Publications

@inproceedings{DBLP:conf/icde/ColomboF17,

title = {Towards a Unifying Attribute Based Access Control Approach for NoSQL 

 Datastores},

author = {Pietro Colombo and Elena Ferrari},

url = {https://doi.org/10.1109/ICDE.2017.123},

doi = {10.1109/ICDE.2017.123},

year  = {2017},

date = {2017-01-01},

booktitle = {33rd IEEE International Conference on Data Engineering, ICDE 2017, 

 San Diego, CA, USA, April 19-22, 2017},

pages = {709--720},

publisher = {IEEE Computer Society},

abstract = {NoSQL datastores allow the efficient management of high volumes of heterogeneous and unstructured data, meeting the requirements of a variety of today ICT applications. However, most of these systems poorly support data security, and recent surveys show that their simplistic support for data protection is considered as a reason not to use them.1 In recent years, Attribute Based Access Control (ABAC) is getting more and more popularity, for its ability to provide highly flexible and customized forms of data protection at different granularity levels. In the current work, with the aim to raise users' confidence in the protection of data managed by NoSQL systems, we define a general approach to enforce ABAC within NoSQL systems. Our approach relies on SQL++[20], a unifying query language for NoSQL platforms. In particular, we develop a novel SQL++ query rewriting mechanism able to enforce heterogeneous types of ABAC policies specified up to cell level. Experimental results show an overhead which is not negligible for policies covering high percentage of the fields characterizing the protected documents, but which is far more contained when field level policies are more sparsely specified.},

keywords = {},

pubstate = {published},

tppubtype = {inproceedings}

}

@inproceedings{DBLP:conf/primelife/Ferrari17,

title = {Identification Services for Online Social Networks (OSNs) Extended 

 Abstract},

author = {Elena Ferrari},

editor = {Marit Hansen and Eleni Kosta and Igor Nai Fovino and Simone Fischer-H \"{u}},

url = {https://doi.org/10.1007/978-3-319-92925-5_16},

doi = {10.1007/978-3-319-92925-5_16},

year  = {2017},

date = {2017-01-01},

booktitle = {Privacy and Identity Management. The Smart Revolution - 12th IFIP 

 WG 9.2, 9.5, 9.6/11.7, 11.6/SIG 9.2.2 International Summer School, 

 Ispra, Italy, September 4-8, 2017, Revised Selected Papers},

volume = {526},

pages = {240--242},

publisher = {Springer},

series = {IFIP Advances in Information and Communication Technology},

abstract = {On-line Social Networks (OSNs) have dramatically changed how users connect, communicate, share content, and exchange goods and services. However, despite all the benefits and the flexibility that OSNs provide, their users become more reliant on online identities with often no means to know who really is behind an online profile. Indeed, to facilitate their adoption and encourage people to join, identities in OSNs are very loose, in that not much more than an email address is required to create an account and related profile. Therefore, the problem of fake accounts and identity related attacks in OSNs has attracted considerable interest from the research community, and resulted in several proposals that mainly aim at detecting malicious nodes that follow identified and formalized attack trends. Without denying the importance of formalizing Sybil attacks and suggesting solutions for their detection, in this extended abstract we also consider the issue of identity validation from a user perspective, by briefly discussing the research proposals aiming at empowering users with tools helping them to identify the validity of the online accounts they interact with.On-line Social Networks (OSNs) have dramatically changed how users connect, communicate, share content, and exchange goods and services. However, despite all the benefits and the flexibility that OSNs provide, their users become more reliant on online identities with often no means to know who really is behind an online profile. Indeed, to facilitate their adoption and encourage people to join, identities in OSNs are very loose, in that not much more than an email address is required to create an account and related profile. Therefore, the problem of fake accounts and identity related attacks in OSNs has attracted considerable interest from the research community, and resulted in several proposals that mainly aim at detecting malicious nodes that follow identified and formalized attack trends. Without denying the importance of formalizing Sybil attacks and suggesting solutions for their detection, in this extended abstract we also consider the issue of identity validation from a user perspective, by briefly discussing the research proposals aiming at empowering users with tools helping them to identify the validity of the online accounts they interact with.},

keywords = {},

pubstate = {published},

tppubtype = {inproceedings}

}

@inproceedings{DBLP:conf/IEEEscc/CarminatiCFS16,

title = {Enhancing User Control on Personal Data Usage in Internet of Things 

 Ecosystems},

author = {Barbara Carminati and Pietro Colombo and Elena Ferrari and Gokhan Sagirlar},

editor = {Jia Zhang and John A Miller and Xiaofei Xu},

url = {https://doi.org/10.1109/SCC.2016.45},

doi = {10.1109/SCC.2016.45},

year  = {2016},

date = {2016-01-01},

booktitle = {IEEE International Conference on Services Computing, SCC 2016, 

 San Francisco, CA, USA, June 27 - July 2, 2016},

pages = {291--298},

publisher = {IEEE Computer Society},

abstract = {Internet of Things (IoT) services are improving our life, supporting people in a variety of situations. However, due to the high volume of managed personal data, they can be a serious threat for individuals privacy. Users data are commonly gathered by devices scattered in the IoT, each of which sees a portion of them. The combination of different data may lead to infer users sensitive information. The distributed nature and the complexity of the IoT scenario cause users to lose the control on how their data are handled. In this paper, we start addressing this issue with a framework that empowers users to better control data management within IoT ecosystems. A novel privacy reference model allows users to state how their data can be processed and what cannot be inferred from them, and a dedicated mechanism allows enforcing the stated references. Experimental results show the efficiency of the enforcement.},

keywords = {},

pubstate = {published},

tppubtype = {inproceedings}

}

@inproceedings{DBLP:conf/asunam/JaradatDMF16,

title = {Trust and privacy correlations in social networks: A deep learning 

 framework},

author = {Shatha Jaradat and Nima Dokoohaki and Mihhail Matskin and Elena Ferrari},

editor = {Ravi Kumar and James Caverlee and Hanghang Tong},

url = {https://doi.org/10.1109/ASONAM.2016.7752236},

doi = {10.1109/ASONAM.2016.7752236},

year  = {2016},

date = {2016-01-01},

booktitle = {2016 IEEE/ACM International Conference on Advances in Social Networks 

