Asian Journal of Research in Computer Science
Not a member yet
792 research outputs found
Sort by
MRI-based Brain Tumor Image Classification Using CNN
No full textThough all brain tumors are not cancerous but they caused a critical disease produced by irrepressible and unusual dividing of cells. For the case of Medical diagnostics of many diseases, the health industry needs help, the current development in the arena of deep learning has assisted to detect diseases. In recent years medical image classification has gained remarkable attention. The most well-known neural network model for image classification problems is the Convolutional Neural Network (CNN). CNN is the frequently employed machine-learning algorithm that is used in Visual learning and Image Recognition research. It is considered to derive features adaptively through convolution, activation, pooling, and fully connected layers. In our paper, we present the convolutional neural network method to determine cancerous and non-cancerous brain tumors. We also used Data Augmentation and Image Processing to classify brain (Magnetic Resonance Imaging (MRI). We used two significant steps in our proposed system. First, different image processing techniques are used to preprocess the images and secondly we classify the preprocessed image using CNN. Brain tumor classification is a process of identifying and separating the cancerous and non-cancerous brain tissues and labeling them automatically. We use the famous machine learning algorithms Convolutional Neural Network which is broadly employed for image classifications. This experiment is conducted on a dataset of 2065 images. In our dataset number of training, examples are 1445, the number of validation examples is 310, and the number of testing example is 310. We also used data augmentation to raise the number of the dataset. We achieved a high testing accuracy of 94.39%. The proposed system displayed sufficient accuracy on the dataset and beat many of the noticeable present methods
A Hybridized Feature Extraction Model for Offline Yorùbá Document Recognition
No full textDocument recognition is required to convert handwritten and text documents into digital equivalents, making them more easily accessible and convenient to store. This study combined feature extraction techniques for recognizing Yorùbá documents in an effort to preserve the cultural values and heritages of the Yorùbá people. Ten Yorùbá documents were acquired from Kwara State University’s Library, and ten indigenous literate writers wrote the handwritten version of the documents. These were digitized using HP Scanjet300 and pre-processed. The pre-processed image served as input to the Local Binary Pattern, Speeded-Up-Robust-Features and Histogram of Gradient. The combined extracted feature vectors were input into the Genetic Algorithm. The reduced feature vector was fed into Support Vector Machine. A 10-folds cross-validation was used to train the model: LBP-GA, SURF-GA, HOG-GA, LBP-SURF-GA, HOG-SURF-GA, LBP-HOG-GA and LBP-HOG-SURF-GA. LBP-HOG-SURF-GA for Yorùbá printed text gave 90.0% precision, 90.3% accuracy and 15.5% FPR. LBP-HOG-SURF-GA for Handwritten Yorùbá document showed 80.9% precision, 82.6% accuracy and 20.4% (FPR) LBP-HOG-SURF-GA for CEDAR gave 98.0% precision, 98.4% accuracy and 2.6% FPR. LBP-HOG-SURF-GA for MNIST gave 99% precision, 99.5% accuracy, 99.0% and 1.1% FPR. The results of the hybridized feature extractions (LBP-HOG-SURF) demonstrated that the proposed work improves significantly on the various classification metrics
An Enhanced Data Transmission in Mobile Banking Using LSB-AES Algorithm
No full textThe advent of mobile banking applications has transformed the way customers’ access banking services from brick-and-mortar to remote banking. The ubiquitous nature of this innovation has encouraged its adoption. This is because of improved banking services and accessibility to the services on a 24/7 basis using the internet. However, mobile banking applications are susceptible to numerous security threats and vulnerabilities that adversaries take advantage of to siphon money from bank customers. The aim of this study is to design and evaluate least significant bit and advanced encryption standard cryptography (LSB-AES) hybrid algorithm to protect data on transmission in mobile banking. This study employs data science research design, and performance evaluations demonstrated through simulations using MATLAB. Findings of this study can be applied to banks offering mobile banking across the world. This study utilized six color images from University of Southern California’s Signal and Image Processing Institute (USC-SIPI) dataset which were stored in Tagged Image File Format (TIFF). Contemporary steganographic systems utilize a minimum of 2 to 10 images for test simulations. Visual quality analysis of cover images and