Μεταπτυχιακές Διατριβές

Μόνιμο URI για αυτήν τη συλλογήhttps://dspace.library.tuc.gr/handle/123456789/121

Νέα

Περιηγούμαι

Τώρα δείχνει 1 - 12 από 12

A bayesian personalized recommendation system
(Πολυτεχνείο Κρήτης, 2014) Babas Konstantinos; Μπαμπας Κωνσταντινος; Chalkiadakis Georgios; Χαλκιαδακης Γεωργιος; Lagoudakis Michael; Λαγουδακης Μιχαηλ; Deligiannakis Antonios; Δεληγιαννακης Αντωνιος
In this research, we propose a novel Bayesian approach for personalized recommendations. We succeed in providing recommendations that are entirely personalized, based on a user’s past item “consumptions”, building a representative user model which reflects agent’s corresponding beliefs. Having a set of items, our agent has to select the one which better matches her beliefs about a specific user, in order to recommend it and receive the corresponding reward. In our approach, we model both user preferences and items under recommendation as multivariate Gaussian distributions; and make use of Normal-Inverse Wishart priors to model the recommendation agent beliefs about user types. We interpret user ratings in an innovative way, using them to guide a Bayesian updating process that helps us both capture a user’s current mood, and maintain her overall user type. We produced several variants of our approach, and applied them in the movie recommendations domain, evaluating them on data from the MovieLens dataset. We developed a generic & domain independent system, able to face the scalability challenge and able to capture user preferences (long-term and short-term). Moreover, we dealt with the exploration vs exploitation dilemma in this domain, via the application of various exploration algorithms (e.g., VPI exploration). Ours is a completely personalized approach, which exploits Bayesian Reinforcement Learning in order to recommend an item or a top-N group of items, without the need of ratings prediction. We do not employ a Collaborative Filtering or Content-based or Preference Elicitation technique, but we are still able to provide successful recommendations. Furthermore, we tackle the famous “cold-start” problem via the use of Bayesian and VPI explorations. Our algorithms are shown to be competitive against a state-of-the-art method, which nevertheless requires a minimum set of ratings from various users to provide recommendations --- unlike our entirely personalized approach.
BucDoop: Bottom Up Computation of Iceberg Data Cubes with Hadoop
(Πολυτεχνείο Κρήτης, 2014) Tsakonas Konstantinos; Τσακωνας Κωνσταντινος; Deligiannakis Antonios; Δεληγιαννακης Αντωνιος; Garofalakis Minos; Γαροφαλακης Μινως; Christodoulakis Stavros; Χριστοδουλακης Σταυρος
Big Data analysis has been a key matter during the recent years for the study of various phenomena in various science contexts as well as in business intelligence. Furthermore it appears for good reason to remain in focus for the future. Online Analytical processing methods and Data Cubes need to be further studied in order to reduce time used for efficient data analysis. This study introduces BucDoop, a novel algorithm that exploits the parallelism benefits of Hadoop Map Reduce, for the efficient iceberg data cube creation in reasonable time. BucDoop includes the use of the Bottom Up Computation (BUC) idea in the context of iceberg cube data lattice traversal, managing to reduce the amount of data handled with early pruning architecture and producing the portion of the cube needed for analysis purposes (iceberg problem). Experiments conducted herein present an efficient scalability factor for the creation of the iceberg cube for very big data, by-passing the data explosion and memory constraints problem while using only commodity hardware.
Channel coding and detection for increased range bistatic scatter radio
(Technical University of Crete, 2014) Alevizos Panagiotis; Αλεβιζος Παναγιωτης; Bletsas Aggelos; Μπλετσας Αγγελος; Karystinos Georgios; Καρυστινος Γεωργιος; Deligiannakis Antonios; Δεληγιαννακης Αντωνιος