 Analysis and Mining, ASONAM 2016, San Francisco, CA, USA, August 

 18-21, 2016},

pages = {203--206},

publisher = {IEEE Computer Society},

abstract = {Online Social Networks (OSNs) remain the focal point of Internet usage. Since the beginning, networking sites tried best to have right privacy mechanisms in place for users, enabling them to share the right content with the right audience. With all these efforts, privacy customizations remain hard for users across the sites. Existing research that address this problem mainly focus on semi-supervised strategies that introduce extra complexity by requiring the user to manually specify initial privacy preferences for their friends. In this work, we suggest an adaptive solution that can dynamically generate privacy labels for users in OSNs. To this end, we introduce a deep reinforcement learning framework that targets two key problems in OSNs like Facebook: the exposure of users' interactions through the network to less trusted direct friends, and the possibility of propagating user updates through direct friends' interactions to indirect friends. By implementing this framework, we aim at understanding how social trust and privacy could be correlated, specifically in a dynamic fashion. We report the ranked dependence between the generated privacy labels and the estimated user trust values, which indicate the ability of the framework to identify the highly trusted users and share with them higher percentages of data.},

keywords = {},

pubstate = {published},

tppubtype = {inproceedings}

}

@inproceedings{DBLP:conf/coinco/AlbertiniCF16,

title = {Privacy Settings Recommender for Online Social Network},

author = {Davide Alberto Albertini and Barbara Carminati and Elena Ferrari},

url = {https://doi.org/10.1109/CIC.2016.079},

doi = {10.1109/CIC.2016.079},

year  = {2016},

date = {2016-01-01},

booktitle = {2nd IEEE International Conference on Collaboration and Internet 

 Computing, CIC 2016, Pittsburgh, PA, USA, November 1-3, 2016},

pages = {514--521},

publisher = {IEEE Computer Society},

abstract = {In recent years Relationship Based Access Control (ReBAC) has become the reference paradigm for controlled information sharing in Online Social Network (OSN) scenarios. Nevertheless, many of the most popular OSN providers do not implement in their platforms an access control model fully compliant with ReBAC. This fact, thus, limits the capability of OSN users to define customized and fine-grained access control policies. Moreover, average users might have difficulties in properly setting, potentially, complex access control policies. As results, many users give up in defining proper privacy setting, simply accepting the default setting proposed by OSN provider. To cope with this problem, we see the need of tools in support of policy specification. At this aim, in this paper we presenta recommendation system that, exploiting an association rules mining process, learns OSN users' habits in releasing resources in online social networks, and exploit them to suggest customized access control policies. We also prove the feasibility of the presented techniques by illustrating an experiment which has been conducted on 30 human users by building customized access control policies from the data learnt from each of them.},

keywords = {},

pubstate = {published},

tppubtype = {inproceedings}

}

@inproceedings{DBLP:conf/icde/ColomboF16,

title = {Towards Virtual Private NoSQL datastores},

author = {Pietro Colombo and Elena Ferrari},

url = {https://doi.org/10.1109/ICDE.2016.7498240},

doi = {10.1109/ICDE.2016.7498240},

year  = {2016},

date = {2016-01-01},

booktitle = {32nd IEEE International Conference on Data Engineering, ICDE 2016, 

 Helsinki, Finland, May 16-20, 2016},

pages = {193--204},

publisher = {IEEE Computer Society},

abstract = {Many modern applications use context related information to provide highly personalized services, and use NoSQL databases for data management, as these systems show outstanding performance and support high volumes of data. However, NoSQL databases integrate poor data protection features with basic coarse grained access control and no support for context aware policies. Therefore, we believe that a general approach is required to enhance NoSQL datastores with fine grained context aware access control. In this paper, we start to fill this void by targeting MongoDB, a very popular datastore. The contribution is twofold. We enhance MongoDB's access control model with advanced features and we define an enforcement monitor for the proposed enhanced model, which can be straightforwardly used in any MongoDB deployment. Technological limitations of MongoDB do not allow implementing the same efficient enforcement mechanism for all query types. As a consequence, experimental results show an enforcement overhead that is significant for aggregate queries, which contrasts with a low overhead measured for find and map-reduce queries.},

keywords = {},

pubstate = {published},

tppubtype = {inproceedings}

}

@inproceedings{DBLP:conf/icde/BahriSSCFG16,

title = {Beat the DIVa - decentralized identity validation for online social 

 networks},

author = {Leila Bahri and Amira Soliman and Jacopo Squillaci and Barbara Carminati and Elena Ferrari and Sarunas Girdzijauskas},

url = {https://doi.org/10.1109/ICDE.2016.7498337},

doi = {10.1109/ICDE.2016.7498337},

year  = {2016},

date = {2016-01-01},

booktitle = {32nd IEEE International Conference on Data Engineering, ICDE 2016, 

 Helsinki, Finland, May 16-20, 2016},

pages = {1330--1333},

publisher = {IEEE Computer Society},

abstract = {Fake accounts in online social networks (OSNs) have known considerable sophistication and are now attempting to gain network trust by infiltrating within honest communities. Honest users have limited perspective on the truthfulness of new online identities requesting their friendship. This facilitates the task of fake accounts in deceiving honest users to befriend them. To address this, we have proposed a model that learns hidden correlations between profile attributes within OSN communities, and exploits them to assist users in estimating the trustworthiness of new profiles. To demonstrate our method, we suggest, in this demo, a game application through which players try to cheat the system and convince nodes in a simulated OSN to befriend them. The game deploys different strategies to challenge the players and to reach the objectives of the demo. These objectives are to make participants aware of how fake accounts can infiltrate within their OSN communities, to demonstrate how our suggested method could aid in mitigating this threat, and to eventually strengthen our model based on the data collected from the moves of the players.},

