°ÄÃÅÀúÊ·¿ª½±¼Ç¼

Outside of the Social Data Science (Q-Step) Programme, SODA organises two innovative undergraduate modules.

Outside of the Social Data Science (Q-Step) Programme, SODA organises two innovative undergraduate modules that are available to students across UCL’s Faculty of Social and Historical Sciences. One is an introductory module designed to introduce data science to students who have little or no background in mathematics or statistics. The other is an advanced module using python to explore techniques of machine learning.

Data Storytelling: Explaining the Social World

This module is for students who want to understand what data science is, and how it can be used to answer important questions about the social world. No knowledge of mathematics or statistics is required: the module introduces data science techniques in an accessible way, making it suitable for students from all disciplinary backgrounds. It integrates data analysis with a ‘data storytelling’ approach that focuses on communicating findings from data using creative visualisations and journalistic techniques. Students will leave the module empowered to apply, interpret and critically analyse statistical methods. They will become better researchers and more informed citizens, but also confident and effective communicators of the insights from data. The course content includes the basic tools of data science – describing and visualising data and uncovering patterns – as well as how to implement them through coding. This material is interwoven with instruction in a series of techniques to communicate the results of data analysis in compelling ways, both verbally and visually.

Machine Learning for Social Sciences with Python

This module offers a grounding in the Python programming language alongside an introduction to a range of machine-learning techniques. It comprises three parts: data preparation and wrangling for machine learning (including fundamental machine learning concepts); unsupervised machine learning techniques (e.g., K-means and hierarchical clustering, principle component analysis) and supervised machine learning techniques (e.g., linear and logistic regressions, tree-based models, support vector machines). The module is problem-focused, mimicking the kinds of analysis students are likely to undertake in dissertations and in the workplace. Students will leave the module empowered to apply, interpret and critically analyse machine learning techniques using Python and they will also be able to critically engage with the results from such analysis and recognise their added value.ÌýThe module is suitable for students from all disciplinary backgrounds, but students enrolling on the module will require knowledge of fundamental statistical concepts such as linear regression. Previous experience of using a programming language (such as R) is desirable but not essential.