As smart cities infrastructures mature, data becomes a valuable asset which can radically improve city services and tools. Registration, acquisition and utilization of data, which will be transformed into smart services, are becoming more necessary than ever. Online social networks with their enormous momentum are one of the main sources of urban data offering heterogeneous real-time data at a minimal cost. However, various types of attacks often appear on them, which risk users{\textquoteright} privacy and affect their online trust. The purpose of this article is to investigate how risks on online social networks affect smart cities and study the differences between privacy and security threats with regard to smart people and smart living dimensions.
}, keywords = {online social networks, privacy threats, security threats, smart cities, smart living, smart people}, doi = {https://doi.org/10.1145/3184558.3191516 }, author = {Moustaka, Vaia and Zenonas Theodosiou and Athena Vakali and Anastasis Kounoudes} } @article {1967, title = {A Systematic Review for Smart City Data Analytics}, journal = {Computing Surveys (CSUR)}, year = {2018}, doi = {https://doi.org/10.1145/3239566}, author = {Moustaka, Vaia and Vakali, Athena and Anthopoulos, Leonidas G.} } @article {1928, title = {Sentiment analysis leveraging emotions and word embeddings}, journal = {Expert Systems with Applications}, volume = {69}, year = {2017}, pages = {214 - 224}, abstract = {Abstract Sentiment analysis and opinion mining are valuable for extraction of useful subjective information out of text documents. These tasks have become of great importance, especially for business and marketing professionals, since online posted products and services reviews impact markets and consumers shifts. This work is motivated by the fact that automating retrieval and detection of sentiments expressed for certain products and services embeds complex processes and pose research challenges, due to the textual phenomena and the language specific expression variations. This paper proposes a fast, flexible, generic methodology for sentiment detection out of textual snippets which express people{\textquoteright}s opinions in different languages. The proposed methodology adopts a machine learning approach with which textual documents are represented by vectors and are used for training a polarity classification model. Several documents{\textquoteright} vector representation approaches have been studied, including lexicon-based, word embedding-based and hybrid vectorizations. The competence of these feature representations for the sentiment classification task is assessed through experiments on four datasets containing online user reviews in both Greek and English languages, in order to represent high and weak inflection language groups. The proposed methodology requires minimal computational resources, thus, it might have impact in real world scenarios where limited resources is the case.
}, keywords = {Online user reviews}, issn = {0957-4174}, doi = {http://dx.doi.org/10.1016/j.eswa.2016.10.043}, url = {http://www.sciencedirect.com/science/article/pii/S095741741630584X}, author = {Maria Giatsoglou and Manolis G. Vozalis and Konstantinos Diamantaras and Athena Vakali and George Sarigiannidis and Konstantinos Ch. Chatzisavvas} } @inproceedings {1923, title = {Smart Cities Tales and Trails}, booktitle = {Internet Science - Third International Conference, INSCI 2016, Florence, Italy, September 12-14, 2016, Proceedings}, year = {2016}, doi = {10.1007/978-3-319-45982-0_24}, url = {http://dx.doi.org/10.1007/978-3-319-45982-0_24}, author = {Athena Vakali and Angeliki Milonaki and Ioannis Gkrosdanis} } @inproceedings {conf/wims/VakaliAK14, title = {Smart Cities Data Streams Integration: experimenting with Internet of Things and social data flows}, booktitle = {WIMS}, year = {2014}, pages = {60}, publisher = {ACM}, organization = {ACM}, isbn = {978-1-4503-2538-7}, author = {Athena Vakali and Anthopoulos, Leonidas G. and Krco, Srdjan}, editor = {Akerkar, Rajendra and Bassiliades, Nick and Davies, John and Ermolayev, Vadim} } @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} } @inproceedings {conf/icc/VakaliAG13, title = {Sensors talk and humans sense Towards a reciprocal collective awareness smart city framework}, booktitle = {ICC Workshops}, year = {2013}, pages = {189-193}, publisher = {IEEE}, organization = {IEEE}, abstract = {Smart city infrastructures provide unique opportunities for innovative applications developing and testing. Sensor city installations offer the ground for experimenting with user-oriented services, which at the same time can test and improve the infrastructure itself. The proposed work summarizes principles and methodology for and experiment, entitled SEN2SOC which will bridge sensor measurements and social networks interactions via natural language generation for supporting smart city services. SEN2SOC aims at exploiting the SmartSantander infrastructure in a sensor to social reciprocal fashion such that the sensor measurements will be and communicated to the public (citizens,authorities, etc), while social networks users activities in relevance to sensors social postings will be analyzed and summarized both to verify sensors reporting and to develop collective aware applications.