keywords = {},

pubstate = {published},

tppubtype = {inproceedings}

}

@inproceedings{DBLP:conf/icde/ColomboF16a,

title = {Efficient enforcement of action-aware purpose-based access control 

 within relational database management systems},

author = {Pietro Colombo and Elena Ferrari},

url = {https://doi.org/10.1109/ICDE.2016.7498402},

doi = {10.1109/ICDE.2016.7498402},

year  = {2016},

date = {2016-01-01},

booktitle = {32nd IEEE International Conference on Data Engineering, ICDE 2016, 

 Helsinki, Finland, May 16-20, 2016},

pages = {1516--1517},

publisher = {IEEE Computer Society},

abstract = {Among the variety of access control models proposed for database management systems (DBMSs) a key role is covered by the purpose-based access control model, which, while enforcing access control, also achieves basic privacy preservation. We believe that DBMSs could greatly take benefit from the integration of an enhanced purpose based model supporting highly customized and efficient access control. Therefore, in this paper, we propose a purpose-based model that supports action-aware policy specification and a related efficient enforcement framework to be integrated into relational DBMSs. The experimental evaluation we have performed shows the feasibility and efficiency of the proposed framework.},

keywords = {},

pubstate = {published},

tppubtype = {inproceedings}

}

@inproceedings{DBLP:conf/ipps/ThuraisinghamKK16,

title = {Emergency-Driven Assured Information Sharing in Secure Online Social 

 Networks: A Position Paper},

author = {Bhavani M Thuraisingham and Murat Kantarcioglu and Latifur Khan and Barbara Carminati and Elena Ferrari and Leila Bahri},

url = {https://doi.org/10.1109/IPDPSW.2016.201},

doi = {10.1109/IPDPSW.2016.201},

year  = {2016},

date = {2016-01-01},

booktitle = {2016 IEEE International Parallel and Distributed Processing Symposium 

 Workshops, IPDPS Workshops 2016, Chicago, IL, USA, May 23-27, 2016},

pages = {1813--1820},

publisher = {IEEE Computer Society},

abstract = {The United States and its Allied Forces have had tremendous success in combat operations. This includes combat in Germany, Japan and more recently in Iraq and Afghanistan. However not all of our stabilization and reconstruction operations (SARO) have been as successful. Several studies have been carried out on SARO by organizations such as the National Defense University and the Naval Post Graduate School. These studies have shown that security as well as power and jobs are key ingredients for success during SARO. One of the major conclusions is that we need to plan for SARO while we are planning for combat. These studies have also analyzed the various technologies that are needed for successfully carrying out SARO which include sensors, robotics and information management. As stated in the work by the Naval Postgraduate School, we need to determine the social, political and economic relationships between the local communities as well as determine who the important people are. One particular task during SARO is managing emergencies and crisis situations as well as carrying out humanitarian operations. In such situations, it is critical that the information between various individuals as well as agencies be shared. Furthermore, social relationships between the different parties need to be analyzed. To address the key technical challenges for emergency management and crisis response during SARO and related humanitarian efforts we are exploring assured information sharing and secure social networking to develop novel technologies to support emergency management and crisis response.},

keywords = {},

pubstate = {published},

tppubtype = {inproceedings}

}

@inproceedings{DBLP:conf/iri/SinghCF16,

title = {A Risk-Benefit Driven Architecture for Personal Data Release (Invited 

 Paper)},

author = {Bikash Chandra Singh and Barbara Carminati and Elena Ferrari},

url = {https://doi.org/10.1109/IRI.2016.14},

doi = {10.1109/IRI.2016.14},

year  = {2016},

date = {2016-01-01},

booktitle = {17th IEEE International Conference on Information Reuse and Integration, 

 IRI 2016, Pittsburgh, PA, USA, July 28-30, 2016},

pages = {40--49},

publisher = {IEEE Computer Society},

abstract = {Personal data storages (PDSs) give individuals the ability to store their personal data in a data unified repository and control release of their data to data consumers. Being able to gather personal data from different data sources (e.g., banks, hospitals), PDSs will play strategic role in individual privacy management. As such, PDS demands for new privacy models for protecting personal data. In this paper, we propose a new technical approach that empowers individuals to better control data in PDS. Particularly, we present a privacy-aware PDS architecture by focusing on two logical data zones based on the categories of personal data. Moreover, we propose a strategy for regulating personal data release that takes in consideration both user preferences and possible risks and benefits of the data release.},

keywords = {},

pubstate = {published},

tppubtype = {inproceedings}

}

@inproceedings{DBLP:conf/medhocnet/LalehCF16,

title = {Anomalous change detection in time-evolving OSNs},

author = {Naeimeh Laleh and Barbara Carminati and Elena Ferrari},

url = {https://doi.org/10.1109/MedHocNet.2016.7528431},

doi = {10.1109/MedHocNet.2016.7528431},

year  = {2016},

date = {2016-01-01},

booktitle = {2016 Mediterranean Ad Hoc Networking Workshop, Med-Hoc-Net 2016, Vilanova 

 i la Geltru, Spain, June 20-22, 2016},

pages = {1--8},

publisher = {IEEE},

abstract = {Recently, the automatic highlighting of anomalous changes in a sequence of social graph snapshots is receiving growing interest due to its numerous applications. For instance, it may be helpful for the identification of attackers or risky users in Online Social Networks (OSNs). Indeed, dynamically monitoring and learning the friendship patterns of users in a large social graph over time for any anomalous change often reflects and predicts significant events or attacker's behaviors. In this paper, we focus on anomalous changes that happen in the neighborhood of OSN users. Our main goal is to detect those users whose changes in the structure of their subgraph deviate from their own previous change patterns and from those of other nearest users in the graph. Our approach returns a list of these users by ranking them based on the value of their change deviation. We analyze the performance of our approach on a real Google+ dataset.},

keywords = {},

pubstate = {published},

tppubtype = {inproceedings}

}

@inproceedings{DBLP:conf/mldm/LalehCFG16,

title = {Gossip-Based Behavioral Group Identification in Decentralized OSNs},

author = {Naeimeh Laleh and Barbara Carminati and Elena Ferrari and Sarunas Girdzijauskas},

editor = {Petra Perner},

url = {https://doi.org/10.1007/978-3-319-41920-6_52},

doi = {10.1007/978-3-319-41920-6_52},

year  = {2016},

date = {2016-01-01},

booktitle = {Machine Learning and Data Mining in Pattern Recognition - 12th International 

