Analyzing-Social-Media-Data-for-Social-Science

slides https://docs.google.com/presentation/d/111MeIrf5WzlFxnRDhxTD6UuAz3kCEXXrKQhGHL6A4Is/edit?usp=sharing

email: lucia.chen AT stanford.edu

Twitter/Youtube: ML_made_simple

Welcome to this tutorial!

If you are from a social science background and want to learn programming and data science step by step, this is the right place :)  Previous programming knowledge is not required.

Prerequisites

Please make sure to install Python, Jupyter notebook, Anaconda (optional) and the required packages before you come to the workshop.

Setting up your environment

Packages required:

• NLTK (NLP)
• pandas (process tabular data)
• emoji
• matplotlib (for making plots)
• wordcloud
• jieba (for Chinese word segmentation)

Contents

Workshop 1

clean text data, extract summary statistics from text and build a word cloud by yourself. This workshop will consist of the following four modules:

• Module 1 – Common Data Structures in Python
• Module 2 – Preprocessing Text Data
• Module 3 – Building a Word Cloud
• Module 4 – Summary Statistics with Pandas