Privacy concerns are constantly increasing in different sectors. Regulations such as the EU's General Data Protection Regulation (GDPR) are pressuring organizations to handle the individual's data with reinforced caution. As information systems deal with increasingly large amounts of personal data in essential services, there is a lack of mechanisms to help organizations in protecting the involved data subjects. In this paper, we propose and evaluate the use of Named Entity Recognition as a way to identify, monitor and validate Personally Identifiable Information. In our experiments, we used three of the most well-known Natural Language Processing tools (NLTK, Stanford CoreNLP, and spaCy). First, we assess the effectiveness of the tools with a generic dataset. Then, machine learning models are trained and evaluated with datasets built on data that contain personally identifiable information. The results show that models' performance was highly positive in accurately classifying both generic and more context-specific data. We observe the relationship between the datasets' training size and respective performance and estimate the appropriate size for model training within this context. Furthermore, we discuss how our proposal can effectively act as a Privacy Enhancing Technology as well as the potential risks and associated impacts.
The Internet of Things (IoT) has been increasingly adopted in domains such as smart infrastructure, healthcare, supply chain, transportation, and many others. However, the constrained computational resources of these devices make conventional security approaches against security threats not applicable. This limitation emphasizes the need to explore new approaches, specifically tailored to these kinds of devices. To increase protection against cyberattacks in IoT devices, Intrusion Detection Systems (IDSs) are considered an effective approach. Machine Learning (ML) techniques can be combined with Federated Learning to enhance the privacy and scalability of these systems. Many IDSs have been introduced, but there is a research gap concerning Host Intrusion Detection System (HIDS), which is the primary focus of our current work. Additionally, existing research predominantly focuses on the application of ML techniques and their evaluation, with limited attention to real-world implementation. We propose a lightweight HIDS that relies on the analysis of system call traces to detect malicious activities. The proposed HIDS achieved an accuracy rate of approximately 98%. Finally, by using eXplainable Artificial Intelligence methods, we sought to provide explanations for these these results.
Private data is transmitted and stored online every second. Therefore, security and privacy assurances should be provided at all times. However, that is not always the case. Private information is often unwillingly collected, sold, or exposed, depriving data owners of their rightful privacy. In this article, various privacy threats, concepts, regulations, and personal data types are analyzed. An overview of Privacy Enhancing Technologies (PETs) and a survey of anonymization mechanisms, privacy tools, models, and metrics are presented together with an analysis of respective characteristics and capabilities. Moreover, this article analyses the applicability of the reviewed privacy mechanisms on today's Cloud Services and identifies the current research challenges to achieve higher privacy levels in the Cloud.
As information systems deal with contracts and documents in essential services, there is a lack of mechanisms to help organizations in protecting the involved data subjects. In this paper, we evaluate the use of named entity recognition as a way to identify, monitor and validate personally identifiable information. In our experiments, we use three of the most well-known Natural Language Processing tools (NLTK, Stanford CoreNLP, and spaCy). First, the effectiveness of the tools is evaluated in a generic dataset. Then, the tools are applied in datasets built based on contracts that contain personally identifiable information. The results show that models' performance was highly positive in accurately classifying both the generic and the contracts' data. Furthermore, we discuss how our proposal can effectively act as a Privacy Enhancing Technology.