Telegram bot for determining the location from a photo using machine learning tools
Students Name: Rudevych Dmytro Mykolaiovych
Qualification Level: magister
Speciality: Software Engineering
Institute: Institute of Computer Science and Information Technologies
Mode of Study: full
Academic Year: 2022-2023 н.р.
Language of Defence: ukrainian
Abstract: The purpose of this master’s qualification work is to research neural networks of deep learning and develop software for determining geolocation from images using a machine learning model. The main feature of this software is the ability to determine the approximate or exact location based on the input image. Telegram bot software is developed using C# programming language, .NET Core framework, Telegram.Bot and MetadataExtractor libraries. The Python programming language and the TensorFlow library were used to develop the machine learning model. Pandas Python and NumPy libraries were used to prepare data for model training. The software requirements specification is developed based on requirements analysis. The design, implementation and testing of the software solution was carried out in accordance with the developed specification. The work contains the results of research on machine learning methods and the development of a deep neural network for image recognition and classification, a software implementation of a Telegram bot, and an explanatory note with a total volume of 68 pages. Keywords: machine learning, convolutional neural networks, deep learning neural networks, image classification, photo geolocation.