Asian Journal of Research in Computer Science
Not a member yet
792 research outputs found
Sort by
Early Parkinson\u27s Disease Detection Using Machine Learning Approach
No full textParkinson\u27s disease (PD) is a progressive neurodegenerative disorder that affects movement and motor skills. Early diagnosis and treatment of Parkinson\u27s disease are crucial for improving patient outcomes; however, traditional diagnostic methods are time-consuming and subject to observer bias. This study aims to use a machine learning model for the detection of Parkinson\u27s disease. The model will be trained on a public repository dataset of biomedical voice measurements from individuals with and without Parkinson\u27s disease and its performance will be evaluated in terms of accuracy and precision. The results of this study have the potential to revolutionize the diagnosis of Parkinson\u27s disease by providing a fast, non-invasive, and reliable diagnostic tool. The study\u27s results could also have implications for the development of similar diagnostic tools for other neurodegenerative disorders
Cloud BPM Application (Appian) Robotic Process Automation Capabilities
No full textCloud Business Process Management (BPM) platforms, such as Appian, have gained significant traction in recent years due to their ability to streamline and automate business processes. One key feature that has revolutionized process automation is Robotic Process Automation (RPA). RPA enables the creation of software robots that mimic human actions and interact with various applications, systems, and data sources to perform repetitive and rule-based tasks.This abstract focuses on the RPA capabilities within the Appian platform and its impact on business process automation. Appian\u27s RPA capabilities provide organizations with the ability to integrate and orchestrate human-centric and robot-centric tasks within their business processes, resulting in improved efficiency, accuracy, and productivity.By leveraging Appian\u27s RPA capabilities, organizations can automate manual, time-consuming tasks, such as data entry, data validation, and data extraction, by creating software robots that execute these tasks with speed and precision. The platform offers a user-friendly interface for designing, deploying, and managing software robots, enabling both technical and non-technical users to participate in the automation journey. Appian’s RPA capabilities also facilitate seamless integration with existing systems and applications, allowing organizations to leverage their investments in legacy systems while enhancing their automation capabilities. Through APIs and connectors, software robots can interact with databases, web services, enterprise applications, and other sources of data, enabling end-to-end process automation. Additionally, the Appian platform provides comprehensive monitoring, reporting, and analytics capabilities for RPA. Organizations can track and measure the performance of software robots, identify bottlenecks, and gain insights into process efficiency and effectiveness. This data-driven approach allows for continuous improvement and optimization of automated processes. Furthermore, the cloud-based nature of the Appian platform offers scalability, flexibility, and accessibility. Organizations can rapidly scale their RPA deployments based on business needs, leverage cloud resources for increased computing power, and enable access to software robots from any device and location. In conclusion, Appian\u27s RPA capabilities within the Cloud BPM application offer organizations the opportunity to achieve significant advancements in process automation. By leveraging software robots to perform repetitive tasks, organizations can enhance operational efficiency, reduce errors, and allocate resources to more value-added activities. The abstract highlights the key features and benefits of Appian\u27s RPA capabilities, setting the stage for further exploration and understanding of how this technology can revolutionize business process management
Synergy Between AI and Robotics: A Comprehensive Integration
No full textThe emergence of artificial intelligence is mostly linked to software-driven robotic systems, including mobile robots, unmanned aerial aircraft, and, to a growing extent, semi-autonomous automobiles. Nevertheless, the significant disparity between the algorithmic realm and the physical realm hinders current systems from achieving the desired outcome of creating intelligent and user-friendly robots that can effectively engage with and manipulate our human-centric environment. The nascent field of machine intelligence (MI), which combines robotics and artificial intelligence, strives to develop reliable and embodiment-aware artificial intelligence systems. These systems possess self-awareness and an understanding of their environment, enabling them to adapt to the interacting body they are operating. The incorporation of artificial intelligence (AI) and robotics into control, perception, and machine-learning systems is necessary for the realization of fully autonomous intelligent systems in our everyday existence. This review provides an overview of the historical development of machine intelligence, tracing its origins to the twelfth century. It then proceeds to examine the present state of robotics and artificial intelligence (AI), discussing significant systems and contemporary research directions. Additionally, the article outlines the remaining challenges in these fields and speculates on the potential future of human-machine interactions that has yet to be realized
Receiver Operating Characteristic Optimization Based on Convex Hull and Evolutionary Algorithm
No full textClassification is basically a Multi-objective problem. The efficiency of classification vastly depends on performance of a classifier which can be evaluated on the basis of Receiver Operating Characteristic (ROC) graph, Area under Curve (AUC), and selection of different threshold values are generally used as a tool. In machine learning, generally, 2-D classifiers are available that deal with bi-objective problems where overlapping of class may occur i.e. sensitivity and specificity may overlap. Recently, multi-class classification in which classes are mutually exclusive is in research trends along with the evolutionary algorithm. The application of ROC graph is extended to evaluate multi-dimensional classification as it is cost sensitive learning. The goal of this paper is to gather recent achievements in the field at one place to analyze the classifier performance for multi-dimensional classification problems using convex hull and evolutionary algorithms. In this paper, we tried to cover all the existing recent advance techniques in maximization of ROC and proposed a convex hull and evolutionary algorithm based new model for ROC maximization.  
