Recent Submissions

  • Implementing enhanced web-based application security using shadow daemon WAF

    balfagih, zain; Abusulaiman, Rana; Shata, Ghalia; Alkhouli, Deema; Computer Science
    The importance of companies taking proactive steps to defend their web applications is growing as the number of web-based threats rises. Utilizing web application firewalls (WAFs) is one such measure. In this study, we investigate the possible advantages of integrating Shadow Daemon into the design of regional start-up businesses to enhance their cybersecurity defence. We start by giving a general review of Shadow Daemon and its features, emphasizing its capacity to recognize and stop typical web application threats like SQL injection, XSS, and LFI. The significance of web application security and the possible repercussions of a successful attack are then covered. Then, we look at the current state of web application security in local start-up businesses, including frequent security flaws and the difficulties they encounter in putting in place workable security measures. Following that, describing the crucial processes required in the implementation process, we suggest a framework for incorporating Shadow Daemon into the architectural design of regional start-up businesses. We also talk about how employing Shadow Daemon might enhance security, lower the chance of data breaches, and boost customer confidence. Finally, we emphasize the value of proactive web application security and the part Shadow Daemon may play in assisting regional start-up businesses to defend their web apps against attacks as we draw to a close. We suggest implementing Shadow Daemon can improve cybersecurity defences and reduce potential security risks for nearby start-up businesses in an efficient and cost-efficient manner.
  • Early Schizophrenia Detection Using Artificial Intelligence

    Mian Qaisar, Saeed; Bukhari, Syeda Maha; Milyani, Danah; Ali, Fatima; Computer Science
    Following years of research, the processes driving schizophrenia (SZ) genesis, recur rence, symptomatically, and therapy remain a mystery. One of the explanations for this condition could be a lack of proper analytic methods to cope with the variety and complexity of SZ. Deep learning algorithms’ (a branch of artificial intelligence (AI) inspired by the nervous system) extraordinary precision in classification and prediction tasks has changed a wide range of scientific domains and is fast penetrating SZ research. Deep learning has the potential to assist doctors in the prediction, dia gnosis, and treatment of SZ. A thorough literature review was conducted to examine existing research on the use of deep learning in the study of psychosis for diagnosis, as well as the use of electroencephalographic data and signal classification. In this study, we propose the suggested methodology; acquisition, pre-processing of the data; extracting the features; dimension reduction; artificial intelligence classifiers. The outcomes are expected to be positive or negative, and different evaluation metrics will be applied. Results from the study are expected to show that the proposed AI classifiers were able to achieve high accuracy in detecting schizophrenia. The findings of this research have important implications for improving early detection and treatment outcomes for individuals with schizophrenia. Future research should focus on further improving the performance of the proposed AI classifiers and exploring their potential for real-world application.
  • Attendance QR Code System for Organizations

    khan, Sohail; Zarra, Leen; Bakashwean, Raneem; Computer Science
    Nowadays, cellphones are a big part of our daily lives thanks to technology. Today's cellphones can swiftly and simply resolve the majority of issues. Everybody's life has been simpler and easier because of the many social apps, business apps, problem-solving apps, educational apps, and marketing apps, among others. The study intended a solution to deal with a difficulty with recording attendance after the technology. The suggested system is an application that uses QR codes to track employee attendance. Employees must scan their unique QR code to verify their attendance; they are not permitted to share or screenshot this code. The system's identity verification process is covered in the paper in order to prevent bogus registrations. The system handles the monitoring and assessment of every employee's attendance. Each person will receive a QR code in their application to scan when they enter the organization. This is a first effort to assess the system's suitability for application in various settings, including offices, classrooms, and laboratories. Based on the findings and data collecting, it can be concluded that the testbed development for the creation of a smart security attendance system using a QR code has been successful.
  • Advanced Abnormal Logins Detection System using ML Algorithms