}, keywords = {collective aware applications, sensors data management, smart city, social networks mining}, author = {Athena Vakali and Angelis, Lefteris and Maria Giatsoglou} } @inproceedings {conf/data/VakaliCKA13, title = {Social Data Sentiment Analysis in Smart Environments - Extending Dual Polarities for Crowd Pulse Capturing}, booktitle = {DATA}, year = {2013}, pages = {175-182}, publisher = {SciTePress}, organization = {SciTePress}, isbn = {978-989-8565-67-9}, author = {Athena Vakali and Despoina Chatzakou and Vassiliki A. Koutsonikola and Andreadis, George}, editor = {Helfert, Markus and Francalanci, Chiara and Filipe, Joaquim} } @inproceedings {conf/www/VakaliGA12, title = {Social networking trends and dynamics detection via a cloud-based framework design}, booktitle = {WWW (Companion Volume)}, year = {2012}, pages = {1213-1220}, publisher = {ACM}, organization = {ACM}, keywords = {cloud service deployment, microblogs and blogosphere dynamics, Social networks social, Web Data Clustering}, isbn = {978-1-4503-1230-1}, author = {Athena Vakali and Maria Giatsoglou and Antaris, Stefanos}, editor = {Mille, Alain and Gandon, Fabien L. and Misselis, Jacques and Rabinovich, Michael and Staab, Steffen} } @inproceedings {1907, title = {Social Networking Trends and Dynamics Detection via a Cloud-based Framework Design}, year = {2012}, abstract = {Social networking media generate huge content streams, whichleverage, both academia and developers efforts in providingunbiased, powerful indications of users{\textquoteright} opinion and interests.Here, we present Cloud4Trends, a framework for collecting andanalyzing user generated content through microblogging andblogging applications, both separately and jointly, focused oncertain geographical areas, towards the identification of the mostsignificant topics using trend analysis techniques. The cloudcomputing paradigm appears to offer a significant benefit in orderto make such applications viable considering that the massive datasizes produced daily impose the need of a scalable and powerfulinfrastructure. Cloud4Trends constitutes an efficient Cloud-basedapproach in order to solve the online trend tracking problem basedon Web 2.0 sources. A detailed system architecture model is alsoproposed, which is largely based on a set of service modulesdeveloped within the VENUS-C research project to facilitate thedeployment of research applications on Cloud infrastructures.
} } @inproceedings {1904, title = {Social Web Mashups Full Completion via Frequent Sequence Mining}, year = {2011}, abstract = {In this paper we address the problem of WebMashups full completion which consists of predicting themost suitable set of (combined) services that successfully meetthe goals of an end-user Mashup, given the current service(or composition of services) initially supplied. We model fullcompletion as a frequent sequence mining problem and weshow how existing algorithms can be applied in this context.To overcome some limitations of the frequent sequence miningalgorithms, e.g., efficiency and recommendation granularity,we propose FESMA, a new and efficient algorithm for computingfrequent sequences of services and recommending completions.FESMA also integrates a social dimension, extractedfrom the transformation of user {\textrightarrow} service interactions intouser {\textrightarrow} user interactions, building an implicit graph thathelps to better predict completions of services in a fashiontailored to individual users. Evaluations show that FESMAis more efficient outperforming the existing algorithms evenwith the consideration of the social dimension. Our proposalhas been implemented in a prototype, SoCo, developed at BellLabs.