Scatter radio, i.e communication by means of reflection, has emerged as a potential key-enabling technology for ultra low-cost, large-scale, ubiquitous sensor networking. This work jointly studies noncoherent detection and channel coding for scatter radio networks, with the ultimate goal to further extend communication range, compared to prior art. Specifically, this work focuses on frequency shift keying modulation (FSK) in bistatic scatter radio architectures, where carrier emitter is dislocated from the software defined radio receiver (SDR). FSK is ideal for the power limited regime and allows for simple, frequency division multiple access (FDMA) of simultaneously operating receiver-less sensors. A novel composite hypothesis testing decoding rule is derived for noncoherent channel-encoded FSK, in bistatic scatter radio architectures. Such decoding rule is evaluated with short block length channel codes; the latter offer ultra-low encoding complexity, and thus, they are appropriate for resource-constraint scatter radio sensors. Reed-Muller and BCH codes are studied, due to their strong algebraic structure. It is shown that the proposed decoding scheme achieves high diversity order through interleaving. Extensive simulations under Rician fading scenarios include the impact of carrier frequency offset estimation errors, channel coherence time and interleaving depth. Closed-form performance analysis is also provided. Theoretical analysis for maximum likelihood coherent detection and decoding in on-off keying modulation (OOK) is also presented. Furthermore, experimental measurements are conducted outdoors, with a commodity SDR reader and custom scatter radio sensor. Sensor-to-reader ranges up to 134 meters are experimentally demonstrated with omnidirectional antennas and 13 dBm (20 milliWatt) transmission power. Coded setup offered 10 additional meters range extension compared to the state-of-the-art uncoded noncoherent detection. As a result, this thesis provided a simple solution that could further leverage the adoption of scatter radio in large-scale, ultra low-cost wireless sensor networks
Coherent detection and channel coding for backscatter sensor networks
(Πολυτεχνείο Κρήτης, 2014) Fasarakis-Chilliarnt Nikos; Φασαρακης-Χιλλιαρντ Νικος; Bletsas Aggelos; Μπλετσας Αγγελος; Karystinos Georgios; Καρυστινος Γεωργιος; Deligiannakis Antonios; Δεληγιαννακης Αντωνιος
Crowdsourcing and management of nature observational data
(Πολυτεχνείο Κρήτης, 2014) Skevakis Giannis; Σκευακης Γιαννης; Christodoulakis Stavros; Χριστοδουλακης Σταυρος; Mania Aikaterini; Μανια Αικατερινη; Deligiannakis Antonios; Δεληγιαννακης Αντωνιος
Observations of plants and animals in nature is highly valued information for the experts in the area of biodiversity. They can be used to define changes in the population of animals, plants, or track their movements throughout long periods of time. Moreover, the richer the information following the observations, the more knowledge can be extracted from them. However, the limited number of experts and the limited funding in the area, makes the observation gathering procedure almost impossible. We present the design and implementation of a framework for the management of bio- diversity observations captured by users roaming in the nature. This aims to alleviate the need for experts capturing biodiversity information, and propagates the collection of information to simple users wandering in the nature. Our framework consists of a model supporting the observations, and an infrastructure that allows the capturing, enrichment and storage of the observations using state- of-the-art technologies. Our architecture provides a scalable, highly efficient management of the collected data. The collection of the observational data is performed in real-time using mobile de- vices that most of the people have available with them, like mobile phones and tablets. Additionally, we describe the meta-model that we have defined, allowing the personalization of the metadata that follow the observations. This provides our framework with the freedom and extensibility needed so as to be implemented for various domains other than biodiversity. Finally, we describe the process of migrating the data collected by the Natural Europe project to our infrastructure.
Practical secure and efficient range search
(Technical University of Crete, 2015) Demertzis Ioannis; Δεμερτζης Ιωαννης; Garofalakis Minos; Γαροφαλακης Μινως; Deligiannakis Antonios; Δεληγιαννακης Αντωνιος; Christodoulakis Stavros; Χριστοδουλακης Σταυρος