stego images was done using the following evaluation metrics: Mean Squared Error (MSE), Peak Signal-to-Noise Ratio (PSNR), and histogram analysis. Security of the proposed algorithm was done using entropy analysis. MSE values of cover and stego images should be closer to zero and indicate that cover image is of good quality for embedding data. On the other hand, PSNR values should be more than 40 decibel (dB) which indicates good imperceptibility. Histogram analysis should demonstrate no visible distortions between the cover and stego image for it to be free from statistical attacks. Entropy analysis should yield values close to 8 for the algorithm to be robust. Results from the proposed LSB-AES hybrid algorithm evaluation metrics reveals that Mean Squared Error (MSE) values ranges from 0.0001297 to 0.0005646 while Peak Signal-to-Noise Ratio (PSNR) values ranges from 80.65 to 87.71 and entropy values ranges from 6.295 to 7.762. Histogram analysis reveals that the cover and stego images are almost similar. These results infer that the proposed algorithm has good quality images with good imperceptibility and that the proposed algorithm is reliable, robust and secure for mobile banking. Entropy and histogram analysis results show that the proposed algorithm is resistant to Man-in-the-Middle attacks. This study recommends that legislation evaluates and amends security of mobile banking policies so that the proposed LSB-AES hybrid algorithm can be adopted as a secure solution for mobile banking
An Enhanced Load Balancing Algorithm for Cloud Enterprise Resource Planning (ERP) Data in a Multi-Cloud Environment
No full textBusinesses and individuals have seen the need to adopt the cloud and mulit-cloud environment for their businesses and storage of data. The load balancing concerns especially in the multi-cloud environment was investigated and a new algorithm proposed. In this research, a proposed new load balancing algorithm is presented and compared with the Round Robin (RR) and Weighted Round Robin (WRR) algorithms. The proposed scheduling algorithm considered several Cloud ERP Data chunks to analyse the data transmission rate or throughput, the transmission delay, data loss and the Cloud ERP Data drop ratio. The proposed algorithm performed better compared to the Round Robin (RR) and Weighted Round Robin (WRR) in a multi cloud environment with data chunks above 150 in terms of throughput. The proposed algorithm again outperformed the RR and WRR with a recorded lower transmission delays and lower data loss
Comparison of Canonical and Generalized Canonical Correlation Analysis Using Some Continuous Distributions
No full textThis study is designed to compare canonical and generalized canonical correlation analysis for two data sets using five continuous distributions namely; Beta, Exponential, Gamma, Weibull and Normal distributions as control. Simulation studies for samples of sizes n = 10, 20, 30, 40, and 50 replicated 10,000 each were analyzed using R-programming language. Relative efficiencies of the two methods (CCA and GCCA) calculated for each of the distributions under consideration showed no significant differences in the two methods
Optimizing Immersion: Analyzing Graphics and Performance Considerations in Unity3D VR Development
No full textAims: Virtual reality (VR) is a new approach that gives users an immersive experience. VR applications frequently experience performance bottlenecks due to the high processing cost of displaying real-time animations multiple times (for both eyes) and the limited resources of wearable devices, performance optimization serves a significant part in VR application development. The goal of this paper is to investigate approaches and techniques available in Unity3D to improve the immersion of VR in virtual reality. This entails studying visuals and performance elements to design more fluid and engaging VR apps.
Study Design: Quantitative analysis.
Methodology: Systematic literature reviews, Consensus on the design space for VR-based education would greatly benefit future advances. To do this, we use a systematic mapping technique to the literature, extracting essential data from papers indexed in four academic online databases.
Results: VR applications include extra aspects like graphics rendering and actual time animation, performance improvements for VR apps might differ significantly from those for standard software. Fully immersive virtual environments vary from standard virtual reality environments in that they can record entire body movements. This skill enables users to engage with other avatars, computer-controlled entities, and artifacts in the virtual world using their complete range of physical activities. VR performance optimization, in essence, relates to the methods, tactics, and procedures used to ensure that VR experiences function as smoothly as feasible.
