuning filterโs dimensions. The simulations with this design are simulated using Sonnet software and simulation results are reported
A cryptographic technique applying Laplace Transform and Exponetial Function
Full text linkSecured Information transformation from one person to another is very much essential in this world of internet. In electronic communications such as system security, smart card, mobile communications etc, the data are stored in the various applications used in them. While using all the applications, we give permission to access all the information. Cryptography is based on transformation of multiple rounds of transformation of messages in the form of plain text as input into encrypted text message. Through suitable mathematical technique, secrecy of the information is maintained. This paper proposes a Laplace and its inverse transform technique with a suitable function for encryption and decryption of a message
In-Dominating Number and Out-Dominating Number of Unidirectional Perfect Binary Trees
Full text linkA diverging unidirectional perfect binary tree is a unidirectional perfect binary tree in which every internal node has exactly outdegree two and indegree 1 and all the leaf nodes are at the same level. A converging unidirectional perfect binary tree is a unidirectional perfect binary tree in which every internal node has exactly indegree two and outdegree 1 and all the leaf nodes are at the same level. In this paper, we study the out-domination number and in-domination number of the unidirectional perfect binary trees
An Overview of Linear Algebra in Image Processing
Full text linkLinear algebra plays a crucial role in image processing, providing a powerful mathematical framework for manipulating and analyzing digital images. This article provides a brief introduction to basic linear algebra concepts and techniques commonly used in image processing, including matrix operations, linear transformations. It discuss how these tools can be used to perform operations such as image filtering, compression, registration, and segmentation, and illustrate their use with concrete examples. This work highlight some of the key challenges and open problems in the field of linear algebra and image processing, and suggest future directions for research in this area. This overview aims to provide a broad perspective on the role of linear algebra in image processing, and to stimulate further research in this exciting and rapidly evolving field. For numerical simulation, MATLAB is utilized
Total and Edge Domination on Anti Fuzzy Line Graph
Full text linkIn this paper, the total dominating set, edge dominating set and domination number (TDN & EDN) for anti fuzzy line graph is computed by maxima method algorithm (MMA). This calculation is executed using strong adjacency matrix of anti fuzzy graphs and its line graph
Bulk queueing system with starting failure and single vacation
Full text linkTo analyze an M^([X])/G(a,b)/1 queue with starting failure, repair and single vacation. This type of queueing model has a wide range of applications in production/manufacturing systems. Important performance measures and stability conditions are obtained. Further, we discuss some particular cases. Finally, numerical results of the proposed model have been derived