Lectures / Seminars

Paraphrasing and textual entailment recognition and generation

Paraphrasing methods recognize, generate, or extract phrases, sentences, or longer natural language expressions that convey almost the same information. Textual entailment methods, on the other hand, recognize, generate, or extract pairs of natural language expressions, such that a human who reads (and trusts) the first element of a pair would most likely infer that the other element is also true. Paraphrasing can be seen as bidirectional textual entailment and methods from the two areas are often very similar.

Structural-functional landscapes of the human genome – Nuclear Biology in the postgenomic era

Despite of the technological advances at their aim accelerating at an unprecedented pace, researchers in the field of biological sciences are still far from being able to provide concise theoretical frameworks for well-described biological problems. Ten years after the completion of the first draft of the human genome sequence there appears to be poor progress in the field of genomics in terms of understanding of crucial processes such as the regulation of DNA replication and RNA transcription as the bases of the multicellular organization of life.

A Performance Characterization Algorithm for Symbol Localization

In this talk, we will present an algorithm for performance characterization of symbol localization systems. This algorithm is aimed to be a more "reliable" and "open" solution to characterize the performance. To achieve that, it exploits only single points as the result of localization and offers the possibility to reconsider the localization results provided by a system. We use the information about context in ground truth, and overall localization results, to detect the ambiguous localization results.

Detecting Patterns in News Media Content

News from newspapers, magazines, newswires and broadcast media deploy a crucial role in shaping the public opinion, informing the democratic process, and influencing markets and industry. The study of news media is a field of research for social scientists, however most of their work is performed manually. Our research is towards the automation of the analysis of news content by deploying state of the art methodologies from Machine Learning, Data Mining and Natural Language Processing. This has enabled us to analyse millions of articles from hundreds of news outlets and tens of languages.

Γεφυρώνοντας το χάσμα μεταξύ ανεξάρτητων ταξινομητών και δομημένης πρόβλεψης στην εξαγωγή βιοϊατρικών γεγονότων

Αναπτύσσουμε μια δομημένη προσέγγιση εξαγωγής βιοϊατρικών γεγονότων χρησιμοποιώντας το πλαίσιο Μάθηση μέσω Αναζήτησης το οποίο αποδομεί το πρόβλημα της δομημένης πρόβλεψης σε πολλαπλά προβλήματα ταξινόμησης βάση κόστους, των οποίων οι παράμετροι μαθαίνονται συνδυασμένα. Αξιολογώντας την απόδοση στα δεδομένα του διαγωνισμού BioNLP 2009, η δομημένη προσέγγιση βελτιώνει την απόδοση των ανεξάρτητων ταξινομητών κατά 7 μονάδες σε F-score, ενώ διατηρεί την ευκολία υλοποίησης.

Intelligent Electricity Networks and Distributed Intelligence

Η παρουσίαση αφορά την ερευνητική δραστηριότητα του Εργαστηρίου Συστημάτων Ηλεκτρικής Ενέργειας στην περιοχή των Ευφυών Δικτύων Ηλεκτρισμού εστιάζοντας σε εφαρμογές κατανεμημένης νοημοσύνης. Ειδικότερα γίνεται χρήση της Θεωρίας των ευφυών πρακτόρων για διαχείριση καταναλωτών καθώς και μονάδων διεσπαρμένης παραγωγής. Τέλος θα παρουσιαστούν εφαρμογές Διαχείρισης Ενέργειας που αναπτύχθηκαν από το εργαστήριο.

Wireless Battle Mesh v5

The WirelessBattleMesh is an event aiming to bring together developers of mesh protocols, embedded firmwares, wireless drivers as well as applications on top of mesh networks. This wide range of expertise also attracts academics, wireless community members and open source enthusiasts. The coming WirelessBattleMesh edition will take place in Greece in March 2012. This meeting will give you insights of what to expect from such an event and how you could support it. Link: http://battlemesh.org/.

A Tool For Sensor Placement and System Monitoring in Assistive Environments

Assistive Environments is a rapidly growing research field where researchers try to introduce state of the art technology into everyday life aiming to improve its quality. Everyone can benefit from it but first and foremost people with special needs such as elderly, disabled or patients. The proposed tool is part of zScope, which is a web based tool that will be able to monitor human behaviour and is responsible for placing and monitoring the various sensors in the environment.

Decision Models in Adaptive Educational Hypermedia Systems

Adaptive Educational Hypermedia Systems (AEHS) have been proposed as the underlying facilitator for personalized web-based learning with the general aim of personalizing learning experiences for a given learner. Adaptive learning objects selection and sequencing is recognized as challenging research issues in adaptive educational hypermedia systems (AEHS). In order to adaptively select and sequence learning objects in AEHS, the definition of adaptation behaviour, referred to as Adaptation Model, is required.

Effective Capacity Theory for Modeling Systems with Time-Varying Servers, with an Application to IEEE 802.11 WLANs

Πολλές απαιτητικές δικτυακές εφαρμογές επιζητούν στοχαστικές εγγυήσεις ποιότητας υπηρεσίας. Όσον αφορά τις εγγυήσεις που σχετίζονται με τις απώλειες των πακέτων δεδομένων, η ασυμπτωτική θεωρία που βασίζεται στις έννοιες του Ισοδύναμου Εύρους Ζώνης και της Ισοδύναμης Χωρητικότητας αποδείχθηκε κατάλληλη για τον υπολογισμό μικρών πιθανοτήτων απωλειών σε συστήματα αναμονής με (σύνθετες) χρονικά μεταβαλλόμενες διεργασίες εισόδου και εξυπηρέτησης. Επίσης, η ασυμπτωτική θεωρία οδήγησε στη δημιουργία απλών μηχανισμών ελέγχου κίνησης για την παροχή των σχετικών εγγυήσεων

Syndicate content

© 2019 - Institute of Informatics and Telecommunications | National Centre for Scientific Research "Demokritos"

Terms of Service and Privacy Policy