 Conference, MLDM 2016, New York, NY, USA, July 16-21, 2016, Proceedings},

volume = {9729},

pages = {676--691},

publisher = {Springer},

series = {Lecture Notes in Computer Science},

abstract = {DOSNs are distributed systems providing social networking services that become extremely popular in recent years. In DOSNs, the aim is to give the users control over their data and keeping data locally to enhance privacy. Therefore, identifying behavioral groups of users that share the same behavioral patterns in decentralized OSNs is challenging. In the fully distributed social graph, each user has only one feature vector and these vectors can not move to any central storage or other users in a raw form duo to privacy issues. We use a gossip learning approach where all users are involved with their local estimation of the clustering model and improve their estimations and finally converge to a final clustering model available for all users. In order to evaluate our approach, we implement our algorithm and test it in a real Facebook graph.},

keywords = {},

pubstate = {published},

tppubtype = {inproceedings}

}

@inproceedings{DBLP:conf/webi/BartoliCFM16,

title = {A Language and an Inference Engine for Twitter Filtering Rules},

author = {Alberto Bartoli and Barbara Carminati and Elena Ferrari and Eric Medvet},

url = {https://doi.org/10.1109/WI.2016.0107},

doi = {10.1109/WI.2016.0107},

year  = {2016},

date = {2016-01-01},

booktitle = {2016 IEEE/WIC/ACM International Conference on Web Intelligence, 

 WI 2016, Omaha, NE, USA, October 13-16, 2016},

pages = {614--617},

publisher = {IEEE Computer Society},

abstract = {We consider the problem of the filtering of Twitter posts, that is, the hiding of those posts which the user prefers not to visualize on his/her timeline. We define a language for specifying filtering policies suitable for Twitter posts. The language allows each user to decide which posts to filter out based on his/her sensibility and preferences. Since average users may not have the skills necessary to translate their filtering needs into a set of rules, we also propose a method for inferring a policy automatically, based solely on examples of the desired filtering behavior. The method is based on an evolutionary approach driven by a multi-objective optimization scheme. We assess our proposal experimentally on a real Twitter dataset and the results are highly promising.},

keywords = {},

pubstate = {published},

tppubtype = {inproceedings}

}

@inproceedings{DBLP:conf/wistp/BahriCFL16,

title = {LAMP - Label-Based Access-Control for More Privacy in Online Social 

 Networks},

author = {Leila Bahri and Barbara Carminati and Elena Ferrari and William Lucia},

editor = {Sara Foresti and Javier L \'{o}},

url = {https://doi.org/10.1007/978-3-319-45931-8_11},

doi = {10.1007/978-3-319-45931-8_11},

year  = {2016},

date = {2016-01-01},

booktitle = {Information Security Theory and Practice - 10th IFIP WG 11.2 International 

 Conference, WISTP 2016, Heraklion, Crete, Greece, September 26-27, 

 2016, Proceedings},

volume = {9895},

pages = {171--186},

publisher = {Springer},

series = {Lecture Notes in Computer Science},

abstract = {Access control in Online Social Networks (OSNs) is generally approached with a relationship-based model. This limits the options in expressing privacy preferences to only the types of relationships users establish in the OSN. Moreover, current proposals do not address the privacy of dependent information types, such as comments or likes, at their atomic levels of ownership. Rather, the privacy of these data elements is holistically dependent on the aggregate object they belong to. To overcome this, we propose LAMP, a model that deploys fine grained label-based access control for information sharing in OSNs. Users in LAMP assign customized labels to their friends and to all types of their information; whereas access requests are evaluated by security properties carefully designed to establish orders between requestor’s and information’s labels. We prove the correctness of the suggested model, and we perform performance experiments based on different access scenarios simulated on a real OSN graph. We also performed a preliminary usability study that compared LAMP to Facebook privacy settings.},

keywords = {},

pubstate = {published},

tppubtype = {inproceedings}

}

@inproceedings{DBLP:conf/asunam/SolimanBCFG15,

title = {DIVa: Decentralized Identity Validation for Social Networks},

author = {Amira Soliman and Leila Bahri and Barbara Carminati and Elena Ferrari and Sarunas Girdzijauskas},

editor = {Jian Pei and Fabrizio Silvestri and Jie Tang},

url = {https://doi.org/10.1145/2808797.2808861},

doi = {10.1145/2808797.2808861},

year  = {2015},

date = {2015-01-01},

booktitle = {Proceedings of the 2015 IEEE/ACM International Conference on Advances 

 in Social Networks Analysis and Mining, ASONAM 2015, Paris, France, 

 August 25 - 28, 2015},

pages = {383--391},

publisher = {ACM},

abstract = {Online Social Networks exploit a lightweight process to identify their users so as to facilitate their fast adoption. However, such convenience comes at the price of making legitimate users subject to different threats created by fake accounts. Therefore, there is a crucial need to empower users with tools helping them in assigning a level of trust to whomever they interact with. To cope with this issue, in this paper we introduce a novel model, DIVa, that leverages on mining techniques to find correlations among user profile attributes. These correlations are discovered not from user population as a whole, but from individual communities, where the correlations are more pronounced. DIVa exploits a decentralized learning approach and ensures privacy preservation as each node in the OSN independently processes its local data and is required to know only its direct neighbors. Extensive experiments using real-world OSN datasets show that DIVa is able to extract fine-grained community-aware correlations among profile attributes with average improvements up to 50% than the global approach.},

