November 26, 2018

21st IRIXYS Workshop – Keynote speeches

Keynote speeches



Prof. Jacques Savoy (Univ. Neuchâtel)

“Who is the secret hand behind Elena Ferrante?”

Abstract:

Elena Ferrante is a pen name known worldwide, authoring novels such as the bestseller of the tetralogy “My Brilliant Friend” (TV series by HBO). A recent study indicates that the true author behind these books is probably Domenico Starnone. Our study aims to select a set of approved authorship methods and appropriate feature sets to prove, with as much certainty as possible, that this conclusion is correct. To achieve this, a corpus of contemporary Italian novels has been generated, containing 150 books written by 40 authors (including seven by Ferrante). Six authorship identification models have been applied to this dataset (Delta, Labbé’s distance, nearest shrunken neighbors (NSC), naïve Bayes, k-NN, and character n-grams). Using either an instance- or profile-based matching technique, the same result (Starnone) appears very often in first place. Modifying the feature set to include between 50 and 2,000 of the most frequent tokens or lemmas does not change this result. When removing Starnone’s novels from the corpus, all approved attribution methods tend to indicate different names as the most probable author. This result seems to confirm not only that the outputs of these methods are independent, but also that the true author is certainly Starnone. Finally, a lexical analysis reveals the reasons justifying this conclusion.
Keywords : Text categorization, authorship attribution, stylometry.

Biography:

Prof. Jacques Savoy is full Professor in Computer Science at University of Neuchatel (Switzerland). J. Savoy received a Ph.D. in quantitative economics from the University of Fribourg (Switzerland) in 1987. From 1987-92 he was on the faculty of Computer Science at the University of Montreal (Canada). His research interests cover mainly natural language processing and particularly information retrieval for other languages than English (European, Asian, and Indian) as well as multilingual and cross-lingual information retrieval. He participated from many years to various evaluations campaigns (CLEF, NTCIR and TREC) dealing with these questions. His current research interests are related to text clustering and categorization as well as authorship attribution.



Prof. Dietmar Jannach (Univ. Klagenfurt)

“Session-basedRecommendation: Challenges and Recent Advances”

Abstract:

In many applications of recommender systems, a larger fraction of the user population are first-time users or are not logged in when they use the service. In these cases, the item suggestions by the recommender cannot be based on individual long-term preference profiles. Instead, the recommendations have to be determined based on the observed short-term behavior  of the users. Due to the high practical relevance of session-based recommendation, different proposals were made in recent years to deal with the particular challenges of the problem setting . In this talk we will review some of these challenges and provide a survey on recent advances in the field. A specific focus on the talk will be on the particularities of the e-commerce domain.

Biography:

Dietmar Jannach is a full professor of Information Systems at AAU Klagenfurt, Austria. Before joining AAU in 2017, he was a professor of Computer Science at TUDortmund, Germany. In his research, he focuses on the application of intelligent system technology to practical problems and the development of methods for building knowledge-intensive software applications. In the last years, Dietmar Jannach worked on various practical aspects of recommender systems. He is the main author of the first text book on the topic published by Cambridge University Press in 2010 and was the co-founder of a tech startup that created an award-winning product for interactive advisory solutions.


Prof. Mounir Frikha (SUPCom Tunis)

“ResearchActivities in Research Lab MEDIATRON in SUP’COM and Collaborations with UP”

Biography:

Mounir Frikha is currently the Director of the Research Lab MEDIATRON at Carthage University in Tunisia, and was the Director of SUP’COM, the first Engineering School in Telecommunications in Tunisia from 2011 to 2017. He received his engineering degree in 1993 and his Ph.D. degree in 1999, both in Telecommunications from the University of Braunschweig, Germany. He was an associate Professor from 1994 to 1999 at the Institute of Communications Systems (ICS), Technical University of Braunschweig, Germany, and was the leader in development of projects in cooperation with the industry (e‑plus: 3rd mobile phone operator in Germany).

He worked from 1999 to 2001 at Siemens, on cellular networks planning projects and was a project leader in many projects between Siemens and the operator of cellular networks in Morocco and Lebanon, and a leader for the planning of the cellular network for the second license in Tunisia.

He was the Head of the Network Department at SUP’COM from 2005 to 2011 and was involved in many European projects to improve the quality of teaching (TEMPUS) and was responsible for a Master Degree at SUP’COM.

He is president and member of several international conferences program committees, and the author of about 140 international conference and journal papers and several books within the Hermes collection and Wiley Library. His research interests include the management of Quality of Service, Networking, Mobility and Traffic Engineering.


Nicolas Mahler (Datapred)

“An Introduction to Datapred: a Machine Learning Software for Industrial Time Series.”

Abstract :

Datapred is a startup, that edits an industry-strength machine learning software for time series. In this presentation, we will particularly show how Datapred addresses the tedious and potentially dangerous aspects of machine learning for time series (data sequentialization, adaptation of batch algorithms, sequential aggregation, backtests, backend optimization), and illustrate operational benefits on industrial use cases.

Biography :

 Nicolas Mahler is co-founder of Datapred. He has 15 years of experience in data analysis and statistical modeling in multiple industries (particularly market finance, real-time bidding, sales and marketing, logistics and industrial performance). He graduated with a PhD in machine learning from ENS Paris-Saclay, a M.Sc. in probability and statistical theory from Université Paris 6 and a B.Sc. in computer science from Ecole des Mines de Paris.


Henri Bourdeau is product manager at Datapred. He is particularly in charge of the prioritization and development of new features in the Datapred software. He graduated with a M.Sc. in computer science from Ecole Polytechnique and with a M.Sc. in management science from Ecole Polytechnique.


“Viewpoints Discovery based on Topic Models”

Prof. Mohand Boughanem (Univ. Paul Sabatier)

Abstract:

In his talk Mohand Boughanem will be dealing with viewpoints detection in social networks. A viewpoint is a collective opinion shared by a group, for i.e the viewpoint on the marriage equality. In this research work, we try to detect automatically those viewpoints and at the same time the topics discussed in the viewpoints. Thisresearch work uses a Latent Dirichlet allocation (LDA) approach.

Biography :

Mohand Boughanem is professor at the University Paul Sabatier in Toulouse, France. His research interests mainly fall in the area of information retrieval, in particular  information retrieval models, contextual and social information retrieval, blog search, opinion detection and aggregated search (information synthesis).