]]> The methods and procedures involve designing a web platform using Java and React as main technologies, implementation of a login and registration process with hashing passwords and JWT token authentication. Featuring the categorization of locations, tours, and car rentals, along with an interactive map of different attractions across Sarajevo. The platform enables regular users to browse, book, and save services for later, while leaving room for providers (admins) to create and offer their personalized utilities. The system ensures usability across modular design and a responsive user interface.
Results show a useful platform that simplifies tourism and navigation in Sarajevo. Using interactive elements like maps and categorized utilities, the application increases user engagement. Enables providers to manage offerings quickly and attract more consumers to their businesses.
]]>
In the increasingly competitive higher education landscape, digital marketing plays a crucial role in attracting prospective students. However, institutions often face major challenges in tracking cross-channel performance and accurately attributing student applications to specific campaigns. Marketing platforms like Meta Ads and Google Ads often exaggerate campaign effectiveness by counting impressions, clicks, and superficial conversions that may not result in actual applications or enrollments.
This senior design project addresses that issue by developing a Business Intelligence Platform for PPC and Organic Marketing, aimed at providing real-time, actionable insights for the International Burch University (IBU) Marketing Team. The system enables end-to-end campaign tracking by integrating digital ad performance data with actual application data, using a unique medium URL parameter embedded in campaign links and captured during the application process to bridge external and internal data sources.
The platform uses Windsor.ai connectors to fetch ad data into Google Sheets, where a Google Apps Script automates data cleaning and merging. This data is combined with student application records from a MySQL database and visualized in real-time through Looker Studio dashboards.
Results demonstrate the system’s ability to provide clear, actionable insights, enabling the marketing team to compare channels, calculate cost per application, and evaluate the full applicant funnel across departments and academic cycles. In doing so, the platform empowers the team to optimize campaigns based on verified outcomes.
The project highlights the importance of reliable tracking infrastructure, disciplined use of URL parameters, and automation in marketing analytics. It represents a scalable, low-cost solution for improving data-driven decision-making in higher education marketing.
]]>
The system is a full-stack application, leveraging technologies such as React.js for the client, Express.js on the server and MySQL as the relational database. It includes role-based authentication using Google’s OAuth2 strategy, form submissions and status tracking.

The resulting system allows students to apply for internships and fill out daily reports. It further allows company representatives to create listings and choose interns, approve and reject daily reports and fill out final reports. The system further allows internship managers to enter final grades and the administrator to modify company records and choose the current internship manager.
]]> The research utilizes the publicly available Sleep Health and Lifestyle Dataset, comprising 400 records with 13 features, including age, gender, BMI category, sleep duration, stress level, blood pressure, and physical activity level. Five supervised learning algorithms were developed and evaluated: Logistic Regression, Random Forest, Support Vector Machine (SVM), XGBoost, and an Artificial Neural Network (ANN). The models were trained to classify individuals into one of three sleep health categories: No Disorder, Insomnia, or Sleep Apnea.
A comprehensive preprocessing pipeline was implemented, involving data cleaning, feature scaling, one-hot encoding, and SMOTE-based class balancing. Model development followed a nested 5-fold cross-validation strategy, with hyperparameter optimization conducted using GridSearchCV. Performance was evaluated using standard classification metrics: accuracy, macro-averaged precision, recall, F1-score, and ROC-AUC.

