Leuven Statistics Days 2012

KU Leuven, June 7 and 8, 2012

About	Date&Location	Speakers	Programme	Registration	Sponsors

Description and Announcement: Following the tradition of celebrating statistics at KU Leuven, we are glad to announce the biennial Leuven Statistics Days 2012, to be held in Leuven on June 7 and 8, 2012. This edition’s theme is “Mixed models and modern multivariate methods in linguistics.” This international Leuven-based meeting warmly welcomes everyone interested in the methodological and applied themes. Through excellent invited speakers, a number of fine contributions, and a panel discussion, the meeting organizers aim to reach their twofold goal: (1) provide an exchange forum for statistical methods in linguistics and (2) stimulate cross-fertilization, centered around the methodological themes, between researchers in linguistics, bioinformatics, computer science, data mining, engineering, bio-engineering, imaging, and medical statistics, to name a few.
Contributed presentations (oral and poster) are very welcome. To stimulate interaction, the entire meeting is held in a plenary format. Because of that, the number of oral slots available is limited, but the poster sessions are designed to foster interaction.

Leuven Statistics Days (LSD) are supported by LStat (Leuven Statistics Research Centre). The 2012 edition is co-organised with the linguistic research group QLVL (Quantitative Lexicology and Variational Linguistics).

Scientific and organizing committee: Dirk Speelman (chairman), Geert Molenberghs (LStat organizing chair), An Carbonez, Lilian Wassink, Dirk Geeraerts, Kris Heylen.

Date & Location

Leuven Statistics Days 2012
Thursday 7 and Friday 8 June, 2012

Location:

College De Valk

Tiensestraat 41

3000 LEUVEN

Auditorium Zeger Van Hee (room 91.56)

How to reach the location: click here

map

Public parking is available in the vicinity of the congres location (Ladeuze parking).

Invited Speakers

Harald Baayen (Eberhard Karls University, Tübingen; University of Alberta, Edmonton) The main themes in the research of Harald Baayen are morphological productivity, morphological processing, language variation and statistical data analysis. His book on Word Frequency Distributions (Kluwer, 2001) is widely considered to be the reference work on statistical models for word frequency distributions. Together with several of his journal articles his book with Cambridge University Press, Analyzing Linguistic Data: A practical introduction to Statistics using R (2008), offers a comprehensive coverage of mixed models in linguistics and has been very instrumental in the wider adoption of this technique in linguistics.

Marco Baroni (CIMeC, Università di Trento, Roverto, Italy.) Marco Baroni (PhD UCLA, 2000) is a tenured researcher and lecturer at the Center for Mind and Brain Sciences of the University of Trento. His research focuses on using mathematical tools, in particular from linear algebra, to model the induction of linguistic meaning from naturally occurring language data. In 2011, he was awared an ERC starting grant to work on the linguistic notion of compositionality from a vector-based perspective.

Geert Verbeke (Leuven Biostatistics and Statistical Bioinformatics Centre (L-BioStat ) KU Leuven). Geert Verbeke has published extensively on various aspects of mixed models for longitudinal data analysis about which he co-authored and co-edited several text books (Springer Lecture Notes 1997; Springer Series in Statistics 2000 and 2005; Chapman & Hall / CRC 2009). Recent research has focused on the use of mixed models for the joint analysis of multiple outcomes. Applications include modeling of multivariate longitudinal data possibly of a different nature (continuous, binary, counts, …), or the use of multiple longitudinally measured markers in the prediction of a time-to-event outcome. For work in this area, he received the International Biometric Society Award for the best Biometrics paper in 2006.

Luc De Raedt (Department of Computer Science, KU Leuven). Luc De Raedt's research interests are in the areas of Artificial Intelligence, Machine Learning and Data Mining as well as their applications. He is currently working on statistical relational learning, which combines probabilistic graphical models with logical representations and machine learning, the integration of constraint programming with data mining and machine learning principles, the development of programming languages for machine learning, and the analysis of graph and network data. He is also interested in applications of these methods to chemo- and bio-informatics, to natural language processing, vision, robotics and action- and activity learning.

Special contributions from:

Antti Arppe (University of Helsinki)
Roeland van Hout (Radboud University Nijmegen)

Stefan Evert (TU Darmstadt)
Sien Moens (KU Leuven)

Programme

Preliminary programme:

Thursday 7 June, 2012

Time		Speaker	Affiliation	Title
9.00-9.30	Registration and coffee
Chair: Dirk Speelman
9.30 - 9.45	Opening of the meeting	Irène Gijbels Dirk Speelman	Chair LStat Chair Scientific and organizing committee
9.45 - 10.30	Keynote lecture	Harald Baayen	Eberhard Karls University, Tübingen	Mixed-effects models in linguistics and psycholinguistics: A useR's perspective. (presentation)
10.30 - 11.00	Coffee break
11.00 - 11.30	Special contributor	Antti Arppe	University of Helsinki	Mixed-effects logistic regression modeling and analysis for polytomous outcomes without a reference category. (presentation)
11.30 - 12.00		Bram Vandekerckhove	University of Antwerp	Identifying and tracing tussentaal in Flemish TV fiction between 1980 and today (presentation)
12.00 - 12.30	Poster teaser			Slides
12.30 - 13.30	Lunch/Poster session
Chair: Geert Molenberghs
13.30 - 14.15	Keynote lecture	Geert Verbeke	KU Leuven	Mixed models with applications to large data sets. (presentation)
14.15 - 14.45	Special contributor	Roeland van Hout	Radboud University Nijmegen	Mixed models, mixed feelings. (abstract)
14.45 - 15.15		Job Schepens	Radboud University Nijmegen	The L2 Impact on Acquiring Dutch as a L3: the L2 Distance Effect (presentation)
15.15 - 15.45	Coffee break
15.45 - 16.15		Dannielle Barth Vsevolod Kapatsinski	University of Oregon	Evaluating Mixed-Models with LOOCV and Effect Size (presentation)
16.15 - 16.45		Martijn Wieling	University of Groningen	Lexical Differences between Tuscan Dialects and Standard Italian: A Sociolinguistic Analysis using Generalized Additive Mixed Modeling (presentation)
16.45 - 17.15	Annual General Meeting for LStat members

Friday 8 June, 2012

Time		Speaker	Affiliation	Title
Chair: Kris Heylen
9.00 - 9.45	Keynote lecture	Marco Baroni	Università di Trento, Roverto	Compositional operations to represent phrases and sentences in distributional semantics (presentation)
9.45 - 10.15		Samuel Iddi	KU Leuven	A Combined Overdispersed and Marginalized Multilevel Model (presentation)
10.15 - 10.45	Coffee break
10.45 - 11.15		Paul De Boeck	KU Leuven	Doubly Mixed Designs with Covariates in Three Steps (presentation)
11.15 - 11.45		Elasma Milanzi	Hasselt University	Quantifying expert opinion for drug discovery with high dimensional data. (presentation)
11.45 - 12.15		Rui Rothe-Neves	UFMG - Brazil	Formant transition as a cue to place of articulation in Brazilian Portuguese coronal fricatives (presentation)
12.15 - 13.30	Lunch/Poster session
Chair: Irène Gijbels
13.30 - 14.15	Keynote lecture	Luc De Raedt	KU Leuven	Analysing structured data – symbolic and probabilistic approaches. (presentation)
14.15 - 14.45	Special contributor	Marie-Francine Moens	KU Leuven	Cross-Language Probabilistic Topic Models. (presentation)
14.45 - 15.15	Coffee break
15.15 - 15.45	Special contributor	Stefan Evert	TU Darmstadt	The role of dimensionality reduction in distributional semantics (presentation)
15.45 - 16.45	Panel discussion	Antti Arppe, Harald Baayen, Marco Baroni, Luc De Raedt, Stefan Evert, Roeland van Hout, Geert Verbeke.		Slides
16.45 -	Closure and Reception	Geert Molenberghs Kris Heylen

Poster presentations:

Presenter	Affiliation	Title
Ann Bertels	KU Leuven	The Exploration of Polysemy in a Technical Corpus: a Stepwise Multiple Regression Analysis (abstract)
Wilfried Cools	KU Leuven KULAK	Learning to Conjugate Verbs in a Second Language (abstract)
Dirk Goldhahn	University of Leipzig	Finding Language Universals: Multivariate Analysis of Language Statistics using the Leipzig Corpora Collection (abstract)
Kris Heylen	KU Leuven	A distributional corpus analysis of Dutch endo- and exocentric compounds. (abstract)
Damazo Kadengye	KU Leuven	Multiple imputation for missing binary item scores in multilevel cross-classified educational data when the Analysis and Imputation models differ (abstract)
Stijn Luca	K.H.Kempen University College	Detecting hypermotor seizures using extreme value statistics (abstract)
Gamze Özel	Hacettepe University	Special Cases of Compound Poisson Process for Word Occurrences in DNA Sequences. (abstract)
Koen Plevoets	Ghent University	The correspondence analysis of partitioned tables with multiple factors (abstract)
Tom Ruette	KU Leuven	Regional varieties and their vowels: Individual Differences Scaling on the phonetic distances between Northern American cities.(abstract)
Clara Vanderschueren	Ghent University	The Portuguese inflected infinitive: empirical approaches compared. (abstract)
Jelle Van Eyck	KU Leuven	Data mining techniques for predicting acute kidney injury after elective cardiac surgery (abstract)
Mathias Verbeke	KU Leuven	A Statistical Relational Learning Approach to Identify Sections in Scientific Abstracts Using Sentence and Document Structure. (abstract)
Jurgen Vercauteren	KU Leuven	The prevalence of multidrug resistant HIV-1 decreased during the last decade in Portugal: a mixed model accounting for multiple measures per patient. (abstract)
Kelly Wauters	KU Leuven KULAK	Evaluating growth modeling and change tracking for the estimation of the learner’s progress in ability level. (abstract)
Laure Wynants	KU Leuven	Variable selection for prediction models based on multicenter data (abstract)
Weiwei Zhang	KU Leuven	(Non)metonymic expressions for GOVERNMENT in Chinese: A mixed regression (abstract)