keywords = {},

pubstate = {published},

tppubtype = {inproceedings}

}

@inproceedings{DBLP:conf/ccs/UlusoyCFKP15,

title = {GuardMR: Fine-grained Security Policy Enforcement for MapReduce Systems},

author = {Huseyin Ulusoy and Pietro Colombo and Elena Ferrari and Murat Kantarcioglu and Erman Pattuk},

editor = {Feng Bao and Steven Miller and Jianying Zhou and Gail -},

url = {https://doi.org/10.1145/2714576.2714624},

doi = {10.1145/2714576.2714624},

year  = {2015},

date = {2015-01-01},

booktitle = {Proceedings of the 10th ACM Symposium on Information, Computer and 

 Communications Security, ASIA CCS '15, Singapore, April 14-17, 

 2015},

pages = {285--296},

publisher = {ACM},

abstract = {Executing data analytics tasks in MapReduce systems introduces new security and privacy concerns as the processed unstructured datasets may contain sensitive information (e.g., social security numbers, business sensitive information) at the level of individual records, and the existing file-level access control mechanisms provide all or nothing access to the entire dataset. To address these concerns, we propose GUARDMR which is a novel, modular framework that can enforce fine-grained security policies at the key-value level in MapReduce systems. The presented security policies can dynamically create authorized views of data resources based on the organizational roles of the MapReduce users. GUARDMR further simplifies the specification of authorized views via automatically generating the bytecode of the functions necessary for creating the views, from the high level specification language (i.e., OCL). It facilitates enforcement of a broad, flexible set of policies that can handle the complexity demanded by high volume, high variety, unstructured datasets and general MapReduce computation without any modification to the underlying MapReduce system and operating system. Our evaluation results indicate that GUARDMR provides fine-grained access control for Apache Hadoop system with easy maintainability and very low overhead.Executing data analytics tasks in MapReduce systems introduces new security and privacy concerns as the processed unstructured datasets may contain sensitive information (e.g., social security numbers, business sensitive information) at the level of individual records, and the existing file-level access control mechanisms provide all or nothing access to the entire dataset. To address these concerns, we propose GUARDMR which is a novel, modular framework that can enforce fine-grained security policies at the key-value level in MapReduce systems. The presented security policies can dynamically create authorized views of data resources based on the organizational roles of the MapReduce users. GUARDMR further simplifies the specification of authorized views via automatically generating the bytecode of the functions necessary for creating the views, from the high level specification language (i.e., OCL). It facilitates enforcement of a broad, flexible set of policies that can handle the complexity demanded by high volume, high variety, unstructured datasets and general MapReduce computation without any modification to the underlying MapReduce system and operating system. Our evaluation results indicate that GUARDMR provides fine-grained access control for Apache Hadoop system with easy maintainability and very low overhead.},

keywords = {},

pubstate = {published},

tppubtype = {inproceedings}

}

@inproceedings{DBLP:conf/coinco/BahriCF15,

title = {CARDS - Collaborative Audit and Report Data Sharing for A-Posteriori 

 Access Control in DOSNs},

author = {Leila Bahri and Barbara Carminati and Elena Ferrari},

url = {https://doi.org/10.1109/CIC.2015.18},

doi = {10.1109/CIC.2015.18},

year  = {2015},

date = {2015-01-01},

booktitle = {IEEE Conference on Collaboration and Internet Computing, CIC 2015, 

 Hangzhou, China, October 27-30, 2015},

pages = {36--45},

publisher = {IEEE Computer Society},

abstract = {Accountability and transparency have been commonly accepted to deter bad acts and to encourage compliance to rules. For this, auditing has been largely, and since ancient times, adopted to ensure the well running of systems and businesses within which duties are governed by set rules. Recently, an a-posteriori approach to data access control has been investigated for information systems as well across number of critical domains (e.g., Healthcare systems). Besides, privacy advocates started calling for the necessity of accountability and transparency in managing users' privacy in nowadays connected and proliferated web data. Following this line of thought, we suggest a system for collaborative a-posteriori access control to data dissemination in decentralized online social networks based on reporting and auditing. We demonstrate the usability of our suggested model using a real OSN graph.},

keywords = {},

pubstate = {published},

tppubtype = {inproceedings}

}

@inproceedings{DBLP:conf/emo/MedvetBCF15,

title = {Evolutionary Inference of Attribute-Based Access Control Policies},

author = {Eric Medvet and Alberto Bartoli and Barbara Carminati and Elena Ferrari},

editor = {Ant\'{o}nio Gaspar - and Carlos Henggeler Antunes and Carlos Coello A Coello},

url = {https://doi.org/10.1007/978-3-319-15934-8_24},

doi = {10.1007/978-3-319-15934-8_24},

year  = {2015},

date = {2015-01-01},

booktitle = {Evolutionary Multi-Criterion Optimization - 8th International Conference, 

 EMO 2015, Guimar\~{a}es, Portugal, March 29 -April 1, 2015. Proceedings, 

 Part I},

volume = {9018},

pages = {351--365},

publisher = {Springer},

series = {Lecture Notes in Computer Science},

abstract = {The interest in attribute-based access control policies is increasingly growing due to their ability to accommodate the complex security requirements of modern computer systems. With this novel paradigm, access control policies consist of attribute expressions which implicitly describe the properties of subjects and protection objects and which must be satisfied for a request to be allowed. Since specifying a policy in this framework may be very complex, approaches for policy mining, i.e., for inferring a specification automatically from examples in the form of logs of authorized and denied requests, have been recently proposed. 

