Publications
Reinforcement learning algorithms and applications in healthcare and robotics: a comprehensive and systematic review
Apr 11, 2024Journal Sensors
Publisher MDPI
DOI https://doi.org/10.3390/s24082461
Issue 8
Volume 24
Reinforcement learning (RL) has emerged as a dynamic and transformative paradigm in artificial intelligence, offering the promise of intelligent decision-making in complex and dynamic environments. This unique feature enables RL to address sequential decision-making problems with simultaneous sampling, evaluation, and feedback. As a result, RL techniques have become suitable candidates for developing powerful solutions in various domains. In this study, we present a comprehensive and systematic review of RL algorithms and applications. This review commences with an exploration of the foundations of RL and proceeds to examine each algorithm in detail, concluding with a comparative analysis of RL algorithms based on several criteria. This review then extends to two key applications of RL: robotics and healthcare. In robotics manipulation, RL enhances precision and adaptability in tasks such as object grasping and autonomous learning. In healthcare, this review turns its focus to the realm of cell growth problems, clarifying how RL has provided a data-driven approach for optimizing the growth of cell cultures and the development of therapeutic solutions. This review offers a comprehensive overview, shedding light on the evolving landscape of RL and its potential in two diverse yet interconnected fields.
Comprehensive systematic review of information fusion methods in smart cities and urban environments
Feb 21, 2024Journal Information Fusion
Publisher Elsevier
DOI https://doi.org/10.1016/j.inffus.2024.102317
Volume 107
Smart cities result from integrating advanced technologies and intelligent sensors into modern urban infrastructure. The Internet of Things (IoT) and data integration are pivotal in creating interconnected and intelligent urban spaces. In this literature review, we explore the different methods of information fusion used in smart cities, along with their advantages and challenges. However, there are notable challenges in managing diverse data sources, handling large data volumes, and meeting the near-real-time demands of various smart city applications. The review aims to examine smart city applications in detail, incorporating quality evaluation and information fusion techniques and identifying critical issues while outlining promising research directions. In order to accomplish our goal, we conducted a comprehensive search of literature and applied selective criteria. We identified 59 recent studies addressing machine learning (ML) and deep learning (DL) techniques in smart city applications. These studies were obtained from various databases such as ScienceDirect (SD), Scopus, Web of Science (WoS), and IEEE Xplore. The main objective of this study is to provide more detailed insights into smart cities by supplementing existing research. The word cloud visualisation of machine learning/deep learning and information fusion in smart cities papers shows a diverse landscape, covering both technical aspects of artificial intelligence and practical applications in urban settings. Apart from technical exploration, the study also delves into the ethical and privacy implications arising in smart cities. Moreover, it thoroughly examines the challenges that must be addressed to realise this urban revolution's potential fully.
Classification and analysis of the MNIST dataset using PCA and SVM algorithms
Mar 27, 2023Journal Vojnotehnički glasnik / Military Technical Courier
Publisher University of Defence in Belgrade (Military Academy)
DOI https://doi.org/10.5937/vojtehg71-42689
Issue 02
Volume 71
Introduction/purpose: The utilization of machine learning methods has become indispensable in analyzing large-scale, complex data in contemporary data-driven environments, with a diverse range of applications from optimizing business operations to advancing scientific research. Despite the potential for insight and innovation presented by these voluminous datasets, they pose significant challenges in areas such as data quality and structure, necessitating the implementation of effective management strategies. Machine learning techniques have emerged as essential tools in identifying and mitigating these challenges and developing viable solutions to address them. The MNIST dataset represents a prominent example of a widely-used dataset in this field, renowned for its expansive collection of handwritten numerical digits, and frequently employed in tasks such as classification and analysis, as demonstrated in the present study. Methods: This study employed the MNIST dataset to investigate various statistical techniques, including the Principal Components Analysis (PCA) algorithm implemented using the Python programming language. Additionally, Support Vector Machine (SVM) models were applied to both linear and non-linear classification problems to assess the accuracy of the model. Results: The results of the present study indicate that while the PCA technique is effective for dimensionality reduction, it may not be as effective for visualization purposes. Moreover, the findings demonstrate that both linear and non-linear SVM models were capable of effectively classifying the dataset. Conclusion: The findings of the study demonstrate that SVM can serve as an efficacious technique for addressing classification problems.
