Please use this identifier to cite or link to this item:
http://hdl.handle.net/10609/138066
Title: | Malicious URL detection mediante técnicas de Deep Learning |
Author: | Francés Luesma, Óscar |
Tutor: | Hernández Jiménez, Enric |
Others: | Caparrós, Joan |
Abstract: | This paper shows the usefulness of Deep Learning techniques for the detection of URL malicious, which is one of the most used techniques in so-called social engineering attacks. Social engineering attacks are the practice of obtaining sensitive information through the manipulation of legitimate users. It consists of sending through a legitimate means of communication (for example, email) a link to a URL of a web page that has malicious code. The objective is for the user to click on the URL link to cause access to the malicious web page. At that time, the malicious code is executed with the user's permissions and can carry out an attack on the information system based on the permissions available to the user. The more privileges the user has, the greater the devastating effect the attack will have on the information system. The neural networks (Deep Learning) proposed in this project will detect potentially malicious URL with greater or lesser precision. To do this, we experiment with different types of neural networks and show which of them are more efficient for this type of attack. To achieve these objectives, it is an essential requirement to obtain a high-quality data set and the prior application of Machine Learning techniques to prepare such data. Training techniques and model validation will also be required. |
Keywords: | security privacy social engineering |
Document type: | info:eu-repo/semantics/masterThesis |
Issue Date: | Jan-2022 |
Publication license: | http://creativecommons.org/licenses/by-nc-nd/3.0/es/ |
Appears in Collections: | Bachelor thesis, research projects, etc. |
Files in This Item:
File | Description | Size | Format | |
---|---|---|---|---|
ofranceslTFM1221memoria.pdf | Memoria del TFM | 2,94 MB | Adobe PDF | View/Open |
Share:
This item is licensed under a Creative Commons License