37e6aff

Την Πέμπτη 11/05/2017, το Τμήμα Πληροφορικής και Τηλεματικής του Χαροκοπείου Πανεπιστημίου, φιλοξενεί τον κ. Ηλία Πετρούνια, Associate Professor, Alliance Manchester Business School, University of Manchester, UK για μία διάλεξη με αντικείμενο: “A Semantic-Based Framework for Fine Grained Sentiment Analysis”, στο πλαίσιο του σεμιναρίου του Μεταπτυχιακού Προγράμματος Σπουδών.

Η διάλεξη θα πραγματοποιηθεί στο αμφιθέατρο του κτιρίου του Τμήματος Πληροφορικής και Τηλεματικής (Ομήρου 9, Ταύρος), στις 18:30 μ.μ.. 

Θα θέλαμε επίσης να σας υπενθυμίσουμε ότι η διάλεξη αποτελεί τμήμα του Σεμιναρίου και η παρουσία των μεταπτυχιακών φοιτητών είναι υποχρεωτική.  

 

Με εκτίμηση, 

Η Γραμματεία του ΠΜΣ 


(ακολουθεί σύντομο βιογραφικό σημείωμα και περίληψη της διάλεξης)

Dr Ilias Petrounias is an Associate Professor at the Alliance Manchester Business School, University of Manchester, UK. He received his BSc from the National Technical University of Athens (School of Electrical and Computer Engineering) and his MSc (by Research) and PhD from the University of Manchester. He was the Organising Chair for the 2014 7th IFIP 8.1 Working Conference on the Practice of Enterprise Modelling (PoEM).  He has supervised to completion 12 PhD and 15 MPhil students. He has participated as primary and co-investigator number of EU and UK research projects totalling more than £1.2 million. Dr Ilias Petrounias researches business intelligence, with particular interests in data mining and big data analytics and competitive intelligence. The techniques developed are based on either algorithmic or neural network approaches. They have been applied in a variety of domains, including healthcare management, electricity supply prediction and food management. Recent research is concerned with data analytics for social media and traffic control management. 

Abstract 

The arrival of Web 2.0 has allowed people to share online opinions and sentiments about products and/or services. This is helpful for them in understanding the qualities of these products or services. At the same time, it has also created opportunities for organisations to try and understand how customers feel towards their products and then attempt to improve them. However, the opinions and sentiments of customers are large and, typically, unstructured, which makes it difficult to analyse automatically. This work proposes an overall approach towards achieving this, which is accompanied by a practical implementation. The approach allows to analyse sentiments expressed within customer reviews about specific products and the aspects of the products that these reviews relate to. Two case studies from two different domains (phone and hotel domains) are used in order to demonstrate the validity of the approach. A system has been developed, which provides 91.3% accuracy in sentiment classification and 92.5% accuracy for the aspect analysis generated automatically across the two domains that were tested.