} } @inproceedings {conf/services/MaaradjiHSV11, title = {Social Web Mashups Full Completion via Frequent Sequence Mining}, booktitle = {SERVICES}, year = {2011}, pages = {9-16}, publisher = {IEEE Computer Society}, organization = {IEEE Computer Society}, abstract = {In this paper we address the problem of WebMashups full completion which consists of predicting themost suitable set of (combined) services that successfully meetthe goals of an end-user Mashup, given the current service(or composition of services) initially supplied. We model fullcompletion as a frequent sequence mining problem and weshow how existing algorithms can be applied in this context.To overcome some limitations of the frequent sequence miningalgorithms, e.g., efficiency and recommendation granularity,we propose FESMA, a new and efficient algorithm for computingfrequent sequences of services and recommending completions.FESMA also integrates a social dimension, extractedfrom the transformation of user ? service interactions intouser ? user interactions, building an implicit graph thathelps to better predict completions of services in a fashiontailored to individual users. Evaluations show that FESMAis more efficient outperforming the existing algorithms evenwith the consideration of the social dimension. Our proposalhas been implemented in a prototype, SoCo, developed at BellLabs.
}, keywords = {Mashups, Sequence mining, Social networks, Web services}, isbn = {978-1-4577-0879-4}, author = {Maaradji, Abderrahmane and Hacid, Hakim and Skraba, Ryan and Athena Vakali} } @inproceedings {1906, title = {Summarization meets Visualization on Online Social Networks}, year = {2011}, abstract = {Getting an overview of a large online social networkand deciding which communities to join is a challengingtask for a new user. We propose a method that maps a largenetwork into a smaller graph with two kinds of nodes: a nodeof the first kind is representative of a community; a node ofthe second kind is neighbor to a representative and reflectsthe semantics of that community. Our approach encompassesa learning and ranking algorithm that derives this smallergraph from the original one, and a visualization algorithmthat returns a graph layout to the observer. We report on ourresults on inspecting the network of a folksonomy.
} } @inproceedings {conf/webi/GabrielSSV11, title = {Summarization Meets Visualization on Online Social Networks}, booktitle = {Web Intelligence}, year = {2011}, pages = {475-478}, publisher = {IEEE Computer Society}, organization = {IEEE Computer Society}, abstract = {Getting an overview of a large online social networkand deciding which communities to join is a challengingtask for a new user. We propose a method that maps a largenetwork into a smaller graph with two kinds of nodes: a nodeof the first kind is representative of a community; a node ofthe second kind is neighbor to a representative and reflectsthe semantics of that community. Our approach encompassesa learning and ranking algorithm that derives this smallergraph from the original one, and a visualization algorithmthat returns a graph layout to the observer. We report on ourresults on inspecting the network of a folksonomy.