Prediction of COVID-19 disease severity using machine learning techniques
Apr 1, 2022Journal Bulletin of Electrical Engineering and Informatics
Publisher Institute of Advanced Engineering and Science
DOI https://doi.org/10.11591/eei.v11i2.3272
Issue 2
Volume 11
A terrifying spread of COVID-19 (which is also known as severe acute respiratory syndrome coronavirus 2 or SARS-COV-2) led scientists to conduct tremendous efforts to reduce the pandemic effects. COVID-19 has been announced pandemic discovered in 2019 and affected millions of people. Infected people may experience headache, body pain, and sometimes difficulty in breathing. For older people, the symptoms can get worse. Also, it can cause death because of the huge effect on some parts of the human body, particularly for those who have chronic diseases like diabetes. Machine learning algorithms are applied to patients diagnosed with Corona Virus to estimate the severity of the disease depending on their chronic diseases at an early stage. Chronic diseases could raise the severity of COVID-19 and that is what has been proved in this paper. This paper applies different machine learning techniques such as random forest, decision tree, linear regression, binary search, and k-nearest neighbor on Mexican patients’ dataset to find out the impact of lifelong illnesses on increasing the symptoms of the virus in the human body. Besides, the paper demonstrates that in some cases, especially for older people, the virus can cause inevitable death.
Designing a secure campus network and simulating it using Cisco packet tracer
Jul 1, 2021Journal Indonesian Journal of Electrical Engineering and Computer Science
Publisher Institute of Advanced Engineering and Science
DOI http://doi.org/10.11591/ijeecs.v23.i1.pp479-489
Issue 1
Volume 23
The network is a massive part of life today. It participates not only on one side of life but in nearly every station, especially in educational organizations. The key aim of education is to share data and knowledge, making the network important for education. In particular, it is essential to ensure the exchange of information; thus, no one can corrupt it. To safe and trustworthy transfers between users, integrity and reliability are crucial questions in all data transfer problems. Therefore, we have developed a secure campus network (SCN) for sending and receiving information among high-security end-users. We created a topology for a campus of multi networks and virtual local area networks (VLANs’) using cisco packet tracer. We also introduced the most critical security configurations, the networking used in our architecture. We used a large number of protocols to protect and accommodate the users of the SCN scheme.
Execution Of A Smart Street Lighting System For Energy Saving Enhancement
Aug 1, 2021Journal Bulletin of Electrical Engineering and Informatics
Publisher Institute of Advanced Engineering and Science
DOI https://doi.org/10.11591/eei.v10i4.2924
Issue 4
Volume 10
In many countries, particularly, third world countries. The common issue is saving energy. Thats why smart systems considered now primary for life requirements. This work aims to solve the energy saving problem. We prepared a street model that contains several lampposts on both sides of the street; we placed three IR sensors between the lampposts alongside the street. The IR sensors are connected to the controller (in this work we used Arduino UNO). The controller takes the signal from the IR sensor, and then it sends the command to the lamppost to turn on or off. Depending on the number of cars passed,(we took a sample of a number of cars that passed on an actual street) and through formulas we calculated the power consumed by the lampposts in two cases, the first case is when the lights is always on. The second case is when the smart system applied. We also applied fuzzy logic to the system to take the intensity of the ambient light (the sun light) under consideration. The results showed that the proposed smart lighting system is efficient and reliable in saving energy. The energy saved for both (smart and fuzzy) systems was enormous.
Design and implement a self-managed computer network for electronic exams and sharing
Jul 1, 2020Journal Indonesian Journal of Electrical Engineering and Computer Science (IJEECS)
Publisher Institute of Advanced Engineering and Science
DOI http://doi.org/10.11591/ijeecs.v19.i1.pp466-475
Issue 1
Volume 19
To implement electronic exams and material resources sharing (which is usually limited), an operating base is must be available for trading information and correspondence management. In this paper, self-managed computer networks, no server was designed. This network aims to share information and correspondence management between the laboratorys computers. A specific software called Packet tracer used for designing and simulating the network, also for choosing the right medium and install the IP address so that a maximum of data flowing with minimum time and no data loss can be achieved. A class-C IP address was selected since it is the famous class and it doesnt need any special equipment. Also it has a wide range of of computers in case of expanding the network. We didnt use a wireless medium to protect the network from hacking. The network was based on an Ethernet medium and a star-connection between the computers with an average signal distributer. An excellent result was achieved after testing the network with a low error percentage, as shown in the result section of this paper.
Design and Implement A Smart Blind Stick
Aug 1, 2019Journal Journal of Advanced Research in Dynamical and Control Systems
Issue 8
Volume 11
Technologies are growing very fast, which helps people to get a better and easier life. The smart stick is a technique to help sightless people to recognize their way. Sightless People suffer from the lack of ability to do their daily activities, from walking in the street to visiting friends or relative or any daily things. Therefore, the solution for this major problem is proposed by designing a stick that can aid the person to walk safely without having fear of hitting someone on the way or any solid objects. The stick has been designed using Solid Work software. The electric circuit was simulated using Proteus software for designing and simulating electrical circuits. In this paper, we have used three ultrasonic sensors. One sensor has been placed in front of the stick and the other two have been placed on both sides, left and right. To detect the motion from almost every side, it has been used vibrating motor and buzzer alarms to alert the person if some obstacle is detected near him.