Search

A complete collection of codes required to generate the results and diagrams presented in the paper: J. Wu, O. Fabrizio, D. O’Sullivan, and S. Dev, On the Use of Virtual Knowledge Graphs to Improve Environmental Sensor Data Accessibility, 2022, [UNDER REVIEW] Please refer to this github for the latest update.

Motivation: Several methods can be used to obtain the content of the biomedical literature articles annotated with standardized concepts. Various information retrieval (IR) applications exploit these resources, e.g. Literature-Based Discovery (LBD). Such applications often require a complex processing pipeline to transform the raw literature into an appropriate structured representation. Results: TDC Tools offer an intermediate level of representation aimed to facilitate the design and implementation of IR systems exploiting a concept-based view of the literature. Conceptually this intermediate representation decouples the data extraction part from the task-specific exploitation part, thus improving the modularity, interoperability and ultimately the reusability of such systems. Availability: The software and its dependencies are published under open-source license and provided with detailed instructions: https://github.com/erwanm/tdc-tools.

The General Data Protection Regulation (GDPR) has changed the ecosystem of services involving personal data and information. It emphasises several obligations and rights, amongst which the Right to Data Portability requires providing a copy of the given personal data in a commonly used, structured, and machine-readable format – for interoperability. The GDPR thus explicitly motivates the use and adoption of data interoperability concerning information. This chapter explores the entities and their interactions in the context of the GDPR to provide an information model for the development of interoperable services. The model categorises information and exchanges and explores existing standards and efforts towards use for interoperable interactions. The chapter concludes with an argument for the use and adoption of structured metadata to enable more expressive services through semantic interoperability.

The use of automated decision making systems to disclose personal data provokes privacy concerns as it is difficult for individuals to understand how and why these decisions are made. This research proposes an approach for empowering individuals to understand such automated access to personal data by utilising semantic web technologies to explain complex disclosure decisions in a comprehensible manner. We demonstrate the feasibility of our approach through a prototype that uses text and visual mediums to explain disclosure decisions made in the health domain and its evaluation through a user study.

Legal compliance is traditionally seen to be sufficiently demonstrable using legal documents that describe how various operations and activities follow a given set of obligations. The General Data Protection Regulation (GDPR) enforces larger responsibilities upon organisations and provides motivation for the use of technological measures that can ease its compliance. While there is no legal requirement to collaborate on compliance technologies or to use a common mechanism for defining knowledge, doing so has several benefits to the larger community. Through this paper, we describe how open and shared technologies targeted towards GDPR and its compliance can be used to create knowledge-based systems. Our approach uses semantic web technologies due to their open and flexible nature towards describing concepts and relationships. We present a model for such a knowledge-based system along with work published to date.

The calculation of data value is based on the assumption of continued presence of information. When the data in consideration concerns personal information, laws such as the European General Data Protection Regulation (GDPR) affect how this data can be exploited. Under the GDPR, data must be used on the basis of explicit consent, which can be withdrawn at any time by the concerned data subject. This posits that personal data dependant on consent can not be used after the consent is withdrawn. The data value associated with such personal data will also change with such a change in consent. This position paper explores the change in data value for personal data under the GDPR. The paper analyses the obligations and conditions provided by the GDPR for data usage and how they affect the data value chain. This is then used to identify potential approaches towards effective utilisation of available data through techniques such as aggregation and anonymisation. Finally, the paper proposes a method for investigating conditional data value by incorporating data existence and availability as a metric.

Privacy laws such as the General Data Protection Regulation(GDPR) specify several obligations involving personal data. A privacy policy is a document that provides information for legal compliance onhow personal data is collected, used, stored, and shared, which is essential for understanding their privacy implications. Approaches such as the Us-ablePrivacy project that extract information from the text of the privacy policy need to structure it in a manner suitable for machine processing.Semantic web has been proven to be suitable to represent this knowledge as a set of queryable concepts and relationships. However, there is a large overlap between different projects and approaches targeting the privacy policy that does not take advantage of the significant similarity of its underlying information. We present an ontology design pattern to aid these efforts in representing and modelling information related to personal data within a privacy policy. The pattern aims to assist the existing ecosystem of machine-based approaches for interpretation and visualisation of privacy policies by providing a common structured representation to ease modelling and sharing of related information.

