The workshops take place on the pre-conference day, the 13th of December, in two parallel sessions, morning (10-13) and afternoon (14-17). You can choose to attend one or both sessions (if registering for one of them, choose "none" for the other). Workshop participants are requested to bring their own laptops.
The workshops (Dec 13) are free to attend, pending sign-up through this form. To attend the main conference (Dec 14-16) a separate conference registration is required (cf.
https://cudan.tlu.ee/conference). Due to popular demand, please note that we can only provide visa support letters to main conference accepted speakers who have undergone peer review.
The following workshops are being offered:
Computer vision techniques for analyzing cultural data with Python (Ksenia Mukhina). In this workshop, participants will learn how to use computer vision and deep learning to analyze cultural data. No prior data analysis experience is required. We will start with a brief introduction that explains the theory by examining scenarios involving kittens, puppies, and the world of social media filters. These examples serve as key building blocks to understanding how computer vision techniques actually work behind the black box. The main part of the workshop is dedicated to hands-on experience. Participants will use the Pytorch framework on a real-life image dataset from Flickr to explore diverse computer vision approaches for studying cultural data. All the necessary code and datasets will be provided, ensuring a smooth and accessible learning experience. Participants are encouraged to bring their laptops to the workshop. There is no need to install anything in advance.
Visualizing big cultural data using R, with a little help from AI (Andres Karjus). This is an intro-level workshop to R, a powerful programming language and toolkit for data analytics. The workshop is hands-on from the start, focusing on using R to visualize and explore data. We will also look into how to use AI tools like ChatGPT and Bing to help learn programming, write code and debug errors. No prior programming skills are required (for those who might know other languages and pick this up faster, we got extra exercises to keep you entertained). Installation instructions will be sent in advance via email, and installing the (free) software is a prerequisite to participate (don't worry, there will be step by step instructions, it's easy).
Navigating Visual Collections using Image Embeddings (Mar Canet Sola). This practical workshop delves into image embeddings, and the Collection Space Navigator (CSN) tool developed by CUDAN researchers. The session is tailored to acquaint participants with the fundamentals of image embeddings, providing a launchpad for navigating vast visual datasets effortlessly. The workshop will begin with a brief introduction to various visual embeddings, followed by a step-by-step demonstration on preparing embeddings of visual datasets. Some time is dedicated to discussing the features and capabilities of the CSN tool, equipping participants with the know-how to harness its potential use in their research projects. Participants are encouraged to bring their own visual datasets (image collections and metadata). Utilizing a subset might be more manageable for the scope of this workshop if the data is large. For those without personal datasets, we will provision data to engage in the hands-on exercises. Given the novelty of the CSN tool, we value your feedback immensely and would appreciate participants completing a brief survey regarding their experience with the tool. Participants should bring their laptops and download the open-source CSN tool (https://github.com/Collection-Space-Navigator/CSN) ahead of the workshop. Should any issues arise, or for any inquiries, feel free to contact mar.canet@tlu.ee. No prior data analysis background is required!
Tableau for exploring, analysing and visualising cultural data (Vejune Zemaityte) This hands-on workshop aims to introduce participants to the Tableau software for data analysis and visualisation. The session will begin with a brief overview of the software's different functionalities and a demonstration of how to access them. Unlike other analytics tools like R, Tableau is easy to pick up because of its click-and-drag features, which means users do not need to learn a programming language or write code to interact with their data. The majority of the workshop will, therefore, be spent by participants using Tableau to analyse and visualise data. Everyone is encouraged to bring their datasets of interest, but we will provide data for those who don't. The workshop will conclude with a round table for sharing our achievements with the group and discussing our experiences using the software. Participants are invited to bring their laptops and download a free trial version of Tableau Desktop via the link (https://www.tableau.com/products/trial) before the workshop. In case of any issues, feel free to email me at vejune.zemaityte@tlu.ee. No prior data analysis background is required!