}, keywords = {Clustering, communities, community representatives, social network summarization, social network visualization, Social networks, visualization}, isbn = {978-0-7695-4513-4}, author = {Gabriel, Hans-Henning and Spiliopoulou, Myra and Stachtiari, Emmanouela and Athena Vakali}, editor = {Boissier, Olivier and Benatallah, Boualem and Papazoglou, Mike P. and Ras, Zbigniew W. and Hacid, Mohand-Said} } @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} } @inproceedings {conf/semco/GiannakidouKV08, title = {SEMSOC: SEMantic, SOcial and Content-Based Clustering in Multimedia Collaborative Tagging Systems}, booktitle = {ICSC}, year = {2008}, pages = {128-135}, publisher = {IEEE Computer Society}, organization = {IEEE Computer Society}, isbn = {978-0-7695-3279-0}, author = {Giannakidou, Eirini and Yiannis Kompatsiaris and Athena Vakali} } @inproceedings {1864, title = {SEMSOC: Semantics Mining on Multimedia Social Data Sources}, year = {2008}, abstract = {A huge amount of data and metadata emerges fromWeb 2.0 applications which have transformed the Webto a mass social interaction and collabo-ration medium.Collaborative Tagging Systems is a typical,popular and promising Web 2.0 application and despiteits adoption it faces some serious limitations thatrestrict their usability. These limitations (no structureon tags, tags validation, spamming and redundancy)are more evident in the case of multi-media contentdue to its challenging automatic annotation and retrievalrequirements. In this paper, we present an approachfor social data clustering which combinesjointly semantic, social and content-based information.We propose an unsupervised model for efficient andscalable mining on multimedia social-related data,which leads to the extraction of rich and trustworthysemantics and the improvement of retrieval in a socialtagging system. Experimental results demonstrate theefficiency of the proposed approach.
} } @inproceedings {1862, title = {A Structure-Based Clustering on LDAP Directory Information}, year = {2008}, abstract = {LDAP directories have rapidly emerged as the essentialframework for storing a wide range of heterogeneous information undervarious applications and services. Increasing amounts of informationare being stored in LDAP directories imposing the need for efficientdata organization and retrieval. In this paper, we propose the LPAIR\& LMERGE (LP-LM) hierarchical agglomerative clustering algorithmfor improving LDAP data organization. LP-LM merges a pair of clustersat each step, considering the LD-vectors, which represent the entries{\textquoteright}structure. The clustering-based LDAP data organization enhances LDAPserver{\textquoteright}s response times, under a specific query framework.
} } @inproceedings {conf/ismis/KoutsonikolaVMV08, title = {A Structure-Based Clustering on LDAP Directory Information}, booktitle = {ISMIS}, series = {Lecture Notes in Computer Science}, volume = {4994}, year = {2008}, pages = {121-130}, publisher = {Springer}, organization = {Springer}, abstract = {LDAP directories have rapidly emerged as the essentialframework for storing a wide range of heterogeneous information undervarious applications and services. Increasing amounts of informationare being stored in LDAP directories imposing the need for efficientdata organization and retrieval. In this paper, we propose the LPAIR\& LMERGE (LP-LM) hierarchical agglomerative clustering algorithmfor improving LDAP data organization. LP-LM merges a pair of clustersat each step, considering the LD-vectors, which represent the entries{\^a}{\texteuro}{\texttrademark}structure. The clustering-based LDAP data organization enhances LDAPserver{\^a}{\texteuro}{\texttrademark}s response times, under a specific query framework.
}, isbn = {978-3-540-68122-9}, author = {Vassiliki A. Koutsonikola and Athena Vakali and Mpalasas, Antonios and Valavanis, Michael}, editor = {An, Aijun and Matwin, Stan and Ras, Zbigniew W. and Slezak, Dominik} } @inproceedings {conf/ideas/StamosPTV06, title = {A similarity based approach for integrated Web caching and content replication in CDNs}, booktitle = {IDEAS}, year = {2006}, pages = {239-242}, publisher = {IEEE Computer Society}, organization = {IEEE Computer Society}, abstract = {Web caching and content replication techniques emergedto solve performance problems related to the Web. We proposea generic non-parametric heuristic method that integratesboth techniques under a CDN. We provide experimentationshowing that our method outperforms the so farseparate implementations of Web caching and content replication.Moreover, we show that the performance improvementcompared with an existing algorithm is significant. Wetest all these techniques in a simulation environment undera flash crowd event and a workload of a typical lightweightedCDN operation.
