Prerequisites: None
Instructors: Prof. Sakshi Bhatia, Prof. Samar Husain, Prof. Apurva
This track is suitable for beginner-level linguistics students who want to do experimental research. It would contain lectures and practice sessions on Statistical data analysis using R and experiment design methods.
The structure and curriculum for this track will be as follows. Two themes will be covered in the five-day workshop: (i) Statistical data analysis using R, and (ii) Experiment design.
Theme 1: Statistical data analysis using R (15 hours)
R programming I: Variables, Loops, Conditionals
R programming II: Plotting and file handling
Statistics: Significance testing, t-test
Statistics: Linear regression
Statistics: Linear mixed models
Theme 2: Experiment design (15 hours)
Basics of experiment design I
Basics of experiment design II
Experiment building with PCibex
Power analysis and sample size determination
Managing experimental data
Prerequisites: R/python programming (e.g., you should be able write loops, conditional statements and functions in R or Python)
Instructors: Prof. Himanshu Yadav, Sovan, Nitesh
This track is suitable for students who specifically want to learn computational methods for their research, which is an important avenue in today’s research scenario. This track would require that a student already knows R/python programming. It will consist of two themes: (i) Bayesian modeling, and (ii) Corpus-based empirical linguistics, where the focus would be on empirical research using large-scale language corpora.
Theme 1: Bayesian modeling (18 hours)
Random variables
Statistical inference using Bayes’ theorem
Parameter estimation
Bayesian regression modeling
Model comparison
Bayesian hierarchical models
Theme 2: Corpus-based empirical linguistics (12 hours)
Querying the treebanks
Measuring linguistic complexity of trees
Random baseline generation
Artificial vs natural language experiments