Deep learning is particularly well suited when working with unstructured and complex data types. Neural network architectures are extremely flexible and can be adapted to different types of problems, which in part explains their popularity today.
Your task is to build and use a deep learning architecture to explore a problem setting (optimnally) within your main area of study. But you are welcome to work on other issues, too. If you work in a cross-disciplinary team, you are welcome to work on an appropriate cross-disciplinary problem statement.
You are asked to hand in two different report formats, namely:
The notebook targets a machine-learning literate audience. Here you can go deeper into the technical details and method considerations. Provide thorough documentation of the whole process, the used methods. Describe the intuition behind the selected and used methods, justify choices made, and interpret results (e.g. Why scaling? Why splitting the data? Why certain tabulations and visualizations? What can be seen from … ?, How did you select a particular algorithm? Why did you scale features in one way or another?).
Please provide the notebook as a PDF ot HTML (Knittered from rmd or converted ipynb, when HTML zipped), optionally with a public link to a functional Colab version
The stakeholder report (simple PDF or HTML, no code) summarises the analysis for a non-technical audience. Here you don’t need to discuss alternative approaches to standardization and alike. Instead, you should try to explain the analysis and results, emphasizing its meaning and interpretation. Imagine it as a report of the project outcome, as you would explain it to a general audience.
Aim at a length of not more than 5 pages, including tables & visualizations.
Consider drawing from different data sources, incorporating various data types. For instance, one could(but does not have to) mix up numerical, text, relational and image data to classify users of an online social network. Reach out if you have questions about that.
Consider looking into research papers from conferences such as Social Informatics (Socinfo) and Computational Social Science. Some of the papers will link to GitHub repositories with data and code.
If you plan to work with deep learning as your semester project, you are welcome to use the M3 group assignment as a warm-up.