Auto Encoder Fixed-Target Training Features Extraction Approach for Binary Classification Problems
No full textThe main issues with machine learning-based feature extraction techniques are the requirement of extensive domain-level knowledge, experience, and the need to be supported by large amounts of data that are sometimes not available. Moreover, it is often difficult to apply domain-level knowledge to extract the necessary features for building a machine-learning classifier. Therefore, it is significantly important to find and develop feature extraction techniques that depend mainly on the training data and don’t require or depend on domain-level knowledge and experience. To address these issues for binary classification problems, a novel feature extraction approach, AE-FT(Fixed Target) for extracting common features using a Deep Belief Network (DBN)-based Autoencoder (AE) is proposed in this paper. In this approach, common features are extracted by a DBN trained on a dataset sample’s binary using the Fixed Target training approach.
The proposed common features extraction approach is tested and evaluated on two different data sets. For each dataset, the extracted features are used to train seven of the common machine learning binary classification algorithms and compared their performances. Moreover, the number of extracted features is very small compared to other existing feature extraction methods. Therefore, the proposed common features extraction method improves the performance of the binary classification algorithms by reducing the number of features reducing laborious processes, and increasing the recognition accuracy effectively.
The results show that the proposed common features extraction approach, without any domain-level knowledge or human expertise, provides a very good performance compared to other feature extraction techniques
The Development and Deployment of an Online Exam System: A Web Application
No full textThe rapid advancements in computer technology and the internet’s acceptance in every aspect of our lives, particularly in recent years, have made students and instructors vital in the teaching and learning sector. Web-based studies have also brought about advances in the education area, and numerous applications have become widespread in this field. In this paper, we suggested an online test multiple-choice question assessment system for students called the Online Exam System (OES). This system may be used by any university, college, or institution that has a computerized education system. The OES can be used by teachers to administer quizzes. The system will calculate the participant’s performance based on his response, and the following question will be created based on the participant’s performance. After the examination, the system will display the results and offer feedback based on the participant’s request. Administrative control over the entire system is available. A teacher has authority over the question bank and is responsible for creating test schedules. Therefore, the project will be very helpful for the beginner and mid-level programming learners. And also, will give a proper guideline to the students who are willing to learn programming and introduce the users with competitive programming and problem-solving skills
The Impact of Learning Management System on Student Academic Performance of Computer Science Department of Federal Polytechnic Kaura Namoda, Zamfara State, Nigeria
No full textAims: This study is aimed at exploring the impact of a Learning Management System (LMS) on the academic performance of the Computer Science Department, Federal Polytechnic Kaura Namoda, Zamfara State, Nigeria.
Study Design: We adopted a hybrid of reliability test study and questionnaire.
Place and Duration of Study: Higher National Diploma (HND) in Computer Science Department, Federal Polytechnic Kaura Namoda, Zamfara State, Nigeria for 1st semester of 2021/2022 academic session.
Methodology: The study was conducted with 60 students, which is made up of 34 males and 36 females. This respondent was given a test at the commencement of the course and was also tested after using the LMS for learning, their pretest and posttest scores were recorded. Furthermore, out of the 60 questionnaires that were served to the participants and 53 numbers were retrieved, 88.30% of the response. Out of the remaining 7 respondents, 4 were not retrieved and 3 were not correctly filled. The questionnaire has questions to which participants are expected to respond on a measure of five-point Likert scale. The Likert scale was assigned numerical values of 1,2,3,4, and 5 respectively from the negative judgment to the positive opinion.
