We offer several workshops on data science methods during the Data Science Week@Waseda. Please note that seats are limited and sign-up is necessary to attend the workshops. These workshops are free and open to all, but priorities will be given to current Waseda students. These workshops are made possible by the funding provided by the Top Global University Project, Waseda University, Center for Positive/Empirical Analysis of Political Economy (https://www.waseda.jp/eape/en). Please note that the registration period has ended (update: 1/24).

Workshop 1: Web Scraping with R (1.5 hr)

Monday, January 28, 2019 13:00-14:30; 1st conference room, 10th floor, Building 3, Waseda University

By Phillip Law, Postdoctoral Fellow, Monash Biomedicine Discovery Institute, Monash University

This workshop is aimed at students and researchers who want to learn more about using R to extract data from the internet, with a specific focus on social media platforms. Although several other languages are more commonly used for harvesting data from the internet, using R for web scraping is a seamless research process that can be achieved within a single framework. By the end of the workshop, you should have a basic understanding of core techniques to conduct simple web scraping on various social media platforms (e.g., Twitter) for your research. Some knowledge of R will help participants in this workshop, but it is not necessary for full involvement. Please bring your own laptop with R and RStudio installed.

[Enrollment limit: 20]

Workshop 2: Co-occurrence Networks as a Method for Content Analysis (3 hr)

Tuesday, January 29, 2019 14:45-18:00; 1st conference room, 10th floor, Building 3, Waseda University

By Elad Segev, Associate Professor, the Department of Communication, Tel Aviv University

This workshop introduces the basic concepts associated with network analysis. Participants will learn to use Visone, software tool to conduct their own network analyses. As part of the workshop participants will develop their own networks of frequent word co-occurances to identify the biases and framing in different texts.Participants are requested to arrive with labtops and have Visone running. Visone is open-source software (https://visone.info/) developed by computer scientists at the University of Konstanz. It requires Java to be installed on the computers.

[Enrollment limit: 15]

Workshop 3: Sentiment Analysis with Python (3 hr)

Wednesday, January 30, 2019 14:45-18:00; 1st conference room, 10th floor, Building 3, Waseda University

By Robert A. Fahey, Ph.D. Candidate, the Graduate School of Political Science, Waseda University

Sentiment Analysis is a group of methods which aim to answer a deceptively simple question; how do the authors of a set of texts feel about a given subject? These techniques are commonly used in fields like public opinion and market research to understand whether social network users are positive or negative about a political actor, a product or a brand. In this workshop, we’ll look at a variety of different approaches to sentiment analysis, assess their strengths and weaknesses, and try them out on real-world text samples. This will prepare attendees to select and implement effective sentiment analysis approaches in their own research projects. The workshop will be conducted in Python, a powerful and very popular language for data science; no previous knowledge of Python is required to participate but a little experience with programming may be beneficial. Attendees are required to bring their own laptop computer, and are advised to install Python (the Anaconda all-in-one package – https://www.anaconda.com/download/ – or any other distribution with which they are familiar) ahead of the workshop.

[Enrollment limit: 20]

Workshop 4: Quantitative Text Analysis using R (6 hr)

Thursday, January 31 and Friday, February 1, 2019, 13:00-16:15; Room 1102, 11th floor, Building 26, Waseda University

By Kohei Watanabe, Assistant Professor, Waseda Institute of Advanced Studies (WIAS), Waseda University

This workshop introduces you to quantitative text analysis using R. By participating in this workshop, you will be able to study a large body of textual data, such as books, newspapers, advertisements, and transcripts, systematically to discover interesting patterns from the social science and humanity perspective.

Among other R packages for quantitative text analysis, Quanteda (https://quanteda.io) will be used throughout as a package popular among political scientists in this workshop. The instructor will guide you through the Quanteda Tutorials (https://tutorials.quanteda.io) from “Introduction” to “Statistical Analysis”.

Participants are required to have experience in R and basic knowledge of statistical analysis, but programing skill is not necessary to understand the workshop material. Participants must bring their laptop computers to the workshop with R (version 3.4 or newer) and RStudio installed.

[Enrollment limit: 20]