This position paper explores changes concerning the relationship between consent and activities in context of the General Data Protection Regulation (GDPR). Detecting and recording such changes with their effects can provide assistance in demonstration and manage-ment of compliance. We present an approach for using metadata-driven change detection and representation towards supporting querying for GDPR compliance. We use P-Plan (an extension to PROV) for rep-resenting provenance of activities and ODRL for representing consent.We explore the presented approach by means of a use-case.

Laws such as the General Data Protection Regulation (GDPR) specify information to be shared, provided, or communicated between entities as part of its operation and compliance. Its compliance additionally requires provision of information between the entities to follow the requirements and procedures outlined in the text. GDPR also provides explicit cases of data interoperability such as that provided by the right of a data subject to move their personal data between different data controllers. In this paper, we explore the interoperability of information between the various entities mentioned within GDPR. We identify various procedures outlined for information flows which also contain explicit requirements such as presence of structured data or specific data formats being used and provide a discussion of existing standards by evaluating the state of the art with respect to the standards provided by the World Wide Web Consortium (W3C) for representing information.

The General Data Protection Regulations (GDPR) imposes greater restrictions on obtaining valid user consents involving the use of personal data. A semantic model of consent can make the concepts of consent explicit, establish a common understanding and enable re-use of consent. Therefore,forming a semantic model of consent will satisfy the GDPR requirements of specificity and unambiguity and is an important step towards ensuring compliance. In this paper,we discuss obtaining an open vocabulary of expressing consent leveraging existing semantic models of provenance, processes, permission and obligations. We also present a reference architecture for the management of data processing according to consent permission. This data management model utilizes the open vocabulary of consent and incorporates the change of context into the data processing activity. By identifying and incorporating changes to the relational context between data controllers and data subjects into the data processing model, it aims to improve the integration of data management across different information systems specifically adhering to the GDPR and helping controllers to demonstrate compliance.

Privacy policies are legal documents that describe activities over personal data such as its collection, usage, processing, sharing, and storage. Expressing this information as provenance metadata can aid in legal accountability as well as modelling of data usage in real-world use-cases. In this paper, we describe our early work on identification, extraction, and representation of provenance information within privacy policies. We discuss the adoption of entity extraction approaches using concepts and keywords defined by the GDPRtEXT resource along with using annotated privacy policy corpus from the UsablePrivacy project. We use the previously published GDPRov ontology (an extension of PROV-O) to model provenance model extracted from privacy policies.

Privacy laws such as the General Data Protection Regulation(GDPR) specify several obligations involving personal data. A privacy policy is a document that provides information for legal compliance onhow personal data is collected, used, stored, and shared, which is essential for understanding their privacy implications. Approaches such as the Us-ablePrivacy project that extract information from the text of the privacy policy need to structure it in a manner suitable for machine processing.Semantic web has been proven to be suitable to represent this knowledge as a set of queryable concepts and relationships. However, there is a large overlap between different projects and approaches targeting the privacy policy that does not take advantage of the significant similarity of its underlying information. We present an ontology design pattern to aid these efforts in representing and modelling information related to personal data within a privacy policy. The pattern aims to assist the existing ecosystem of machine-based approaches for interpretation and visualisation of privacy policies by providing a common structured representation to ease modelling and sharing of related information.