Results showed that XGBoost and ANN achieved the highest performance, with almost all scores exceeding 0.9, indicating strong predictive accuracy and generalization across validation folds. Feature importance analysis revealed that sleep duration, blood pressure, and BMI category were the most influential predictors. Visualization tools—including confusion matrices, radar charts, and feature importance plots—were used to enhance model interpretability and diagnostic transparency.
Despite the promising results, limitations exist. The relatively small dataset (n = 400) and the absence of critical variables such as sleep stage architecture, oxygen saturation, and environmental or comorbidity data constrain generalizability and clinical applicability. Future research should focus on incorporating larger, more diverse datasets and integrating longitudinal or real-time data from wearable devices to improve predictive robustness.
In conclusion, this study demonstrates the feasibility and effectiveness of machine learning algorithms in classifying sleep disorders using non-invasive inputs. The findings support the development of scalable, AI-driven diagnostic tools that can enhance sleep disorder screening in both clinical and consumer health settings, contributing to the advancement of telemedicine, digital health innovation, and personalized preventive care.
]]> The "Travel Souvenir" application was built for the Android platform using a modern, modular architecture designed for stability and scalability. The methodology involved integrating a secure cloud backend for user management, real-time data synchronization, and media storage. The app utilizes the device's built-in location services for automatic city categorization, an on-device machine learning model for real-time landmark recognition, and external data APIs to add descriptive context to each souvenir.
The result is a fully functional application where user photos are automatically organized into location-based albums. The app successfully syncs data across devices and includes features such as a public feed for shared albums, personal note-taking, and AI-powered landmark identification. The project concludes that by integrating modern mobile and cloud technologies, it is possible to create an engaging and automated solution to the challenge of preserving digital travel memories.
]]> ]]> The project focuses on four core components: a mobile application, a backend system, Point of Sale (POS) integration, and a charger tablet application. The mobile application enables EV users to locate charging stations, initiate and monitor charging sessions, and process payments with ease. For operators, the backend system ensures seamless management of real-time charging station statuses, secure payment processing, and detailed transaction logging. POS integration expands payment flexibility by supporting on-site transactions via embedded tablet applications and POS devices. Additionally, the tablet application, installed on chargers, offers an intuitive interface for users to initiate charging sessions and track charging progress.
The design and development process prioritizes scalability, user experience, and security. The mobile and tablet applications are built with intuitive user interfaces to cater to a diverse audience. The backend employs robust frameworks and real-time data handling to ensure reliability and scalability, while the integration of POS devices leverages secure payment protocols.
Upon completion, this platform is expected to deliver a seamless and flexible user experience for EV owners, streamline operations for administrators, and improve the overall adoption of EV charging infrastructure. This work represents a significant step forward in creating sustainable and user-centered solutions for the growing EV ecosystem.
]]> The system is built using modern web technologies, with a front-end developed in React and styled with Bootstrap for responsiveness and user-friendly interaction. Authentication is managed through Google Sign-In, restricted to university-issued student emails (ending in @stu.ibu.edu.ba) to ensure authorized access. The backend employs Node.js and Sequelize ORM for handling database operations, with parking data being managed dynamically t o reflect real-time changes in availability.
Functionality includes viewing parking layouts, selecting available slots, and booking or canceling reservations. A calendar and/or map interface provides a visual and interactive overview of parking slot statuses. The system distinguishes between student and professor parking zones to prevent cross-access and ensure fair usage.
]]> ]]>
A three-layer architecture is used: Data Access Layer (DAL), Business Logic Layer (BLL), and API Layer. The DAL stores and manages the data in a database. The BLL contains the main logic for how the system works. The API Layer allows users and other systems to interact with the application through web requests. The user interface is implemented with modern web technology, React, which makes the system easy to use and accessible from any device. C# and ASP.NET Core are used for development.

The results show that the system can make waste management more efficient. It becomes easier to assign tasks, monitor progress, and generate reports. In conclusion, this project can help cities or companies improve their waste management process and reduce problems related to waste.
]]> ]]> The application addresses the problem of limited access to personalized, diverse, and engaging story content, especially for children from various cultural and linguistic backgrounds. By integrating advanced AI services, the platform generates unique stories based on user-provided inputs such as title, age group, genre, and illustration style. Using technologies like Google AI, Hugging Face’s text generation models, and custom text-to-speech tools, the system delivers cohesive narratives paired with AI-generated visuals and narration. The backend is developed using Node.js and Express.js, while the frontend is built with React.js, offering a responsive and user-friendly interface. Data is securely stored and managed using a Drizzle ORM with a PostgreSQL database.
Results show that the system can produce high-quality storybooks with consistent plots, age-appropriate language, and stylized imagery, enhancing the reading experience. The platform also includes features like story exploration, user account management, coin-based generation limits, and payment integration for purchasing additional credits.
In conclusion, the AI Storybook Generator showcases the potential of merging generative AI with interactive design to promote literacy and creativity in young users. Future work may involve mobile app development, multilingual support, story sharing features, and integration of educational objectives into story structure. The application serves as a scalable, customizable tool for families, educators, and storytellers worldwide.
]]>