Use of machine learning models during the development of iOS applications

This master's thesis focuses on the use of data mining methods and machine learning during the development of mobile applications for the iOS platform. The main goal of this thesis is the creation, integration, and evaluation of various machine learning models, which were developed using different tools, into an iOS application. The theoretical part of the thesis includes a literature review focused on the use of machine learning in mobile applications, the definition of machine learning, and the specifics of its use on devices with limited computational capabilities, such as mobile phones. The practical part describes the process of developing models intended for deployment in mobile applications using the tools Create ML, BigML, and TensorFlow. Three datasets with different types of data were chosen for training these models. One dataset contains image data, and the other two contain tabular data that differ in the type of included data. One dataset contains purely numerical data, while the other includes a combination of numerical and categorical data. This is followed by a description of how the mobile application was created and which techniques were used to integrate the models. The final part includes an evaluation of the methods for integrating machine learning models into a mobile application for the iOS platform and discusses the limitations of each approach. The source code of the application and the code used for data preprocessing and model training in Python are available on GitHub and can serve as inspiration or a foundation for other mobile developers integrating machine learning into their apps.