}, author = {Stamos, Konstantinos and Pallis, George and Thomos, Charilaos and Athena Vakali}, editor = {Desai, Bipin C. and Gupta, Shyam K.} } @inproceedings {1858, title = {A similarity based approach for integrated Web caching and content replication in CDNs}, year = {2006}, abstract = {Web caching and content replication techniques emergedto solve performance problems related to the Web. We proposea generic non-parametric heuristic method that integratesboth techniques under a CDN. We provide experimentationshowing that our method outperforms the so farseparate implementations of Web caching and content replication.Moreover, we show that the performance improvementcompared with an existing algorithm is significant. Wetest all these techniques in a simulation environment undera flash crowd event and a workload of a typical lightweightedCDN operation.
} } @inbook {books/idea/encyclopedia2005/PallisSV05, title = {Storage and Access Control Issues for XML Documents}, booktitle = {Encyclopedia of Information Science and Technology (V)}, year = {2005}, pages = {2616-2621}, publisher = {Idea Group}, organization = {Idea Group}, isbn = {1-59140-553-X}, author = {Pallis, George and Stoupa, Konstantina and Athena Vakali}, editor = {Khosrow-Pour, Mehdi} } @inproceedings {conf/mmdb/HammicheBHV04, title = {Semantic retrieval of multimedia data}, booktitle = {MMDB}, year = {2004}, pages = {36-44}, publisher = {ACM}, organization = {ACM}, abstract = {This paper deals with the problem of finding multimediadata that fulfill the requirements of user queries. We assumeboth the user query and the multimedia data are expressedby MPEG-7 standard. The MPEG-7 formalism lacks thesemantics and reasoning support in many ways. For example,the search of the implicit data can not be achieved,due to its description based on XML schema. We propose aframework for querying multimedia data based on a tree embeddingapproximation algorithm, combining the MPEG-7standard and an ontology
}, keywords = {Approximation Ontologies, MPEG-7, Multimedia Data, Tree embedding}, isbn = {1-58113-975-6}, author = {Hammiche, Samira and Benbernou, Salima and Hacid, Mohand-Said and Athena Vakali}, editor = {Chen, Shu-Ching and Shyu, Mei-Ling} } @inproceedings {1808, title = {Semantic Retrieval of Multimedia Data}, year = {2004}, abstract = {This paper deals with the problem of finding multimediadata that fulfill the requirements of user queries. We assumeboth the user query and the multimedia data are expressedby MPEG-7 standard. The MPEG-7 formalism lacks thesemantics and reasoning support in many ways. For example,the search of the implicit data can not be achieved,due to its description based on XML schema. We propose aframework for querying multimedia data based on a tree embeddingapproximation algorithm, combining the MPEG-7standard and an ontology
} } @article {journals/jss/TerziVA04, title = {A simulated annealing approach for multimedia data placement}, journal = {Journal of Systems and Software}, volume = {73}, number = {3}, year = {2004}, pages = {467-480}, author = {Terzi, Evimaria and Athena Vakali and Angelis, Lefteris} } @inproceedings {conf/appinf/PallisVAH03, title = {A Study on Workload Characterization for a Web Proxy Server}, booktitle = {Applied Informatics}, year = {2003}, pages = {779-784}, publisher = {IASTED/ACTA Press}, organization = {IASTED/ACTA Press}, abstract = {The popularity of the World-Wide-Web has increaseddramatically in the past few years. Web proxy servershave an important role in reducing server loads, networktraffic, and client request latencies. This paper presentsa detailed workload characterization study of a busyWeb proxy server. The study aims in identifying themajor characteristics which will improve modelling ofWeb proxy accessing. A set of log files is processed forworkload characterization. Throughout the study,emphasis is given on identifying the criteria for a Webcaching model. A statistical analysis, based on theprevious criteria, is presented in order to characterizethe major workload parameters. Results of this analysisare presented and the paper concludes with a discussionabout workload characterization and content deliveryissues.
