In recent years, offensive, abusive and hateful language, sexism, racism and other types of aggressive and cyberbullying behavior have been manifesting with increased frequency, and in many online social media platforms. In fact, past scientific work focused on studying these forms in popular media, such as Facebook and Twitter. Building on such work, we present an 8-month study of the various forms of abusive behavior on Twitter, in a holistic fashion. Departing from past work, we examine a wide variety of labeling schemes, which cover different forms of abusive behavior, at the same time. We propose an incremental and iterative methodology, that utilizes the power of crowdsourcing to annotate a large scale collection of tweets with a set of abuse-related labels. In fact, by applying our methodology including statistical analysis for label merging or elimination, we identify a reduced but robust set of labels. Finally, we offer a first overview and findings of our collected and annotated dataset of 100 thousand tweets, which we make publicly available for further scientific exploration.
}, author = {Antigoni-Maria Founta and Constantinos Djouvas and Despoina Chatzakou and Ilias Leontiadis and Jeremy Blackburn and Gianluca Stringhini and Athena Vakali and Michael Sirivianos and Nicolas Kourtellis} } @proceedings {1953, title = {Class-based Prediction Errors to Categorize Text with Out-of-vocabulary Words}, series = {ALW1{\textquoteright}17}, year = {2017}, address = {Vancouver, Canada}, abstract = {Common approaches to text categorization essentially rely either on n-gram counts or on word embeddings. This presents important difficulties in highly dynamic or quickly-interacting environments, where the appearance of new words and/or varied misspellings is the norm. A paradigmatic example of this situation is abusive online behavior, with social networks and media platforms struggling to effectively combat uncommon or non-blacklisted hate words. To better deal with these issues in those fast-paced environments, we propose using the error signal of class-based language models as input to text classification algorithms. In particular, we train a next-character prediction model for any given class, and then exploit the error of such class-based models to inform a neural network classifier. This way, we shift from the ability to describe seen documents to the ability to predict unseen content. Preliminary studies using out-of-vocabulary splits from abusive tweet data show promising results, outperforming competitive text categorization strategies by 4{\textendash}11\%.
}, author = {Joan Serr{\`a} and Ilias Leontiadis and Dimitris Spathis and Gianluca Stringhini and Jeremy Blackburn and Athena Vakali} } @inproceedings {conf/pakdd/GiatsoglouCSFV15, title = {Retweeting Activity on Twitter: Signs of Deception}, booktitle = {PAKDD (1)}, series = {Lecture Notes in Computer Science}, volume = {9077}, year = {2015}, pages = {122-134}, publisher = {Springer}, organization = {Springer}, isbn = {978-3-319-18037-3}, author = {Maria Giatsoglou and Despoina Chatzakou and Neil Shah and Christos Faloutsos and Athena Vakali}, editor = {Cao, Tru and Lim, Ee-Peng and Zhou, Zhi-Hua and Ho, Tu-Bao and Cheung, David and Motoda, Hiroshi} } @inproceedings {conf/wise/GiatsoglouCV13, title = {Community Detection in Social Media by Leveraging Interactions and Intensities}, booktitle = {WISE (2)}, series = {Lecture Notes in Computer Science}, volume = {8181}, year = {2013}, pages = {57-72}, publisher = {Springer}, organization = {Springer}, keywords = {community detection, user weighted interaction networks}, isbn = {978-3-642-41153-3}, author = {Maria Giatsoglou and Despoina Chatzakou and Athena Vakali}, editor = {Lin, Xuemin and Manolopoulos, Yannis and Srivastava, Divesh and Huang, Guangyan} } @inproceedings {conf/mmm/PapadopoulosSKV13, title = {Semi-supervised Concept Detection by Learning the Structure of Similarity Graphs}, booktitle = {MMM (1)}, series = {Lecture Notes in Computer Science}, volume = {7732}, year = {2013}, pages = {1-12}, publisher = {Springer}, organization = {Springer}, abstract = {We present an approach for detecting concepts in images bya graph-based semi-supervised learning scheme. The proposed approach builds a similarity graph between both the labeled and unlabeled images of the collection and uses the Laplacian Eigemaps of the graph as features for training concept detectors. Therefore, it offers multiple options for fusing different image features. In addition, we present an incremental learning scheme that, given a set of new unlabeled images, efficiently performs the computation of the Laplacian Eigenmaps. We evaluate the performance of our approach both on synthetic datasets and on MIR Flickr, comparing it with high-performance state-of-the-art learning schemes with competitive and in some cases superior results.
