Ioannis Kompatsiaris

User-Driven Ontology Population from Linked Data Sources

In order for ontology-based applications to be deployed in real-life scenarios, significant volumes of data are required to populate the underlying models. Populating ontologies manually is a time-consuming and error-prone task and, thus, research has shifted its attention to automatic ontology population methodologies. However, the majority of the proposed approaches and tools focus on analysing natural language text and often neglect other more appropriate sources of information, such as the already structured and semantically rich sets of Linked Data. The paper presents PROPheT, a novel ontology population tool for retrieving instances from Linked Data sources and subsequently inserting them into an OWL ontology. The tool, to the best of our knowledge, offers entirely novel ontology population functionality to a great extent and has already been positively received according to user evaluation.

Community detection in complex networks based on DBSCAN* and a Martingale process

Community detection is a valuable tool for analyzing and understanding the structure of complex networks. This work investigates the application of the density-based algorithm DBSCAN* to the community detection problem. Given, though, that this algorithm requires a lower bound for the community size to be determined a priori, this work proposes the application of a Martingale process to DBSCAN* so as to progressively detect communities at various levels of granularity, without the need to define in advance a single threshold for the minimum community size. In particular, the proposed DBSCAN*-Martingale community detection algorithm corresponds to an iterative process that progressively lowers the threshold of the size of the acceptable communities, while maintaining the communities detected for higher thresholds. Evaluation experiments are performed based on four realistic benchmark networks and the results indicate improvements in the effectiveness of the proposed DBSCAN*-Martingale community detection algorithm in terms of the Normalized Mutual Information and RAND metrics against several state-of-the-art community detection approaches.

Detection of Terrorism-related Twitter Communities using Centrality Scores

Social media are widely used among terrorists to communicate and disseminate their activities. User-to-user interaction (e.g. mentions, follows) leads to the formation of complex networks, with topology that reveals key-players and key-communities in the terrorism domain. Both the administrators of social media platforms and Law Enforcement Agencies seek to identify not only single users but groups of terrorism-related users so that they can reduce the impact of their information exchange efforts. To this end, we propose a novel framework that combines community detection with key-player identification to retrieve communities of terrorism-related social media users. Experiments show that most of the members of each retrieved key-community are already suspended by Twitter, violating its terms, and are hence associated with terrorism-oriented content with high probability.

Large-Scale Open Corporate Data Collection and Analysis as an Enabler of Corporate Social Responsibility Research

During the last years, citizens and transparency initiatives put increasing pressure on governments, organizations, and companies to be more transparent and to publicize information pertaining to their operations. Although several organizations have started engaging in open data practices, data quality, structure and availability is still highly inconsistent across organizations, which makes it challenging and effort-intensive to obtain and analyze large-scale high-quality datasets. To this end, this paper examines how publicly available financial and corporate data can be leveraged to extract useful inferences regarding the financial and social performance of companies. Numerous reports have been collected from the Securities Exchange Commission (SEC) and analyzed to study hypotheses regarding the corporate practices and social responsibility of companies.

The SemaDrift Protégé Plugin to Measure Semantic Drift in Ontologies: Lessons Learned

Semantic drift is an active research field, which aims to identify and measure changes in ontologies across time and versions. Yet, only few practical methods have emerged that are directly applicable to Semantic Web constructs, while the lack of relevant applications and tools is even greater. This paper presents the findings, current limitations and lessons learned throughout the development and the application of a novel software tool, developed in the context of the PERICLES FP7 project, which integrates currently investigated methods, such as text and structural similarity, into the popular ontology authoring platform, Protege. The graphical user interface provides knowledge engineers and domain experts with access to methods and results without prior programming knowledge. Its applicability and usefulness are validated through two proof-of-concept scenarios in the domains of Web Services and Digital Preservation; especially the latter is a field where such long-term insights are crucial.

