International Journal of Advances in Scientific Research and Engineering (IJASRE), ISSN:2454-8006, DOI: 10.31695/IJASRE
Not a member yet
1411 research outputs found
Sort by
Design and Implementation of AI-Driven Personalized Learning Tools for Tanzanian Secondary Schools
Full text linkThis study investigates the effectiveness of AI-powered personalised learning tools in Tanzanian secondary schools. The research explores the potential of these tools to address the unique challenges these schools face, including large class sizes, limited resources, and significant language diversity. Through a comparative analysis of various AI tools, the study examines their adaptability to Tanzania's educational context, considering language diversity, cultural relevance, and infrastructure constraints. The research employs qualitative design, incorporating comparative case study elements to evaluate the functionalities and adaptability of selected AI-driven personalised learning tools. Data collection involves a systematic review of available tools, semi-structured interviews with educators and AI experts, and a survey to gather information specific to the Tanzanian educational context. Key findings indicate that AI-powered personalised learning tools offer significant potential for enhancing education in Tanzanian secondary schools. These tools can adapt to individual student needs, providing personalized learning experiences that traditional methods cannot achieve. However, the study also identifies challenges, including limited language support, the need for culturally relevant content, and infrastructural constraints. Addressing these challenges is crucial for maximizing the effectiveness of these tools in the Tanzanian context. The study concludes that AI tools can significantly contribute to personalized learning in Tanzania, but their successful implementation requires careful consideration of local needs and challenges. Recommendations include prioritizing AI tools with high adaptability, robust multilingual support, and mobile-first designs to cater to Tanzania's diverse linguistic landscape and technological infrastructure. Future research should focus on empirical testing within Tanzanian classrooms and refining AI tools to align better with local educational needs
Implementation of the Islamic Boarding School-Based Driving School Policy
Full text linkThis study aims to describe and analyze the implementation of the pesantren-based School Mover program and the supporting and inhibiting factors based on Kepmendikbudristek Number 162/M/2021. The research was conducted at SMA Darut Taqwa Purwosari, Pasuruan Regency. The data analysis technique in this study used descriptive qualitative analysis. The research results on implementing the Mover School Program at Darut Taqwa High School were influenced by a strong pesantren environment, with Islamic values as the primary foundation. Support from school leaders, teachers, and the community encourages positive changes in learning, although the prohibition of using electronic devices is a significant obstacle. The program prioritizes student character development by integrating the Merdeka's and the Pesantren curriculum, shaping students into Pancasila learners with cognitive, social, and spiritual abilities. Technical challenges, such as limited technology facilities and budget changes, were overcome through policy evaluation and social approaches. Strong parental cultural support and teacher commitment are key to the success of this program despite the gap in student adaptation. This study emphasizes the importance of intensive teacher training, synergy between schools and related parties, contextual learning innovation, and continuous evaluation to improve the effectiveness of the Mobilizing School Program
Design and Development of a Climate IQ Smart Solution Powered by Data Insights
Full text linkIn response to the growing challenges posed by climate change, there is an urgent need for innovative solutions that leverage data insights to enhance climate resilience and sustainability. This paper presents the design and development of Climate IQ Smart Solutions, a comprehensive system that harnesses advanced data analytics, machine learning, and Internet of Things (IoT) technologies to provide actionable insights for various stakeholders, including farmers, city planners, and environmental researchers.
Climate IQ Smart Solutions aims to empower users by delivering precise, real-time information and predictive analytics to optimize resource management, mitigate adverse climate impacts, and promote sustainable practices. The system integrates diverse data sources, including IoT sensors, satellite imagery, and historical climate data, into a unified platform. This data is processed using state-of-the-art cloud infrastructure and advanced machine-learning algorithms to generate valuable insights.
The architecture of ClimateIQ Smart Solutions comprises several core components: data acquisition, data storage, data processing, and user interface. Data acquisition involves collecting real-time data from a network of IoT sensors and external data sources. This data is then stored in a scalable cloud-based storage system, ensuring efficient handling of large volumes of information. Advanced data processing techniques, including machine learning and predictive analytics, are employed to analyze the data and extract meaningful patterns and trends. Finally, an intuitive user interface presents these insights in a user-friendly manner, allowing stakeholders to make informed decisions.