}, isbn = {978-3-642-35725-1}, author = {Symeon Papadopoulos and Sagonas, Christos and Yiannis Kompatsiaris and Athena Vakali}, editor = {Li, Shipeng and El-Saddik, Abdulmotaleb and Wang, Meng and Mei, Tao and Sebe, Nicu and Yan, Shuicheng and Hong, Richang and Gurrin, Cathal} } @article {journals/tsmc/StamosLV12, title = {Mani-Web: Large-Scale Web Graph Embedding via Laplacian Eigenmap Approximation}, journal = {IEEE Transactions on Systems, Man, and Cybernetics, Part C}, volume = {42}, number = {6}, year = {2012}, pages = {879-888}, abstract = {The Web as a graph can be embedded in a lowdimensionalspace where its geometry can be visualized and studiedin order to mine interesting patterns such as web communities.The existing algorithms operate on small-to-medium-scalegraphs; thus, we propose a close to linear time algorithm calledMani-Web suitable for large-scale graphs. The result is similarto the one produced by the manifold-learning technique Laplacianeigenmap that is tested on artificial manifolds and real webgraphs.Mani-Web can also be used as a general-purpose manifoldlearning/dimensionality-reductiontechnique as long as the datacan be represented as a graph.
}, keywords = {Laplacian eigenmap, large scale, manifold learning, spectral graph theory, web communities}, author = {Stamos, Konstantinos and Laskaris, Nikolaos A. and Athena Vakali} } @inproceedings {conf/fia/SrivastavaV12, title = {Towards a Narrative-Aware Design Framework for Smart Urban Environments}, booktitle = {Future Internet Assembly}, series = {Lecture Notes in Computer Science}, volume = {7281}, year = {2012}, pages = {166-177}, publisher = {Springer}, organization = {Springer}, isbn = {978-3-642-30240-4}, author = {Srivastava, Lara and Athena Vakali}, editor = {Alvarez, Federico and Cleary, Frances and Daras, Petros and Domingue, John and Galis, Alex and Garcia, Ana and Gavras, Anastasius and Karnouskos, Stamatis and Krco, Srdjan and Li, Man-Sze and Lotz, Volkmar and M{\"u}ller, Henning and Salvadori, Elio and Sassen, Anne-Marie and Schaffers, Hans and Stiller, Burkhard and Tselentis, Georgios and Turkama, Petra and Zahariadis, Theodore B.} } @inproceedings {conf/fia/AnthopoulosV12, title = {Urban Planning and Smart Cities: Interrelations and Reciprocities}, booktitle = {Future Internet Assembly}, series = {Lecture Notes in Computer Science}, volume = {7281}, year = {2012}, pages = {178-189}, publisher = {Springer}, organization = {Springer}, isbn = {978-3-642-30240-4}, author = {Anthopoulos, Leonidas G. and Athena Vakali}, editor = {Alvarez, Federico and Cleary, Frances and Daras, Petros and Domingue, John and Galis, Alex and Garcia, Ana and Gavras, Anastasius and Karnouskos, Stamatis and Krco, Srdjan and Li, Man-Sze and Lotz, Volkmar and M{\"u}ller, Henning and Salvadori, Elio and Sassen, Anne-Marie and Schaffers, Hans and Stiller, Burkhard and Tselentis, Georgios and Turkama, Petra and Zahariadis, Theodore B.} } @inproceedings {conf/mediaeval/PapadopoulosZKV11, title = {CERTH @ MediaEval 2011 Social Event Detection Task}, booktitle = {MediaEval}, series = {CEUR Workshop Proceedings}, volume = {807}, year = {2011}, publisher = {CEUR-WS.org}, organization = {CEUR-WS.org}, abstract = {This paper describes the participation of CERTH in the {\^a}{\texteuro}{\'s}SocialEvent Detection Task @ MediaEval 2011{\^a}{\texteuro}{\v t}, which aimsat discovering social events in a large photo collection. Thetask comprises two challenges: (i) identification of soccerevents in the cities of Barcelona and Rome, and (ii) identificationof events taking place in two specific venues. Weadopt an approach that combines spatial and temporal filterswith tag-based location classification models and an ef-ficient photo clustering method. In our best runs, we achieveF-measure and NMI scores of 77.4\% and 0.63 respectivelyfor Challenge 1, and 64\% and 0.38 for Challenge 2.