Conclusion: This optimization might be performed on different kinds of technology, particularly low-end devices, without sacrificing visual fidelity or immersive experience. VR is well-known for. It is critical to optimize visuals as well as performance in Unity3D VR apps to produce immersive and engaging experiences. Balancing great aesthetics with seamless performance necessitates a thorough grasp of Unity\u27s features as well as a dedication to constant optimization
The Rise of Serverless Architectures: Security Challenges and Best Practices
No full textThe field of serverless computing has had significant growth and recognition in the past decade. This emerging area has garnered attention because to its notable impact on cost reduction, latency reduction, scalability improvement, and elimination of server-side management, among other benefits. Nevertheless, there is still a dearth of comprehensive study that would facilitate developers and academics in gaining a more profound comprehension of the importance of serverless computing in many scenarios. Therefore, it is imperative to provide scholarly study data that has been published within this particular field. This study conducted a comprehensive analysis of 275 scholarly articles retrieved from reputable literature sources, with the aim of extracting valuable insights pertaining to serverless computing. Subsequently, the acquired data underwent analysis in order to address many study inquiries pertaining to the contemporary advancements in serverless computing, encompassing its fundamental principles, available platforms, and patterns of utilization, among other relevant aspects. In addition, we analyze the current obstacles confronting serverless computing and explore potential avenues for future research to facilitate its deployment and utilization
ChatGPT: Early Adopters, Teething Issues and the Way Forward
No full textFrom one-way communication to two-way communication, text-based to multimedia, unencrypted to end-to-end encrypted chats, computer, computer-to-human chats have evolved over the years to more sophisticated chatbots, mimicking natural human language based on predefined algorithms. ChatGPT is one of the most recent of such tools and, due to its capabilities has grown tremendously in use. Despite the numerous benefits of ChatGPT, some users have raised numerous concerns about its usage. The purpose of this study was to investigate the underlying determinants contributing to the effective adoption of ChatGPT, elicit from users the challenges to the use of ChatGPT, and make recommendations to developers and users on how to sustain ChatGPT. Questionnaires, face-to-face interviews, and focus group discussions were used to collect data for discussion.
The study revealed very high acceptance rates of the chat tool stemming from its initial wildfire spread of over 1 million users within 5 days of launch to the 85% acceptance or perception among users that ChatGPT has the potential to improve how they perform their own daily professional and personal tasks.
Although 90% of users claimed to find the tool relatively easy to understand, navigate and use they also pointed out areas that need upgrading such as inclusion of multimedia and more accessibility features
Exploratory Search Prompt Generation using n-Degree Connection in Knowledge Graph
No full textSearch engines play a vital role in retrieving in- formation, but users often struggle to express their precise information needs, resulting in less-than-optimal search results. Therefore, enhancing search query refinement is crucial to elevate the accuracy and relevance of search outcomes. One particular challenge that existing search engines face is presenting refined results for queries containing two or more unrelated entities.
This paper presents a novel approach for efficient search prompt generation by leveraging connected nodes and attributes in the knowledge graph. We propose a comprehensive exploration technique that explores the n-degrees connections and their at- tributes to generate all possible imaginative prompts. We realized that n-degrees connection exploration is an expensive task, hence we conducted experiments to determine the effectiveness of 2- degree exploration prompts in covering all the user-asked queries within the provided dataset.
In testing with approximately 2000 queries on a related knowl- edge graph dataset, we found out that our proposed methodol- ogy significantly improved query coverage. At n=1 depth, the coverage increased from 58% without the methodology to 84% with it. At n=2 depth, the coverage rose from 92% without the methodology to nearly 99% with it. Additionally, due to our question caching strategy at n=2, we observed faster response times for all questions
The Impact and Benefits of the E-System for Administration Management in Primary and Secondary Schools for Teachers and Parents
No full textThis research aims to identify the impact of the E-management system on the work of teachers, parents, and students. Enrolling students with all the information and keeping grades, behaviors, activities, and, most importantly, analytical reports should be an easy process for teachers through the E-system. Parents can track their child\u27s progress at the right time while students receive teacher notifications, complete quizzes, access their success, track their shortcomings, and many other benefits.
The paper focuses on the following research objectives: the impact of the Management System on the organization of the teacher\u27s class, research on the effects of the Management System in creating and generating analytical reports for each student, analysis of the impact of the Management System on the communication of the teacher with students as well as with parents, Investigating the effects of the Management System on the teacher\u27s will and researching the impact of the Management System on commitment and work by the teacher