Due to their potential for near-infinite scalability, cloud computing platforms are rapidly becoming the defacto standard for large-scale, big data analytics. Still, serious concerns regarding the outsourcing and querying of private company and personal data remain a key roadblock in the adoption of such cloud platforms for numerous big-data applications. In this work, we extend cryptographic Searchable Symmetric Encryption (SSE) schemes to create the first adaptive Range Searchable Symmetric Encryption (RSSE) schemes that allow the execution of range queries in a practical, efficient, and secure manner. We propose a number of new RSSE schemes, that we analytically prove to be adaptively secure according to a novel, cryptographic security definition (RQ-CKA2), and also exhibit interesting security and performance trade-offs. We also tackle the challenge of updates in our RSSE schemes by proposing a general solution that does not introduce any additional leakage over the static case, other than the number of inserts/deletes. The practicality and scalability of our proposed schemes is demonstrated both theoretically and experimentally. More specifically, our techniques outperform state-of-the-art Privacy Preserving Range Querying approaches in terms of both security and efficiency and, at the same time, offer worst-case guarantees on possible leakages and also protect sensitive information regarding the order of encrypted values.
SDR readers for Gen2 RFID and backscatter sensor networks
(Technical University of Crete, 2015) Kargas Nikolaos; Καργας Νικολαος; Bletsas Aggelos; Μπλετσας Αγγελος; Karystinos Georgios; Καρυστινος Γεωργιος; Deligiannakis Antonios; Δεληγιαννακης Αντωνιος
Scatter radio has emerged as a key enabling technology for low-cost and large scale ubiquitous sensing. Radio frequency identification (RFID) tags/sensors utilize scatter radio technology to transfer sensed information to readers, typically employing Gen2, the industrial RFID protocol. This work offers a complete software-defined radio Gen2 reader, based on GNU Radio and USRP2 commodity software defined radio (SDR) platform. In sharp contrast to prior art, a single radio front end card is used with coherent detection and optimal exploitation of the FM0 line coding memory. The reader can act as a research tool to experiment with state-of-the-art signal processing algorithms and RFID devices. The two tag collision problem is studied and problems that arise in a real world system, such as channel estimation and tag symbol synchronization are highlighted. Experimental measurements are conducted and it is shown that the reader can identify a commercial, passive UHF RFID tag up to 6 meters with acceptable reliability. In addition, it is shown that collision recovery algorithms can increase performance of the implemented reader. Furthermore, an implementation of a SDR reader for a wireless backscatter sensor network (BSN) is presented. The developed reader implements noncoherent frequency shift keying (FSK) detection. The reader can decode multiple tags in real time and achieves communication ranges with semi-passive tags/sensors up to 130 meters.
Self-powered plant sensor for scatter radio
(Technical University of Crete, 2015) Konstantopoulos Christos; Κωνσταντοπουλος Χρηστος; Koutroulis Eftychios; Κουτρουλης Ευτυχιος; Bletsas Aggelos; Μπλετσας Αγγελος; Deligiannakis Antonios; Δεληγιαννακης Αντωνιος
In agriculture applications (e.g. greenhouses, vineyards etc.) it is required to automatically gather information about environmental variables such as soil and air humidity, as well as temperature in the vicinity of plants within the same field, with low-cost and high-scalability. Thus, sensor networks that are extending over a broad area and gather environmental data for microclimate monitoring, are indispensable for the application of optimal crop management techniques. The field of plant electro-physiology investigates the correlation of environmental variables with the electrical signals that are produced by diverse types of plants. Existing research in measurement of electrical signals generated by plants has been conducted using high-cost equipment, such as laboratory multi-meters and data-loggers, in order to perform the signal-conditioning and data acquisition operations required. This thesis introduces for first time in the existing research literature a novel low cost and self-powered sensor node that belongs to a large-scale scatter radio network and simultaneously is powered in a parasitic way to the plants, as well as is able to acquire and transmit these types of signals from each plant. Furthermore, in the context of this thesis, several experimental prototypes of the proposed node were developed, as well as used to gather measurements of electrical signals that are generated from multiple Avocado plants. The experimental results demonstrate the successful operation of the proposed WSN node, as well as indicate the correlation of plants signals with solar irradiation and plant irrigation events. Thus, the proposed system can be employed in precision agriculture applications for automated irrigation scheduling, control of the plant ambient conditions etc. based on data derived directly by the plants.