}, keywords = {Web Caching, Web Data Workload Analysis, Web Technologies}, isbn = {0-88986-345-8}, author = {Pallis, George and Athena Vakali and Angelis, Lefteris and Hacid, Mohand-Said}, editor = {Hamza, M. H.} } @inproceedings {1877, title = {A Study on Workload Characterization for a Web Proxy Server}, year = {2003}, abstract = {The popularity of the World-Wide-Web has increaseddramatically in the past few years. Web proxy servershave an important role in reducing server loads, networktraffic, and client request latencies. This paper presentsa detailed workload characterization study of a busyWeb proxy server. The study aims in identifying themajor characteristics which will improve modelling ofWeb proxy accessing. A set of log files is processed forworkload characterization. Throughout the study,emphasis is given on identifying the criteria for a Webcaching model. A statistical analysis, based on theprevious criteria, is presented in order to characterizethe major workload parameters. Results of this analysisare presented and the paper concludes with a discussionabout workload characterization and content deliveryissues.
} } @inproceedings {1886, title = {Security Model for XML data}, year = {2001}, abstract = {The significance of XML technology for sharing data over the Internet is being rapidly recognised. In this paper, we examine the security problems related to XML data and present our approach, the XML Security model, for enforcing security policies in XML based Information systems. Our methodology has been based on the study of the XML data model, on the identification of the security requirements of XML Information systems and on the survey of security models which have been proposed to support the conventional data models(relational, object-oriented, hypertext etc). The proposed approach takes into account and exploits the specific characteristics of XML data and incorporates the flexibility of Role based Access Control policies.} } @inproceedings {conf/ic/IlioudisPV01, title = {Security Model for XML Data}, booktitle = {International Conference on Internet Computing (1)}, year = {2001}, pages = {400-406}, abstract = {The significance of XML technology for sharing data over the Internet is being rapidly recognised. In this paper, we examine the security problems related to XML data and present our approach, the XML Security model, for enforcing security policies in XML based Information systems. Our methodology has been based on the study of the XML data model, on the identification of the security requirements of XML Information systems and on the survey of security models which have been proposed to support the conventional data models(relational, object-oriented, hypertext etc). The proposed approach takes into account and exploits the specific characteristics of XML data and incorporates the flexibility of Role based Access Control policies.
}, keywords = {Role Based Access Control, XML Security}, author = {Ilioudis, Christos and Pangalos, George and Athena Vakali} } @inproceedings {1853, title = {A Study on Web Caching Architectures and Performance}, year = {2001}, abstract = {As World Wide Web usage has grown dramaticallyin recent yearsso has grown the recognition thatWeb caches especially proxy caches will have animportant role in reducing server loadsclient request latenciesand network trac In this survey we present the most common architectures for webcaching and their most important characteristics areoutlined These architectures include proxy cachingcooperative cachingadaptive cachingpush cachingand active caching Furthermoreemphasis is givenon the basic metrics and factors for evaluating proxycache performance.
} } @inproceedings {1885, title = {A Simulation Study of Shadowed Disks}, year = {1999}, abstract = {Multiple disks have been introduced for use in I/O subsystems to increase data availability and fault tolerance. Identical disks (comprising the shadowed disks) sets have been adopted as an effective secondary storage replication scheme. In the present paper, we introduce a shadowed disk model and mathematically analyze its theoritical expected performance measures by applying the nearer-server rule to satisfy the requests. In addition, we introduce the corresponding simulation model, where each disk is modeled by means of a protoype disk configuration and specific commercial disks metrics. Furthermore, the mirrored disks model is studied extensively as a boundary disk topology of the shadowed disks set. The developed simulation model is experimented and tested under various workloads produced by real I/O traces. The performance of the simulation model is compared to the analytical results and, thus, it is validated since analysis and simulation give close results.
} } @article {journals/ipl/ManolopoulosV91, title = {Seek Distances in Disks with Two Independent Heads Per Surface}, journal = {Inf. Process. Lett.}, volume = {37}, number = {1}, year = {1991}, pages = {37-42}, author = {Manolopoulos, Yannis and Athena Vakali} }