}, author = {Symeon Papadopoulos and Christos Zigkolis and Yiannis Kompatsiaris and Athena Vakali}, editor = {Larson, Martha and Rae, Adam and Demarty, Claire-Helene and Kofler, Christoph and Metze, Florian and Troncy, Rapha{\"e}l and Mezaris, Vasileios and Jones, Gareth J. F.} } @inbook {books/daglib/p/NikolopoulosGKPV11, title = {Combining Multi-modal Features for Social Media Analysis}, booktitle = {Social Media Modeling and Computing}, year = {2011}, pages = {71-96}, publisher = {Springer}, organization = {Springer}, isbn = {978-0-85729-435-7}, author = {Nikolopoulos, Spiros and Giannakidou, Eirini and Yiannis Kompatsiaris and Patras, Ioannis and Athena Vakali}, editor = {Hoi, Steven C. H. and Luo, Jiebo and Boll, Susanne and Xu, Dong and Jin, Rong} } @inproceedings {conf/vsgames/ZigkolisKCKGKV11, title = {Towards a User-Aware Virtual Museum}, booktitle = {VS-GAMES}, year = {2011}, pages = {228-235}, publisher = {IEEE Computer Society}, organization = {IEEE Computer Society}, keywords = {user groups, user preferences, virtual museum}, isbn = {978-1-4577-0316-4}, author = {Christos Zigkolis and Vassiliki A. Koutsonikola and Despoina Chatzakou and Karagiannidis, Savvas and Maria Giatsoglou and Kosmatopoulos, Andreas and Athena Vakali}, editor = {Liarokapis, Fotis and Doulamis, Anastasios D. and Vescoukis, Vassilios} } @inbook {reference/db/Vakali09, title = {Access Control Policy Languages}, booktitle = {Encyclopedia of Database Systems}, year = {2009}, pages = {15-18}, publisher = {Springer US}, organization = {Springer US}, isbn = {978-0-387-39940-9}, author = {Athena Vakali}, editor = {Liu, Ling and Ozsu, M. Tamer} } @inproceedings {conf/wise/KoutsonikolaVGK09, title = {Clustering of Social Tagging System Users: A Topic and Time Based Approach}, booktitle = {WISE}, series = {Lecture Notes in Computer Science}, volume = {5802}, year = {2009}, pages = {75-86}, publisher = {Springer}, organization = {Springer}, abstract = {Under Social Tagging Systems, a typical Web 2.0 application,users label digital data sources by using freely chosen textual descriptions(tags). Mining tag information reveals the topic-domain ofusers interests and significantly contributes in a profile construction process.In this paper we propose a clustering framework which groups usersaccording to their preferred topics and the time locality of their taggingactivity. Experimental results demonstrate the efficiency of the proposedapproach which results in more enriched time-aware users profiles.
}, keywords = {Social tagging systems, time, topic, user clustering}, isbn = {978-3-642-04408-3}, author = {Vassiliki A. Koutsonikola and Athena Vakali and Giannakidou, Eirini and Yiannis Kompatsiaris}, editor = {Vossen, Gottfried and Long, Darrell D. E. and Yu, Jeffrey Xu} } @inbook {reference/db/Vakali09a, title = {Security Services}, booktitle = {Encyclopedia of Database Systems}, year = {2009}, pages = {2546-2547}, publisher = {Springer US}, organization = {Springer US}, isbn = {978-0-387-39940-9}, author = {Athena Vakali}, editor = {Liu, Ling and Ozsu, M. Tamer} } @article {journals/jdim/HammicheLBHV07, title = {Domain Knowledge Based Queries for Multimedia Data Retrieval}, journal = {JDIM}, volume = {5}, number = {2}, year = {2007}, pages = {75-81}, abstract = {This paper describes an approach for semantic description and retrieval of multimedia data described by means ofMPEG-7. This standard uses XML schema to define the descriptions. Therefore, it lacks ability to represent the data semanticsin a formal and concise way and it does not allow integration and use of domain specific knowledge. Moreover,inference mechanisms are not provided and hence the extraction of implicit information is not (always) possible. To addressthese issues, we propose to add a conceptual layer on top of MPEG-7 metadata layer, where the domain knowledgeis represented using a formal language. A set of mapping rules is proposed. They serve as a bridge between the twolayers.Querying MPEG-7 descriptions using XML query languages such as XPath or XQuery requires to know MPEG-7syntax and documents structure. To provide a flexible query formulation, we exploit the conceptual layer vocabularyto express user queries. A user query, making reference to terms specified at the conceptual level, is rewritten into anXQuery expression over MPEG-7 descriptions.