Privacy policies are legal documents that describe activities over personal data such as its collection, usage, processing, sharing, and storage. Expressing this information as provenance metadata can aid in legal accountability as well as modelling of data usage in real-world use-cases. In this paper, we describe our early work on identification, extraction, and representation of provenance information within privacy policies. We discuss the adoption of entity extraction approaches using concepts and keywords defined by the GDPRtEXT resource along with using annotated privacy policy corpus from the UsablePrivacy project. We use the previously published GDPRov ontology (an extension of PROV-O) to model provenance model extracted from privacy policies.

Internet services have become an important part of the daily life for a large number of people, and often deal with varying amounts of personal information. A privacy policy is a legal document governed by territorial laws that outlines the collection, usage, storage, and sharing of personal data. A known problem with such documents is its ambiguity and difficulty in comprehension for end users. The General Data Protection Regulation (GDPR) requires transparency regarding the provision of such information to the data subject through its various obligations and rights. We propose a remodelling of the privacy policy based on provision of relevant information regarding personal data specific to the user. Such a policy will dynamically reflect the state of activities over personal data using a legal and comprehensive document, and can be used as a tool for the provision of rights and requests from data subjects. We support our discussion with an example use-case of a GDPR-based privacy policy adopted from online services. We present our analysis on identifying changes and our approach towards the representation and creation of such dynamic policies.

Internet services have become an important part of the daily life for a large number of people, and often deal with varying amounts of personal information. A privacy policy is a legal document governed by territorial laws that outlines the collection, usage, storage, and sharing of personal data. A known problem with such documents is its ambiguity and difficulty in comprehension for end users. The General Data Protection Regulation (GDPR) requires transparency regarding the provision of such information to the data subject through its various obligations and rights. We propose a remodelling of the privacy policy based on provision of relevant information regarding personal data specific to the user. Such a policy will dynamically reflect the state of activities over personal data using a legal and comprehensive document, and can be used as a tool for the provision of rights and requests from data subjects. We support our discussion with an example use-case of a GDPR-based privacy policy adopted from online services. We present our analysis on identifying changes and our approach towards the representation and creation of such dynamic policies.

Citation. Navarro Gallinad, Albert, A Usable Knowledge Graph Framework for Linking Health Events with Environmental Data, Trinity College Dublin, School of Computer Science & Statistics, Computer Science, 2023. http://hdl.handle.net/2262/102971 Abstract Environmental exposures transported across air, land and water can affect our health making us more susceptible to developing a disease. Researchers studying these health-environment interactions integrate and link multiple and diverse data sources as part of their research workflows. Emerging technologies such as Knowledge Graphs (KG) can make the data integration process efficient for researchers by making the datasets interoperable. However, KG technologies are not easy to incorporate into researchers? workflows due to the technical knowledge and practical expertise required to access, explore and establish relevant links between the datasets. The major contribution of this PhD thesis is the proposed framework SERDIF (Semantic Environmental and Rare Disease data Integration Framework) that allows health data researchers themselves to directly link health data with relevant environmental data in support of their research workflows. SERDIF advances the state of the art in being the first usable KG framework, that is W3C standards-based, to be developed and implemented for the study of environmental triggers associated with rare diseases. This PhD thesis yielded two minor contributions towards improving the adoption of KG technologies and promoting transparency of research methods and data reuse towards improving the efficiency of scientific research. The first minor contribution is a step by step description of the methods and results of the evaluation approach, providing KG practitioners with a reproducible example in how to make their technologies usable for domain experts. The second minor contribution is a a collection of open source artefacts as a by-product during the development of SERDIF published to promote open science. While SERDIF has been implemented for rare disease studies, the framework has the potential to be used in other contexts to address the data integration challenges of environmental studies. Thesis available at http://hdl.handle.net/2262/102971, Navarro Gallinad, Albert, A Usable Knowledge Graph Framework for Linking Health Events with Environmental Data, Trinity College Dublin, School of Computer Science & Statistics, Computer Science, 2023. http://hdl.handle.net/2262/102971

submitted toSEMIC2022 IOPEU Academy corner