 In this work, we propose a multi-objective evolutionary approach for solving the policy mining task. We designed and implemented a problem representation suitable for evolutionary computation, along with several search-optimizing features which have proven to be highly useful in this context: a strategy for learning a policy by learning single rules, each one focused on a subset of requests; a custom initialization of the population; a scheme for diversity promotion and for early termination. We show that our approach deals successfully with case studies of realistic complexity.},

keywords = {},

pubstate = {published},

tppubtype = {inproceedings}

}

@inproceedings{DBLP:conf/icws/CarminatiFT15,

title = {A Privacy-Preserving Framework for Constrained Choreographed Service 

 Composition},

author = {Barbara Carminati and Elena Ferrari and Ngoc Hong Tran},

editor = {John A Miller and Hong Zhu},

url = {https://doi.org/10.1109/ICWS.2015.48},

doi = {10.1109/ICWS.2015.48},

year  = {2015},

date = {2015-01-01},

booktitle = {2015 IEEE International Conference on Web Services, ICWS 2015, 

 New York, NY, USA, June 27 - July 2, 2015},

pages = {297--304},

publisher = {IEEE Computer Society},

abstract = {One of the major goals of Web services is to make easier their composition to form more complex services, modeled as workflows. A key role in the Web services composition is the selection of a proper service for each activity in the workflow. In general, this requires the exchange of sensitive information of users, requiring the composition, as well as of involved service providers. So far this problem has been investigated in the setting of orchestrated service composition, under the assumption of the presence of a broker coordinating the composition. However, a promising alternative approach is the one of choreography, where each service involved in the service composition has to locally manage service selection and invocation. In this paper, we propose a framework to enforce user and provider requirements in the scenario of service choreography in a privacy-preserving way, that is, without the releasing of any private information of users and providers. To achieve this result we make use of different privacy-preserving protocols. As it will be shown in the paper, the proposed solution does not implies relevant overhead.One of the major goals of Web services is to make easier their composition to form more complex services, modeled as workflows. A key role in the Web services composition is the selection of a proper service for each activity in the workflow. In general, this requires the exchange of sensitive information of users, requiring the composition, as well as of involved service providers. So far this problem has been investigated in the setting of orchestrated service composition, under the assumption of the presence of a broker coordinating the composition. However, a promising alternative approach is the one of choreography, where each service involved in the service composition has to locally manage service selection and invocation. In this paper, we propose a framework to enforce user and provider requirements in the scenario of service choreography in a privacy-preserving way, that is, without the releasing of any private information of users and providers. To achieve this result we make use of different privacy-preserving protocols. As it will be shown in the paper, the proposed solution does not implies relevant overhead.},

keywords = {},

pubstate = {published},

tppubtype = {inproceedings}

}

@inproceedings{DBLP:conf/iri/BahriCF15,

title = {What Happens to My Online Social Estate When I Am Gone? An Integrated 

 Approach to Posthumous Online Data Management},

author = {Leila Bahri and Barbara Carminati and Elena Ferrari},

url = {https://doi.org/10.1109/IRI.2015.16},

doi = {10.1109/IRI.2015.16},

year  = {2015},

date = {2015-01-01},

booktitle = {2015 IEEE International Conference on Information Reuse and Integration, 

 IRI 2015, San Francisco, CA, USA, August 13-15, 2015},

pages = {31--38},

publisher = {IEEE Computer Society},

abstract = {Technology and the digital world have been making an important part of people's lives nowadays. As death is unquestionably a crucial and fundamental part of life, technology and the digital world ought to play an equally important role in end of life issues as well. For instance, the adoption of online social networks (OSNs) has been amplifying to cover large numbers of the world's population playing big roles in shaping their daily life, in documenting their life experiences, and in sharing their moments with their friends in the network. While current systems focus on the provision of usable and attractive features of their OSN services, considerations of the faith of the online accounts, identities, and data created and shared in their realms when the owner is mo more available to manage them have not been equally taken. In this paper, we raise and discuss issues related to the design and to the provision of integrated services for a posthumous data management that would respect the wills of users all while being concealed to their survivors. We survey the existing practices, we discuss their limitations, and we suggest an integrated approach to posthumous data management based on posthumous data planning assisted by data categorization and automated tools.},

keywords = {},

pubstate = {published},

tppubtype = {inproceedings}

}

@inproceedings{DBLP:conf/mobihoc/BahriCFT15,

title = {Privacy Preserving Decentralized Identity Validation for Geo-social 

 Networks over MANET},

author = {Leila Bahri and Barbara Carminati and Elena Ferrari and Ngoc Hong Tran},

editor = {Michael Sirivianos and Pan Hui and Nishanth Sastry and Yang Chen and Hongqiang Harry Liu},

url = {https://doi.org/10.1145/2757513.2757520},

doi = {10.1145/2757513.2757520},

year  = {2015},

date = {2015-01-01},

booktitle = {Proceedings of the 7th International Workshop on Hot Topics in Planet-scale mObile computing and online Social neTworking, HOTPOST@MobiHoc 2015, 

 Hangzhou, China, June 22, 2015},

pages = {7--12},

publisher = {ACM},

abstract = {Mobile phones, and more specifically smart gadgets, have known a rapid proliferation over the past years in terms of their adoption and usage. Their prices have also known noticeable declines making the ownership of a smart-phone at the ability of all pocket sizes. This has created tremendous potential for the design and creation of services that users can consume through their smart-phones and that would improve their daily lives tasks. In this work, we focus on the potential of using smart-phones in geographically bounded areas, such as shopping malls, museums, conference venues, etc, to establish collaborative ad-hoc networks over MANET. These networks are meant to allow for the provision of P2P exchange of information and help between visitors of such places to improve their visiting experience. We discuss how such a network could be designed and we focus on two main challenges: 1. identity validation over the network to ensure the worthiness of provided information, and 2. privacy preservation both against personal information inference from provided information and over the p2p overlay.},

keywords = {},

pubstate = {published},

tppubtype = {inproceedings}

}

@inproceedings{DBLP:conf/sebd/ColomboF15,

title = {Complementing MongoDB with Advanced Access Control Features: Concepts 

 and Research Challenges},

author = {Pietro Colombo and Elena Ferrari},

editor = {Domenico Lembo and Riccardo Torlone and Andrea Marrella},

year  = {2015},

date = {2015-01-01},

booktitle = {23rd Italian Symposium on Advanced Database Systems, SEBD 2015, 

 Gaeta, Italy, June 14-17, 2015},

pages = {343--350},

publisher = {Curran Associates, Inc.},

abstract = {Based on recent surveys the interest for NoSQL datastores is continuously growing. This trend is an expected consequence of key features of these systems, such as their ability to efficiently handle high data volumes, and to easily scale. Indeed, NoSQL datastores outclass 

traditional relational database management systems (RDBMs) wrt performance and scalability. In addition, with innovative computational paradigms, such as MapReduce and the aggregation pipeline, NoSQL datastores also overcome RDBMSs in terms of achievable forms of data analysis. At the same time, they provide only poor support for data protection, which is typically limited to a very basic form of access control. 

