To plagiarize is to take the credit for another person's work. In the case of text, to plagiarize means including text fragments (and even entire documents) from other people in a document without giving the corresponding credit. Considering the volume of scientific and cultural information available on Internet, the temptation to take a shortcut and plagiarize is high. Due to this fact, it is highly relevant to develop methods that assist the detection of this text misuse. Automatic plagiarism detection is based on different techniques of Information Retrieval and Extraction as well as Pattern Recognition and Information Theory. It has received special attention in the last years due to the possibility of generating efficient mechanisms for the detection of plagiarism cases. The production of this kind of resources could minimise the temptation to plagiarize. In this talk an overview of some of the state-of-the-art in automatic plagiarism detection is presented. Special attention is paid to some of the techniques designed in the Natural Language Engineering Lab at the Technical University of Valencia.