The rise of the internet and social media has fundamentally transformed how communication occurs, enabling the rapid and widespread dissemination of information, often anonymously. This prevalence of anonymous information has led to an increase in the unacknowledged copying of text, posing significant risks of copyright infringement, where the authenticity of information is paramount. Authorship attribution (AA), is a crucial classification challenge in Natural Language Processing (NLP), aims to determine the authorship of texts, addressing these concerns by identifying the original creators of content. While extensive research has been conducted on long texts, the AA of short texts such as tweets remains challenging due to their concise nature and diverse styles. This chapter investigates the performance of machine learning (ML) methods with different feature extraction techniques for short texts. We employed three ML algorithms: Naive Bayes (NB), Support Vector Machine (SVM), and Logistic Regression (LR) compound with three feature extraction techniques: Bag of Words (BoW), TF-IDF, n-grams. The highest accuracy achieved with ML methods was 92.34% using an SVM with TF-IDF. This research not only advances the technical capabilities of AA but also extends its practical applications, providing tools that can be adapted across various domains to enhance the security and integrity of digital communications.