This aspect is seen by several managers as the top obstacle to the use of datastores in industry [14]. We believe that this issue can be addressed by integrating proper data 

protection mechanisms into NoSQL datastores. However, due to the variety of existing NoSQL datastores, each characterized by a given data model and query language, this appears as a very ambitious task. In order to do a first step towards the achievement of this goal, in this position paper we start focusing on MongoDB (http://www.mongodb.org), the most popular NoSQL datastore (see http://db-engines.com/en/ranking). 

More precisely, we present a research roadmap related to the enhancement of MongoDB with fine grained context-aware access control.Based on recent surveys the interest for NoSQL datastores is continuously growing. This trend is an expected consequence of key features of these systems, such as their ability to efficiently handle high data volumes, and to easily scale. Indeed, NoSQL datastores outclass 

traditional relational database management systems (RDBMs) wrt performance and scalability. In addition, with innovative computational paradigms, such as MapReduce and the aggregation pipeline, NoSQL datastores also overcome RDBMSs in terms of achievable forms of data analysis. At the same time, they provide only poor support for data protection, which is typically limited to a very basic form of access control. 

This aspect is seen by several managers as the top obstacle to the use of datastores in industry [14]. We believe that this issue can be addressed by integrating proper data 

protection mechanisms into NoSQL datastores. However, due to the variety of existing NoSQL datastores, each characterized by a given data model and query language, this appears as a very ambitious task. In order to do a first step towards the achievement of this goal, in this position paper we start focusing on MongoDB (http://www.mongodb.org), the most popular NoSQL datastore (see http://db-engines.com/en/ranking). 

More precisely, we present a research roadmap related to the enhancement of MongoDB with fine grained context-aware access control.},

keywords = {},

pubstate = {published},

tppubtype = {inproceedings}

}

@inproceedings{DBLP:conf/smartcity/LalehCF15,

title = {Graph Based Local Risk Estimation in Large Scale Online Social Networks},

author = {Naeimeh Laleh and Barbara Carminati and Elena Ferrari},

url = {https://doi.org/10.1109/SmartCity.2015.124},

doi = {10.1109/SmartCity.2015.124},

year  = {2015},

date = {2015-01-01},

booktitle = {2015 IEEE International Conference on Smart City/SocialCom/SustainCom 

 2015, Chengdu, China, December 19-21, 2015},

pages = {528--535},

publisher = {IEEE Computer Society},

abstract = {Online Social Networks (OSNs) have become extremely popular in recent years, leading to the presence of huge volumes of users' personal information on the Internet. This increases the need for efficient and effective measures helping users to judge their direct contacts so as to avoid friendship with malicious users that could misuse their personal information. At this purpose, in this paper we propose a risk measure, called local risk factor, having as a key idea the fact the malicious users in OSNs (aka attackers) show some common features on the topology of their social graphs, which is different from those of legitimate users. This consideration brought us to design a set of features defined based on attacker activity patterns. To prove the effectiveness of the proposed risk measure, we run several experiments on a real OSN dataset (i.e., Orkut social network) with more than 3 million vertices and 117 million edges, by injecting synthetic fake users according to different settings and showing how the proposed measures can indeed help in their detection.},

keywords = {},

pubstate = {published},

tppubtype = {inproceedings}

}

@inproceedings{DBLP:conf/asunam/AkcoraF14,

title = {Discovering trust patterns in ego networks},

author = {Cuneyt Gurcan Akcora and Elena Ferrari},

editor = {Xindong Wu and Martin Ester and Guandong Xu},

url = {https://doi.org/10.1109/ASONAM.2014.6921587},

doi = {10.1109/ASONAM.2014.6921587},

year  = {2014},

date = {2014-01-01},

booktitle = {2014 IEEE/ACM International Conference on Advances in Social Networks 

 Analysis and Mining, ASONAM 2014, Beijing, China, August 17-20, 

 2014},

pages = {224--229},

publisher = {IEEE Computer Society},

abstract = {In the past decade, online social networks have provided invaluable data in understanding how social networks change in time while attracting new users and fostering relationships among members. The concept of social trust was developed to explain why and how much users trust each other to become friends or expose their personal data. Existing work on social trust analyze behavioral features and profile attributes to find trust between pairs of users. Although useful, these works suffer from the problem of incomplete, inaccurate and inconsistent social network data. 

 We approach the problem of analyzing trust from an ego network perspective. We observe new friendships, group formations and structural roles of users in ego networks to outline three trust questions. Answers to these questions provide insights into how social trust can be measured from user connections.},

keywords = {},

pubstate = {published},

tppubtype = {inproceedings}

}

@inproceedings{DBLP:conf/cikm/LuciaF14,

title = {EgoCentric: Ego Networks for Knowledge-based Short Text Classification},

author = {William Lucia and Elena Ferrari},

editor = {Jianzhong Li and Xiaoyang Sean Wang and Minos N Garofalakis and Ian Soboroff and Torsten Suel and Min Wang},

url = {https://doi.org/10.1145/2661829.2661990},

doi = {10.1145/2661829.2661990},

year  = {2014},

date = {2014-01-01},

booktitle = {Proceedings of the 23rd ACM International Conference on Conference 

 on Information and Knowledge Management, CIKM 2014, Shanghai, China, 

 November 3-7, 2014},

pages = {1079--1088},

publisher = {ACM},

abstract = {Classification of short text messages is becoming more and more relevant in these years, where billion of users use online social networks to communicate with other people. Understanding message content can have a huge impact on many data analysis processes, ranging from the study of online social behavior to targeted advertisement, to security and privacy purposes. 