Sketch-based geometric monitoring of distributed stream queries
(Technical University of Crete, 2012) Athanasoglou Konstantinos; Αθανασογλου Κωνσταντινος; Garofalakis Minos; Γαροφαλακης Μινως; Samoladas Vasilis; Σαμολαδας Βασιλης; Deligiannakis Antonios; Δεληγιαννακης Αντωνιος
The SPARQL-RW framework: mapping modeling and query rewriting for ontology based mediators
(Πολυτεχνείο Κρήτης, 2014) Makris Konstantinos; Μακρης Κωνσταντινος; Christodoulakis Stavros; Χριστοδουλακης Σταυρος; Deligiannakis Antonios; Δεληγιαννακης Αντωνιος; Lagoudakis Michael; Λαγουδακης Μιχαηλ
The Web of Data is an open environment consisting of heterogeneous, distributed and highly structured information sources. Uniform information access in this kind of setting is of major importance for data consumer applications and end users. To this end, ontology based mediator systems supporting transparent query access over federated data sources are considered essential. In this thesis we present SPARQL-RW, a Framework supporting mapping modeling and query rewriting in the context of ontology based mediator architectures. SPARQL-RW provides a formal model for describing mappings between ontology schemas, as well as a generic method for SPARQL 1.1 query rewriting, with respect to a set of predefined ontology mappings and data source endpoints. The mapping model supports a set of rich and flexible mapping types based on Description Logic semantics, as well as notable mapping formalisms, including Global-As-View (GAV), Local-As-View (LAV), and Global-and-Local-As-View (GLAV). Additionally, it defines a mapping language capable of representing all the supported types of inter-schema correspondences. The Framework provides functionality for performing mapping inference and for identifying inconsistencies in a given set of mappings and ontology schemas. Regarding query rewriting, the proposed algorithms are proved to provide semantics preserving queries with respect to the GAV mapping types supported by the model. The reformulated queries can be executed directly on any SPARQL federated query engine, or exploited as logical query plans by any ontology based mediator system. SPARQL-RW has been implemented, formally evaluated and tested in a prototype mediator system integrating several data providers from the biodiversity community along with DBpedia.
Stochastic PageRank maintenance over shared-nothing architectures
(Technical University of Crete, 2014) Perros Ioakeim; Περρος Ιωακειμ; Garofalakis Minos; Γαροφαλακης Μινως; Lagoudakis Michael; Λαγουδακης Μιχαηλ; Deligiannakis Antonios; Δεληγιαννακης Αντωνιος
VectorL: a macroprogramming language for testing wireless sensor network applications
(Technical University of Crete, 2015) Asimoglou Aikaterini; Ασημογλου Αικατερινη; Samoladas Vasilis; Σαμολαδας Βασιλης; Deligiannakis Antonios; Δεληγιαννακης Αντωνιος; Garofalakis Minos; Γαροφαλακης Μινως
Wireless sensor networks (WSNs) have gained world-wide attention in recent years and are expected to find wide applicability and increasing deployment in the near future. These networks are consisted of small sensors, with limited processing and computing resources. Thus, the considerable cost of deploying and maintaining large-scale WSNs for experimental purposes makes simulation a necessary phase of the application development cycle. Sensor node actions are triggered by the information they sense, measure and gather from the environment. Therefore, physical process modeling and simulation is an integral part of a realistic application simulation. Though, sensing is usually neglected in WSN simulators. The usual practice is to feed random numbers to nodes or each node to have a static value. The purpose of this master thesis is to introduce VectorL, a high-level domain-specific language designed to serve the need for an effective and simple way to model and simulate external environment in order to produce realistic sensor readings. Another great advantage of VectorL is that it can be used, not only during simulation phase, but also during wsn testing phase, inside the motes.

Περιηγούμαι

Πλοήγηση Μεταπτυχιακές Διατριβές ανά Συγγραφέα "Deligiannakis Antonios"