AUTOML - A Desktop Application for Algorithm Selection and Optimization

Automated Machine Learning (AutoML) has emerged as a practical approach to lowering the entry barrier for applying machine learning techniques by automating repetitive and technically complex steps such as data preprocessing, model selection, and evaluation. This paper presents AutoMLApp v1, a desktop-based automated machine learning application developed as a student-led applied research project. The system is implemented in Python, with a graphical user interface (GUI) built using PyQt5 and a machine learning backend based on the Scikit-learn library. The current version focuses on supervised classification tasks, providing automated dataset handling, model training across multiple algorithms, performance comparison using standard evaluation metrics, and learning curve visualization for training behavior analysis. The system architecture is designed with extensibility in mind, allowing future integration of regression tasks and hyperparameter optimization. Experimental results demonstrate that AutoMLApp v1 can effectively identify suitable classification models for user-provided datasets, making it a useful educational and prototyping tool. This work emphasizes practical system design, applied experimentation, and learning-oriented contributions rather than claiming state-of-the-art performance.