}, keywords = {Logic Languages, Mapping Rules, MPEG-7, Multimedia Data Descriptions, Ontology, Semantic and Structural Aspects}, author = {Hammiche, Samira and Lopez, Bernardo and Benbernou, Salima and Hacid, Mohand-Said and Athena Vakali} } @inproceedings {conf/iscis/StoupaSV06, title = {Credential-Based Policies Management in an Access Control Framework Protecting XML Resources}, booktitle = {ISCIS}, series = {Lecture Notes in Computer Science}, volume = {4263}, year = {2006}, pages = {603-612}, publisher = {Springer}, organization = {Springer}, abstract = {XML has been widely adopted for Web data representation undervarious applications (such as DBMSs, Digital Libraries etc). Therefore, accessto XML data sources has become a crucial issue. In this paper we introduce acredential-based access control framework for protecting XML resources. Underthis framework, we propose the use of access policy files containing policiesconcerning a specific credentials type. Moreover, we propose the reorganizationof the policies in these files based on their frequency of use (the morefrequently it is used the higher in the file it is placed). Our main goal is to improverequest servicing times. Several experiments have been conducted whichare carried out either on single request or on multiple requests base. The proposedframework is proven quite beneficial for protecting XML-based frameworkssuch as digital libraries or any other data resources whose format is expressedin XML.
}, isbn = {3-540-47242-8}, author = {Stoupa, Konstantina and Simeoforidis, Zisis and Athena Vakali}, editor = {Levi, Albert and Savas, Erkay and Yenig{\"u}n, H{\"u}sn{\"u} and Balcisoy, Selim and Saygin, Y{\"u}cel} } @inproceedings {conf/iccsa/PetridouKVP06, title = {A Divergence-Oriented Approach for Web Users Clustering}, booktitle = {ICCSA (2)}, series = {Lecture Notes in Computer Science}, volume = {3981}, year = {2006}, pages = {1229-1238}, publisher = {Springer}, organization = {Springer}, abstract = {Clustering web users based on their access patterns is a quite significanttask in Web Usage Mining. Further to clustering it is important to evaluatethe resulted clusters in order to choose the best clustering for a particular framework.This paper examines the usage of Kullback-Leibler divergence, aninformation theoretic distance, in conjuction with the k-means clusteringalgorithm. It compares KL-divergence with other well known distance measures(Euclidean, Standardized Euclidean and Manhattan) and evaluates clusteringresults using both objective function{\^a}{\texteuro}{\texttrademark}s value and Davies-Bouldin index.Since it is imperative to assess whether the results of a clustering process aresusceptible to noise, especially in noisy environments such as Web environment,our approach takes the impact of noise into account. The clusters obtainedwith KL approach seem to be superior to those obtained with the otherdistance measures in case our data have been corrupted by noise.}, isbn = {3-540-34072-6}, author = {Petridou, Sophia G. and Vassiliki A. Koutsonikola and Athena Vakali and Papadimitriou, Georgios I.}, editor = {Gavrilova, Marina L. and Gervasi, Osvaldo and Kumar, Vipin and Tan, Chih Jeng Kenneth and Taniar, David and Lagan{\u A} , Antonio and Mun, Youngsong and Choo, Hyunseung} } @proceedings {conf/edbtw/2004, title = {Current Trends in Database Technology {\^a}{\texteuro}{\textquotedblleft} EDBT 2004 Workshops, EDBT 2004 Workshops PhD, DataX, PIM, P2P\&DB, and ClustWeb, Heraklion, Crete, Greece, March 14-18, 2004, Revised Selected Papers}, booktitle = {EDBT Workshops}, series = {Lecture Notes in Computer Science}, volume = {3268}, year = {2004}, publisher = {Springer}, isbn = {3-540-23305-9}, editor = {Lindner, Wolfgang and Mesiti, Marco and T{\"u}rker, Can and Tzitzikas, Yannis and Athena Vakali} } @inproceedings {conf/edbtw/VakaliPD04, title = {An Overview of Web Data Clustering Practices}, booktitle = {EDBT Workshops}, series = {Lecture Notes in Computer Science}, volume = {3268}, year = {2004}, pages = {597-606}, publisher = {Springer}, organization = {Springer}, abstract = {Clustering is a challenging topic in the area of Web data management.Various forms of clustering are required in a wide range of applications, includingfinding mirrored Web pages, detecting copyright violations, and reporting searchresults in a structured way. Clustering can either be performed once offline, (independentlyto search queries), or online (on the results of search queries). Importantefforts have focused on mining Web access logs and to cluster search engine resultson the fly. Online methods based on link structure and text have been appliedsuccessfully to finding pages on related topics. This paper presents an overview ofthe most popular methodologies and implementations in terms of clustering eitherWeb users or Web sources and presents a survey about current status and futuretrends in clustering employed over the Web.