    khan, Sohail; Alshamrani, Adel; Alowlaqi, Habeba; Almufadda, Flowra; Computer Science
    In recent years, technology and computers have evolved rapidly, allowing tasks to be performed with minimal complexity and effort, but it has also led to an increase in cyber-attacks, such as data breaches and identity theft. This highlights the necessity of effective methods for detecting abnormal logins and protecting sensitive information. The objective of this study is to develop an advanced abnormal logins detection system using machine learning algorithms. To address the limitations of existing approaches, the proposed algorithm incorporates additional contextual information and adopts a hybrid approach that combines machine learning and network analysis. The CERT r4.2 dataset is utilized for evaluation purposes. The performance of the developed algorithm is assessed through experiments, showcasing a high accuracy rate and effective detection of various insider threat scenarios. Comparative analysis demonstrates its superiority over existing approaches, underscoring its potential in enhancing security measures in web applications. Furthermore, this project’s findings highlight its significance in safeguarding sensitive data, mitigating identity theft risks, and promoting resilient infrastructure. The project aligns with the United Nations Sustainable Development Goals, specifically SDG 9 (Industry, Innovation, and Infrastructure) and SDG 11 (Sustainable Cities and Communities), as well as the Saudi Vision 2030. In conclusion, the advanced abnormal logins detection system showcases the potential of machine learning algorithms and hybrid methodologies in addressing cybersecurity challenges. The project’s implications extend to enhancing data protection, mitigating identity theft risks, and contributing to the sustainable development goals outlined by international and national initiatives.
  • Using Machine Learning and Sentiment Analysis Methods to Evaluate Financial Apps Based on User Reviews

    Nouman, Mohammad; Alsalem, Fatmah; Nassir, Jana; Bawazir, Joud; Computer Science
    Sentiment analysis is critical for comprehending people’s opinions and attitudes regarding fi- nancial applications. The study offered a thorough approach to sentiment analysis utilizing machine learning and deep learning models in this research. Data pretreatment procedures, model imple- mentation, and evaluation are all part of the process. Using the Keras and TensorFlow frameworks, we constructed numerous machine learning and deep learning models such as Naive Bayes, SVM, Decision Tree, BERT, and RNN. The accuracy and F1 score measures were used to evaluate the performance of these models. A thematic analysis was also performed to uncover common themes and subjects in financial application reviews. The findings show that sentiment analysis is useful in analyzing user sentiments and providing insights for improving user experience
  • Prediction of Students’ Academic Performance using Machine Learning

    balfagih, zain; Alkaf, Dhekra; Bajammal, Faigah; Asrar, Manal; Computer Science
    Education is an important factor of a civilization and has been of importance throughout history. The right education results in a fruitful future of change, development, and globalization for coming generations. Thus, it is of great importance to constantly re-evaluate the techniques used in modern education systems with the aim of continually improving the provision and quality of education. In a time where technology thrives to support human development in numerous ways, machine learning can be a great asset to the education system. Through the application of machine learning on educational data through prediction, universities and educational institutions will be able to improve teaching and learning outcomes, as well as provide the right support for the different types of students at the institution. The main objective of this project is to provide an accurate machine learning model that can predict students’ performance early on to support them on their journeys. This project will be fulfilled by continuous research, working closely with large numbers of data, and experimenting with different algorithms to reach the best results possible. In this paper Random Forest resulted in being the best performing algorithm for the intended model.
  • Iot Data Transfer System Using Lorawan And Cloud Computing

    Rajput, Adil; Brahimi, Tayeb; Laila, Bakhashab; Maryam, Alnahdi; Raghad, Zarei; Computer Science
    With the rapid growth of technology, the world is in hunger for more data, connection, intelligence, and speed. The solution of LoRaWAN sensors promotes the infrastructure of the Internet of Things (IoT). To emphasize, LoRaWAN is a “point-to-multipoint networking protocol” requiring low-power to connect end-devices and manage communication through network gates. This study aims to assess and build a fulfilled system using LoRaWAN sensors to improve the wireless network experience for users in smart cities, specifically tackling agriculture as a testing medium for evaluating data accuracy and more effective transmission of data packets. The method used in this research is based on applying an experiment of a LoRaWAN model using nodemcu (esp32) applied in smart agriculture. Results show that LoRaWAN is a promising technology that can be implemented in several fields and effectiveness over current technologies like Wi-Fi and Zigbee. The study presents a prototype that addresses both strengths and limitations and issues with a plan for multiple industries to immediately put the technology in action in helping existing areas in utilizing the technology. While using cloud computing to transfer data and have accessible, secure and up to date data. Cloud computing is the sharing of data, resources, and software as the Internet serves as an unseen thread that links everything.