Key features of Climate IQ Smart Solutions include real-time monitoring of environmental conditions, predictive analytics for forecasting climate-related events, and recommendations for optimizing resource usage. For instance, farmers can use the system to monitor soil moisture levels and receive irrigation recommendations, while city planners can leverage predictive models to prepare for extreme weather events
Subject Review: Cyberbullying and Detection Methods
Full text linkCyberbullying is a prevalent issue on social media, causing significant mental and social harm to victims. This review highlights cutting-edge cyberbullying detection technologies, concentrating on machine learning (ML), deep learning (DL), and natural language processing (NLP). The learners are categorized into supervised, unsupervised, and hybrid models, highlighting their pros and cons. This article covers common research datasets, including social media comments, and addresses difficulties including data imbalance, linguistic diversity, and context interpretation. Future efforts include developing context-sensitive models, improving on-the-fly detection, and addressing ethical concerns in automated system deployment
Subject Review: Comparison of Data Encryption Algorithms
Full text linkData encryption protects data by converting it into an unreadable format, ensuring confidentiality, integrity, and security during transmission and storage. There are several encryption algorithms, each with its own distinct characteristics, strengths, and weaknesses. This study compares the most widely used encryption algorithms: AES, DES, RSA, and ECC, analyzing their strengths, weaknesses, and ideal use cases
Enhancement the Attendance System in Educational Institutions Based on Image Processing and Facial Recognition
Full text linkCurrently, accounting for student attendance is one of the key components for raising the calibre of specialized training. It is possible to automate this procedure. The article suggests using facial recognition technology, which enables you to identify multiple people at once without having to make direct contact with them or utilize pricey equipment, to create the attendance system in educational institutions. Based on the article's analysis of contemporary facial recognition techniques, this solution makes use of convolutional neural networks Retina Face and ResNet. Our attendance system's architecture is enhanced by image pre-processing techniques that, when needed, apply algorithms to the image to even out colours, sharpen edge, boost brightness, and reduce noise. These techniques are based on our suggested methodology, which is based on the BREN measure. The computer experiment results are shown, demonstrating the higher efficiency of the suggested approach in comparison to its equivalents
Metric-driven Computational Models for multi -focus Image Fusion
Full text linkA significant and popular technique for merging the focused portions of several source images into a single, completely focused image is multifocal image fusion. Accurately identifying the focus areas is currently crucial to solving the multifocal image fusion problem, particularly when the source pictures produced by cameras exhibit anisotropic blur. Both focused and unfocused portions are present in the photos that the camera captures because of the focal lengths of the optical lenses. Combining the focused portions of several photos in the same scene using multifocal image fusion technology is a popular way to get fully focused images. Because of its rich features and worldwide clarity, the fully focused merged image is more suited for computer processing and visual perception. This research is focused on the technique of merging multi-focal images using the iterative scheme, where each level of the input images is represented by a distinct iterative scheme from the second input image. If (each point of the input image is determined and combined with the second input image, then the pixel values are determined to obtain a new merged image by combining the two images to obtain a merged image with greater clarity and details
Sentiment Analysis and Topic Modelling of Tourist Reviews on Heritage Destinations of Lawang Sewu
Full text linkUnstructured online reviews for heritage destinations like Lawang Sewu offer a rich source of data for an in depth analysis of visitor perceptions. A combined approach of supervised sentiment classification and topic modeling was applied to a bilingual review corpus to identify the specific drivers of satisfaction and dissatisfaction. The analysis revealed a clear dichotomy in visitor feedback. The overall sentiment is overwhelmingly positive. This positivity is driven by the site's core heritage value, its historical architecture, and a family friendly atmosphere. In contrast, negative sentiment is highly concentrated and not directed at the core attraction, but rather at specific operational frictions. These frictions are primarily related to ticketing, pricing, and the perceived quality of guided tour services. This integrated approach provides management with actionable, data driven insights, enabling them to protect celebrated strengths while systematically addressing logistical weaknesses. This work demonstrates how unstructured public feedback can be transformed into a strategic tool for targeted service improvement at heritage destinations
Secure Image Steganography Using AES-CBC Encryption and Dynamic Key Derivation from Image Features
Full text linkThis research presents a technique for concealing messages within images by combining AES encryption in Cipher Block Chaining (CBC) mode with Least Significant Bit (LSB) embedding. A key feature of this method is the dynamic generation of the encryption key based on intrinsic statistical properties of the image—specifically, the standard deviation of the RGB channels— This eliminates the necessity for external key exchange, making the encryption process inherently dependent on the image itself. The secret message is first encrypted using AES-CBC, and the resulting ciphertext is embedded into the image using LSB manipulation. The proposed approach was evaluated using standard image quality metrics. Experimental results reveal excellent preservation of visual quality, with a Mean Squared Error (MSE) of 0.000091, Mean Absolute Error (MAE) of 0.000091, Peak Signal-to-Noise Ratio (PSNR) of 88.53 dB, Structural Similarity Index Measure (SSIM) of 1.000000, and a coefficient of determination (R²) of 1.000000. A total of 336 pixel-level changes were detected, all strictly confined to the LSBs, resulting in a 100% LSB-only modification ratio. These outcomes confirm that the proposed method ensures both strong security and the imperceptibility of the hidden data
Analytical study of Numerical Approximation for Ordinary Differential Equation
Full text linkThis research primarily focuses on methods for solving Initial Value Problems (IVPs) for Ordinary Differential Equations (ODEs), with a particular emphasis on three key methods: the Euler method, the Fourth-Order Runge-Kutta method (RK4), and the Fifth-Order Runge-Kutta method (RK5). The latter two methods are highlighted as highly effective and practical for solving such problems. To ensure accuracy, the numerical solutions obtained using these methods were compared with exact solutions. The results showed that the numerical solutions agreed well with the exact solutions. Additionally, numerical comparisons between the Euler method and the Runge-Kutta methods were demonstrated using MATLAB. Finally, the errors of the proposed methods were studied and calculated for various examples