The following ECIR 2019 tutorials will take place on April 14, 2019:
Conducting Laboratory Experiments Properly with Statistical Tools: An Easy Hands-on Tutorial
This hands-on half-day tutorial consists of two sessions.
Part I covers the following topics:
- Paired and two-sample t-tests
- confidence intervals
- One-way ANOVA and two-way ANOVA without replication
- Familiwise error rate
Part II covers the following topics:
- Tukey's HSD test, simultaneous confidence intervals
- Randomisation test and randomised Tukey HSD test
- What's wrong with statistical significance tests?
- Effect sizes, statistical power
- Topic set size design and power analysis
- Summary: how to report your results
Text Categorization with Style
Text categorization problems (e.g., automatic indexing, filtering ) are recurrent in IR. This tutorial focuses on the text style to provide answers to various questions (authorship attribution, author profiling (determining author gender, age or psychological traits) or verification, etc.). First, the best stylistic markers must be determined according to the underlying problem. Second, when training data is available, a machine learning model can be applied. Otherwise, the decision must be taken without a learning stage (e.g., author clustering). Third, an evaluation procedure must be defined to measure the effectiveness of different approaches. Based on the R stylo package, some hand-on applications (e.g., US history, literature) will be presented.
PLS-based Structural Equation Modeling: A Hands-On Tutorial on Basics, Applications and Current Developments
Markus Kattenbeck and David Elsweiler, University of Regensburg
This full-day tutorial introduces participants to Structural Equation Models (SEMs) and their estimation using Partial Least Squares Path Modeling (PLS Path Modeling). Structural Equation Models are a statistical technique to simultaneously assess the relationships between unobserved factors (i.e. latent variables) and the way these factors are measured based on observable variables. This is useful, for example, for researchers in our community interested in learning about how different factors influence human information behavior and as a means to validate and optimize measurement instruments for subjective concepts important in IIR, such as engagement, satisfaction, trust etc.
The tutorial is introductory in nature. Audience members need only have basic statistical knowledges. Familiarity with GNU R is essential, as participants will be required to run their analyses using this software on their own computers.
Concept to Code: Neural Networks for Sequence Learning
Omprakash Sonie, Muthusamy Chelliah and Bidyut Kr. Patra
Deep Learning has shown significant results. Advances in deep learning are applied to many aspects of modern IR systems. In this tutorial we provide conceptual understanding of Embedding methods and Recurrent Neural Networks (RNNs) which are currently applied in IR. As RNNs are effective in modeling sequential data (e.g. clicks, add to cart, purchase data) that is generated by users in a session and across sessions we provide a hands-on case study for sequence aware Recommender System.
Proposed tutorial begins with learning of Embedding (e.g. user, product, product features), traditional sequence-based and session-aware recommendation systems and their evaluation methods. It then covers various models based on hierarchical representation and RNNs: attention & attribute aware, memory-based models, multi-layer LSTMs, and combining RNNs with CNNs.
We walk-through the code for techniques covered in each section and for a sequence-aware recommender system on e-commerce dataset, summarize these models, parameters and understand what is going on behind the scene with various visualizations. We will use Jupyter notebook with already executed code for walk through.
We believe that a self contained tutorial giving good conceptual understanding of deep learning techniques with sufficient mathematical background along with actual code will be of immense help to participants.