WILLKOMMEN
ONLINE REGISTRIERUNG
DEADLINES
PROGRAMM
KONTAKT

Workshops


Workshop 1: An Introduction to Bayesian Statistics for Ecologists

Short titel:

Bayesian Statistics

Lecturers:

• Florian Hartig, Uni Regensburg, Universitätsstraße 31, 93053 Regensburg, DE, 0049 (0)941 943-4316, florian.hartig@biologie.uni-regensburg.de
• Jörn Pagel, Uni Hohenheim, August-von-Hartmann-Str. 3, 70599 Stuttgart, DE, 0049 (0)711 459-22069, joern.pagel@uni-hohenheim.de

Duration:

1.5 days

Date/Time:

Sunday 08.09.2019, Time 9:30 – 17:30
Monday 09.09.2019, Time 9:30-12:30

Location:

Fürstenberghaus, F072

Costs:

50 €

Maximum number of participants:

25

Course content:

Bayesian statics has become an important item in the toolbox of a modern quantitative ecologists. In this short course, we will introduce you to the basics of a Bayesian analysis in Jags, including the philosophical underpinnings of Bayesian inference, Jags, working with MCMC samplers, choice of priors, model diagnostics and Bayesian model selection. The goal of the course is to enable you to conduct basic Bayesian regression analysis in Jags.

The course is targeted towards ecologists that
  1. Have little or no prior experience in Bayesian statistics, but want to learn it
  2. Have a good grasp of R (we will assume that participants can independently manipulate and plot data, and run typical statistical models)
  3. Have a good understanding of frequentist statistics (MLE, p-values), as well as standard regression analysis (ANOVA, LM, GLM, ideally also GLMMs, including standard extensions such as autoregressive and zero-inflated models)
Note that if you don’t fulfill the requirement 1, you may be bored, while not fulfilling 2 and 3 will make it very hard to impossible for you to follow the course.
Workshop 2: Mixed effects models in R

Short titel:

Mixed-effects models

Lecturers:

• Christoph Scherber, University of Münster, Heisenbergstraße 2, 48149 Münster, DE, 00 49 (0) 251 – 83 33996, christoph.scherber@uni-muenster.de

Duration:

2h

Date/Time:

Monday 09.09.2019, 10:00-12:00

Location:

Fürstenberghaus, F073

Costs:

20 €

Maximum number of participants:

25

Course content:

This two-hour workshop will provide an introduction to mixed effects models in R using packages such as nlme, MASS, lme4 and mgcv. Participants will learn how to specify models with these packages, how to create random effects structures and how to use variance functions, correlation structures and canonical link functions. In addition, participants will learn how to inspect model adequacy (plots of residuals and random effects), how to deal with overdispersion, how to create predictions from these models, and how to generate graphical output.

Requirements:

Participants should have basic knowledge of R. Please bring an own laptop.
Workshop 3: Tracing back biodiversity to the 19th century via text mining

Short titel:

Mining literature for biodiversity data

Lecturers:

• Christine Driller, Senckenberg Gesellschaft für Naturforschung, Senckenberganlage 25, 60325 Frankfurt am Main, DE, 0049 (0)69 75421906, christine.driller@senckenberg.de
• Sajawel Ahmed, Text Technology Lab, Goethe University Frankfurt, Robert-Mayer-Str. 10, 60325 Frankfurt am Main, DE, 0049 (0)69 79824661, sahmed@em.uni-frankfurt.de,
• Gerwin Kasperek, University Library Johann Christian Senckenberg, Bockenheimer Landstr. 134-138, 60325 Frankfurt am Main, DE, 0049 (0)69 79839365, g.kasperek@ub.uni-frankfurt.de

Duration:

3 h

Date/Time:

Tuesday 10.09.2019, 09:30-12:30

Location:

Fürstenberghaus, F072

Costs:

-

Maximum number of participants:

20

Course content:

Climate change and biodiversity loss are among the major challenges of our time. Investigating the causes and causal relationships are therefore the focus of current research and policy making. The extraction of biodiversity information from historical data sources becomes more and more relevant in this context in order to capture the baseline of the pre-industrial time. The Specialised Information Service Biodiversity Research (BIOfid) taps into this growing demand by improving the accessibility of legacy literature relevant to biodiversity research and by developing reusable text mining tools for data extraction. As part of the pilot project BIOfid particularly focusses on Central European literature on the distribution and ecology of vascular plants, birds, moths and butterflies. In framework of this workshop we want to share our technological developments and discuss researchers´ requirements and expectations for this type of information service. This workshop addresses scientists working in the field of biodiversity research and will focus on the following topics: 1) Introduction to the BIOfid web portal, i.e. accessing literature and extracting data from historical texts through a visual interface. 2) Use of state-of-the-art and easy-to-use Natural Language Processing (NLP) tools, e. g. deep learning of text content. 3) Developing customer workflows from source materials to processable texts and data output. The participants deal with challenges associated with data quality, natural language processing as well as information- and semantic relation extraction. They will automatically analyse large text corpora to extract data linked to established ontologies and knowledge bases. Participants are also invited to integrate their own texts into the analysis procedure provided by the authors.

Requirements:

Please bring an own laptop.