The need to manage large amounts of data, make useful inferences and deepen the fields of Artificial Intelligence, such as the field of Text Classification, are undoubtedly some of the most important goals of this science. This paper aims to categorize a set of news texts based on their content and title using basic Text Classification algorithms. More specifically, a study of the "News_Category Dataset" from the "kaggle" website, a dataset which includes 209,528 news text records collected from the global news website "Huffpost", was carried out. This dataset has for each record/text, various information such as the title and a short description of the text, its content-based genre, its link to the website as well as its author(s) and the date of publication. In total, the records are for published articles from 2012 to 2018. Based on the above study, the creation of an Artificial Neural Network using Deep Learning techniques is presented, which will be able to perform this task. The Neural Networks implemented in this work are RNN architecture, LSTM (Long Short-Term Memory) and Naive Bayes categorizer using TF-IDF technique. In total, the dataset contains 42 different categories of news texts, which are derived based on the content of each text. To process the data, 42 separate numeric labels were created to represent each category individually and the appropriate datasets were created for training and subsequent model testing. Within this project, the data preprocessing techniques followed, the corresponding codes for the implementation of the neural networks and the overall results of the above approaches are presented. This project was created for the purposes of the academic course of "Deep Learning" of the department of Computer Science and Biomedical Informatics, University of Thessaly.