Results: From our reliability test on the data, it showed that the students had an average gain score of 12.5. This signifies that, on average, the participants who used the LMS platform had 12.5 points higher during the posttest than they had during the pretest. This reliability test score for the student measures their consistency on both tests. A score of 1 was used to indicate reliability while 0 is not reliable. From our data, it showed test score reliability for students to be above 0.95, indicating that student scores were consistent. Meanwhile, on the respondent\u27s thoughts and feelings about the LMS, With LMS, the student can understand clearly the important learning goal with 4.06, with LMS students who say they are aware of important course during for completion of the course 3.81, 3.77 can easily be guided on the course title while using LMS and 3.92 can easily get abreast of how the task is to be performed to enable the student to learn. The LMS\u27s impact on students\u27 social interaction, showed that the new system has assisted student-teacher relationships and has made them communicate and ask questions freely from people in their area of study. It also showed that students\u27 confidence and interaction with their peers have grown and their online discussion horizon has improved their social interaction.
Conclusion: The LMS has improved the academic performance of students of the Computer Science Department, Federal Polytechnic Kaura Namoda, Zamfara State, Nigeria. Their scores were higher compared to what they used to be and meanwhile, the duration of the study also significantly improved. They are now aware of the reason for collaborative and academic study
Novel Impact Model for Mobile Learning Adoption in Higher Education
No full textIn successful mobile learning (ML) integration, factors associated with live-ware, software, and infrastructure are important. The main objective of this study is to investigate and model the influencing factors for learners and teachers at once to adopt ML in higher education. The proposed model consists of five impact factors: teacher, learner, mobile devices, ML tools, ML contents, communication technologies, and higher education institutes. Then the proposed model was implemented using a modified Moodle mobile application and evaluated using 60 teachers and 60 learners attached to the University of Kelaniya, Sri Lanka in 2021. According to the experimental research design approach, the proposed impact model was assessed as pre-test and post-test surveys using seven questionnaires. According to the Pearson correlation coefficient test, the most significant factor for learners and teachers to adopt ML is the mobile device. Learning content and communication technology were elected as the second most significant adoption factors for teachers and learners consecutively. However, higher correlation values were obtained for all factors denoting that they are greatly influenced the participant to adopt ML. The significant influencing factor of each impact factor was also investigated. In conclusion, it was recommended that featured smart devices, quality learning content, user-satisfied communication technology, academic enriched ML tools and higher education institutes with sound educational facilities are crucial for the university community to adopt ML in higher education. These findings help design academic community acceptable ML environments for higher education context
Cyber Kill Chain Analysis Using Artificial Intelligence
No full textArtificial Intelligence (AI) tools are promising multifaceted techniques for addressing the most mundane tasks for greater efficiency and high productivity. Cyber security space is one of the areas that AI is promising to revolutionize. This study will develop a conceptual and theoretical framework to support a research design that can simulate research in understanding how AI can be applied to Cyber Kill Chain phases. This study has reviewed 21 journal and conference articles mostly from IEEE Xplore database. An overview of the application of artificial intelligence (AI) in cybersecurity, particularly within the framework of the Cyber Kill Chain was provided in this study. It also emphasizes the limitations of traditional security approaches and the necessity for innovative and intelligent defense methodologies. The results of reviewing the relevant literatures discovered that the key components of cybersecurity, includes identity, asset management, automated configuration management, security control validation, governance, risk assessment, and vulnerability identification. A theoretical framework was developed which introduces the Cyber Kill Chain model with a Unified Kill Chain model to address its shortcomings. Application of AI in cybersecurity offers an optimistic solutions to address the evolving threat landscape. AI techniques, such as machine learning, anomaly detection, and behavioural analysis, have shown great potential in enhancing various aspects of cybersecurity. However, challenges related to data quality, adversarial attacks, and privacy concerns need to be addressed for successful implementation. Further research and development are crucial to fully harness the power of AI in cybersecurity and stay ahead of evolving cyber threats
Deepfakes in Cyber Warfare: Threats, Detection, Techniques and Countermeasures
No full textTechnology known as deepfake (DT) has reached an entirely new level of complexity. Cybercriminals now have the ability to modify sounds, images, and videos in order to mislead individuals and businesses and spread false information. This constitutes a rising threat to international organizations as well as individuals, and it is imperative that something be done about it. This article presents an overview of deepfakes, discussing their usefulness to society as well as the operation of DT. This article focuses on the dangers that can be posed by deep fakes to the economic, political, and legal institutions of countries all over the world. In addition to this, the study will investigate various solutions to the problem of deepfakes, and it will finish by discussing potential directions for further research