}, keywords = {Web Data Clustering}, isbn = {3-540-23305-9}, author = {Athena Vakali and Pokorny, Jaroslav and Dalamagas, Theodore}, editor = {Lindner, Wolfgang and Mesiti, Marco and T{\"u}rker, Can and Tzitzikas, Yannis and Athena Vakali} } @inproceedings {conf/smc/StoupaVLA04, title = {Web-based delegation using XML}, booktitle = {SMC (6)}, year = {2004}, pages = {5189-5194}, publisher = {IEEE}, organization = {IEEE}, abstract = {Existing access control mechanisms should beextended in order to authorize external (and possiblyunknown) clients, when entering distributedenvironments. This paper proposes the structure andissuing of appropriate authorization certificate to supportthe delegation process under a role-based access controlenvironment. The proposed processes aim to enhanceaccessing automation and to avoid (centraladministrator) bottlenecks (in cases of altering anauthorization or a policy). The delegation requests andthe certificates are expressed according to the XMLsyntax for enhancing the interoperability of thedelegation processes, which is highlighted in a step-bystepalgorithmic fashion using flowcharts.
}, keywords = {Delegation, XML access control}, isbn = {0-7803-8566-7}, author = {Stoupa, Konstantina and Athena Vakali and Li, Fang and Andreadis, George} } @inproceedings {conf/iceis/ZengLV04, title = {An XML-Based Bootstrapping Method for Pattern Acquisition}, booktitle = {ICEIS (2)}, year = {2004}, pages = {303-308}, author = {Zeng, Xingjie and Li, Fang and Zhang, Dongmo and Athena Vakali} } @inproceedings {conf/edbtw/StoupaVLT04, title = {XML-Based Revocation and Delegation in a Distributed Environment}, booktitle = {EDBT Workshops}, series = {Lecture Notes in Computer Science}, volume = {3268}, year = {2004}, pages = {299-308}, publisher = {Springer}, organization = {Springer}, abstract = {The rapid increase on the circulation of data over the web has highlightedthe need for distributed storage of Internet-accessible information due tothe rapid increase on the circulation of data over the web. Thus, access controlmechanisms should also be distributed in order to protect them effectively. A recentidea in the access control theory is the delegation and revocation of rights,i.e. the passing over of one clients rights to the other and vice versa. Here, wepropose an XML-based distributed delegation module which can be integratedinto a distributed role-based access control mechanism protecting networks. Theidea of X.509v3 certificates is used for the transfer of authorization informationreferring to a client. The modules are XML-based and all of the associated datastructures are expressed through Document Type Definitions (DTDs).
}, isbn = {3-540-23305-9}, author = {Stoupa, Konstantina and Athena Vakali and Li, Fang and Tsoukalas, Ioannis}, editor = {Lindner, Wolfgang and Mesiti, Marco and T{\"u}rker, Can and Tzitzikas, Yannis and Athena Vakali} } @inproceedings {conf/adbis/VakaliM98, title = {Replication in Mirrored Disk Systems}, booktitle = {ADBIS}, series = {Lecture Notes in Computer Science}, volume = {1475}, year = {1998}, pages = {224-235}, publisher = {Springer}, organization = {Springer}, abstract = {In this paper we study data replication in a mirrored disk system.Free disk space is exploited by keeping replicas of specific cylindersat appropriate disk locations. Assuming an organ-pipe arrangement wecalculate the expected seek distance by varying the probability cylinderaccess under different distributions. Also, analytic formulae are derivedfor the expected seek distance under replication and comparison with theconventional (without replication) mirrored disk system is performed.}, isbn = {3-540-64924-7}, author = {Athena Vakali and Manolopoulos, Yannis}, editor = {Litwin, Witold and Morzy, Tadeusz and Vossen, Gottfried} }