In this paper, we propose a new unsupervised knowledge-based classifier for short text messages, where each category is represented by an ego-network. 

A short text is classified into a category depending on how far its words are from the ego of that category. We show how this technique can be used both in single label and in multi-label classification, and how it outperforms the state of the art for short text messages classification.Classification of short text messages is becoming more and more relevant in these years, where billion of users use online social networks to communicate with other people. Understanding message content can have a huge impact on many data analysis processes, ranging from the study of online social behavior to targeted advertisement, to security and privacy purposes. 

In this paper, we propose a new unsupervised knowledge-based classifier for short text messages, where each category is represented by an ego-network. 

A short text is classified into a category depending on how far its words are from the ego of that category. We show how this technique can be used both in single label and in multi-label classification, and how it outperforms the state of the art for short text messages classification.},

keywords = {},

pubstate = {published},

tppubtype = {inproceedings}

}

@inproceedings{DBLP:conf/icdcs/BahriCF14,

title = {Community-Based Identity Validation on Online Social Networks},

author = {Leila Bahri and Barbara Carminati and Elena Ferrari},

url = {https://doi.org/10.1109/ICDCS.2014.11},

doi = {10.1109/ICDCS.2014.11},

year  = {2014},

date = {2014-01-01},

booktitle = {IEEE 34th International Conference on Distributed Computing Systems, 

 ICDCS 2014, Madrid, Spain, June 30 - July 3, 2014},

pages = {21--30},

publisher = {IEEE Computer Society},

abstract = {Identity management in online social networks (OSNs) is a challenging, yet important requirement for effective privacy protection and trust management. Literature offers several proposals addressing issues related to identity breaches and/or identity related attacks on OSNs, but only a few aim at giving means to judge users' reliability in terms of trustworthiness of their claimed identities. In this paper, we propose an identity validation process that relies on OSN community feedback to assign to OSN users identity trustworthiness levels. For this purpose, we define a community based supervised learning process to detect the set of attributes in a user profile for which it is expected to see a correlation among their values (e.g., job and salary). Once these correlated attribute sets are identified, the profile of a target user is judged by a selected group of raters to estimate her identity trustworthiness level. We demonstrate the effectiveness of our proposal through experimentation under two different scenarios and using real data. The experiments' results under the two scenarios demonstrate the effectiveness and meaningfulness of our proposal.},

keywords = {},

pubstate = {published},

tppubtype = {inproceedings}

}

@inproceedings{DBLP:conf/icws/CarminatiFT14,

title = {Secure Web Service Composition with Untrusted Broker},

author = {Barbara Carminati and Elena Ferrari and Ngoc Hong Tran},

url = {https://doi.org/10.1109/ICWS.2014.31},

doi = {10.1109/ICWS.2014.31},

year  = {2014},

date = {2014-01-01},

booktitle = {2014 IEEE International Conference on Web Services, ICWS, 2014, 

 Anchorage, AK, USA, June 27 - July 2, 2014},

pages = {137--144},

publisher = {IEEE Computer Society},

abstract = {Composite web services are usually coordinated according to a workflow, composed by several activities, each of which carried out by a service. A way to coordinate this cooperation is orchestration, which implies that the workflow underlying the composite web service is processed by a broker hosting a workflow engine (e.g., BPEL engine). According to the orchestration paradigm, the broker coordinates the invocation of services involved in the composition by passing the needed parameters. In general, all previous proposals for the service orchestration model consider the broker as a trusted entity. As such, they never payed attention to the fact that the broker is able to access several pieces of sensitive data. We believe there is the need to protect them against improper access and usage from partner services as well as the broker. To cope with these issues, in this paper, we propose a protocol based on a selective encryption able to ensure that both the broker and service partners can access only the information needed to fulfill their activities.Composite web services are usually coordinated according to a workflow, composed by several activities, each of which carried out by a service. A way to coordinate this cooperation is orchestration, which implies that the workflow underlying the composite web service is processed by a broker hosting a workflow engine (e.g., BPEL engine). According to the orchestration paradigm, the broker coordinates the invocation of services involved in the composition by passing the needed parameters. In general, all previous proposals for the service orchestration model consider the broker as a trusted entity. As such, they never payed attention to the fact that the broker is able to access several pieces of sensitive data. We believe there is the need to protect them against improper access and usage from partner services as well as the broker. To cope with these issues, in this paper, we propose a protocol based on a selective encryption able to ensure that both the broker and service partners can access only the information needed to fulfill their activities.},

keywords = {},

pubstate = {published},

tppubtype = {inproceedings}

}

@inproceedings{DBLP:conf/cgc/LuciaAF13,

title = {Multi-Dimensional Conversation Analysis Across Online Social Networks},

author = {William Lucia and Cuneyt Gurcan Akcora and Elena Ferrari},

url = {https://doi.org/10.1109/CGC.2013.65},

doi = {10.1109/CGC.2013.65},

year  = {2013},

date = {2013-01-01},

booktitle = {2013 International Conference on Cloud and Green Computing, Karlsruhe, 

 Germany, September 30 - October 2, 2013},

pages = {369--376},

publisher = {IEEE Computer Society},

abstract = {With the advance of the Internet, ordinary users have created multiple personal accounts on online social networks, and interactions among these social network users have recently been tagged with location information. In this work, we observe user interactions across two popular online social networks, Facebook and Twitter, and analyze which factors lead to retweet/like interactions for tweets/posts. In addition to the named entities, lexical errors and expressed sentiments in these data items, we also consider the impact of shared user locations on user interactions. In particular, we show that geolocations of users can greatly affect which social network post/tweet will be liked/ retweeted. We believe that the results of our analysis can help researchers to understand which social network content will have better visibility.},

keywords = {},

pubstate = {published},

tppubtype = {inproceedings}

}