Analysis of Suspended Terrorism-Related Content on Social Media

Social media are widely used by terrorist organizations and extremist groups for disseminating propaganda and recruiting new members. Given the recent pledges both by the major social media platforms and governments towards combating online terrorism, our work aims at understanding the terrorism-related content posted on social media and distinguishing accounts of relevance to terrorism investigations from innocuous ones. We conducted an analysis of textual, spatial, temporal and social network features on data and metadata gathered from suspended Twitter content, and compared them with non-suspended content. Our analysis reveals a number of distinct characteristics of terrorism-related Twitter accounts. This work is a first step towards automated tools for the early detection of terrorism-related and extremist content on Twitter.

Enhancing Virtual Learning Spaces: The Impact of the Gaming Analytics

Online virtual labs have been important to educational practice by providing students with distance courses that otherwise would be difficult to be offered. However, the majority of them cannot be easily applied to different courses or pedagogical approaches. In order to overcome this, we propose a high-level, easy-to-use authoring tool that will allow building course-independent high-standard virtual labs. This solution is based on learning and gaming analytics . Ιn the gaming industry, there have been developed strong game analytics methods and tools, which could be easily transferred into the learning domain. Game analytics monitor the users’ activity; model their current behavior through the use of shallow analytics and predict the future behavior of the users through the use of deep analytics. We propose that both of these approaches combined with visualization methodologies will offer insights on what features are important and what functionalities users expect to find in a virtual lab.

Adaptive Focused Crawling Using Online Learning: A Study on Content Related to Islamic Extremism

Focused crawlers aim to automatically discover online content resources relevant to a domain of interest by automatically navigating through the Web link structure and selecting which hyperlinks to follow based on an estimation of their relevance to the topic of interest; to this end, classifier-guided approaches are typically employed for identifying hyperlinks having the higher likelihood of leading to relevant content. However, the training data used for building these classifiers might not be entirely representative of the domain of interest, or the domain of interest might change over time. To meet these challenges, this work proposes a novel adaptive focused crawling framework that allows the classifiers that underlie the hyperlink selection policy to be adapted based on the evidence they encounter during their crawls. Our framework uses two different approaches to retrain its models: (i) Interactive Adaptation, where a user manually evaluates the discovered resources, and (ii) Automatic Adaptation, where the framework uses the already trained classifiers to assess the relevance of newly discovered resources. The evaluation experiments in the domain of Islamic extremism indicate the effectiveness of online learning in focused crawling.

A multimodal analytics platform for journalists analysing large-scale, heterogeneous multilingual and multimedia content

Analysts and journalists face the problem of having to deal with very large, heterogeneous, and multilingual data volumes that need to be analyzed, understood, and aggregated. Automated and simplified editorial and authoring process could significantly reduce time, labor, and costs. Therefore, there is a need for unified access to multilingual and multicultural news story material, beyond the level of a nation, ensuring context-aware, spatiotemporal, and semantic interpretation, correlating also and summarizing the interpreted material into a coherent gist. In this paper, we present a platform integrating multimodal analytics techniques, which are able to support journalists in handling large streams of real-time and diverse information. Specifically, the platform automatically crawls and indexes multilingual and multimedia information from heterogeneous resources. Textual information is automatically summarized and can be translated (on demand) into the language of the journalist. High-level information is extracted from both textual and multimedia content for fast inspection using concept clouds. The textual and multimedia content is semantically integrated and indexed using a common representation, to be accessible through a web-based search engine. The evaluation of the proposed platform was performed by several groups of journalists revealing satisfaction from the user side.

Flood Relevance Estimation from Visual and Textual Content in Social Media Streams.

Disaster monitoring based on social media posts has raised a lot of interest in the domain of computer science the last decade, mainly due to the wide area of applications in public safety and security and due to the pervasiveness not solely on daily communication but also in life-threating situations. Social media can be used as a valuable source for producing early warnings of eminent disasters. This paper presents a framework to analyse social media multimodal content, in order to decide if the content is relevant to flooding. This is very important since it enhances the crisis situational awareness and supports various crisis management procedures such as preparedness. Evaluation on a benchmark dataset shows very good performance in both text and image classification modules.