Symeon Papadopoulos

Sensing Trending Topics in Twitter

Online social and news media generate rich and timely information about real-world events of all kinds. However, the huge amount of data available, along with the breadth of the user base, requires a substantial effort of information filtering to successfully drill down to relevant topics and events. Trending topic detection is therefore a fundamental building block to monitor and summarize information originating from social sources. There are a wide variety of methods and variables and they greatly affect the quality of results. We compare six topic detection methods on three Twitter datasets related to major events, which differ in their time scale and topic churn rate. We observe how the nature of the event considered, the volume of activity over time, the sampling procedure and the pre-processing of the data all greatly affect the quality of detected topics, which also depends on the type of detection method used. We find that standard natural language processing techniques can perform well for social streams on very focused topics, but novel techniques designed to mine the temporal distribution of concepts are needed to handle more heterogeneous streams containing multiple stories evolving in parallel. One of the novel topic detection methods we propose, based on -grams cooccurrence and topic ranking, consistently achieves the best performance across all these conditions, thus being more reliable than other state-of-the-art techniques.

An empirical study on the combination of surf features with VLAD vectors for image search

The study of efficient image representations has attracted significant interest due to the computational needs of large-scale applications. In this paper we study the performance of the recently proposed VLAD method for aggregating local image descriptors when combined with SURF features, in the domain of image search. The experiments show that when SURF features are used as local image descriptors, VLAD attains better performance compared to using SIFT features. We also study how the average number of local image descriptors extracted per image affects the performance and show that by controlling this number we are able to adjust the trade off between feature extraction time and search accuracy. Finally, we examine the retrieval performance of the proposed scheme with varying levels of distractor images.

PScore: A Framework for Enhancing Privacy Awareness in Online Social Networks

The phenomenal increase in the use of social media in recent years has raised a number of issues related to privacy. In this paper, we propose a framework for raising the awareness of Online Social Network (OSN) users with respect to the information about them that is disclosed and that can be inferred by OSN service operators as well as by third parties that can access their data. This framework takes the form of a semantic, hierarchical scoring structure, that enables users to easily browse over different privacy-related aspects of their presence in a social network. Contrary to previous privacy scoring approaches, the proposed framework provides a finer and more intuitive organization of privacy information. Importantly, it also takes into account both information that is explicitly mentioned in users shared content, as well as implicit information, that may be inferred from it. We make available an open source implementation of the framework.

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.

Web Video Verification using Contextual Cues

As news agencies and the public increasingly rely on User-Generated Content, content verification is vital for news producers and consumers alike. We present a novel approach for verifying Web videos by analyzing their online context. It is based on supervised learning on contextual features: one feature set is based on an existing approach for tweet verification adapted to video comments. The other is based on video metadata, such as the video description, likes/dislikes, and uploader information. We evaluate both on a dataset of real and fake videos from YouTube, and demonstrate their effectiveness (F-scores: 0.82, 0.79). We then explore their complementarity and show that under an optimal fusion scheme, the classifier would reach an F-score of 0.9. We finally study the performance of the classifier through time, as more comments accumulate, emulating a real-time verification setting.

Open-Source Monitoring, Search and Analytics Over Social Media

The paper describes a technical demonstration of an open-source framework for monitoring, analysis and search over multiple social media platforms. The framework is intended to be a valuable tool for media intelligence professionals, as well as a framework and testbed for scientists and developers with interest in social media research.

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.

User Community Discovery

This book redefines community discovery in the new world of Online Social Networks and Web 2.0 applications, through real-world problems and applications in the context of the Web, pointing out the current and future challenges of the field. Particular emphasis is placed on the issues of community representation, efficiency and scalability, detection of communities in hypergraphs, such as multi-mode and multi-relational networks, characterization of social media communities and online privacy aspects of online communities. User Community Discovery is for computer scientists, data scientists, social scientists and complex systems researchers, as well as students within these disciplines, while the connections to real-world problem settings and applications makes the book appealing for engineers and practitioners in the industry, in particular those interested in the highly attractive fields of data science and big data analytics.

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.

Towards Improved Air Quality Monitoring Using Publicly Available Sky Images.

Air pollution causes nearly half a million premature deaths each year in Europe. Despite air quality directives that demand compliance with air pollution value limits, many urban populations continue being exposed to air pollution levels that exceed by far the guidelines. Unfortunately, official air quality sensors are sparse, limiting the accuracy of the provided air quality information. In this chapter, we explore the possibility of extending the number of air quality measurements that are fed into existing air quality monitoring systems by exploiting techniques that estimate air quality based on sky-depicting images. We first describe a comprehensive data collection mechanism and the results of an empirical study on the availability of sky images in social image sharing platforms and on webcam sites. In addition, we present a methodology for automatically detecting and extracting the sky part of the images leveraging deep learning models for concept detection and localization. Finally, we present an air quality estimation model that operates on statistics computed from the pixel color values of the detected sky regions.