Social Networks & Text Analysis
Coursework topic
Ø The project has to mainly use methods from network or text analysis
Ø Neural networks are not networks in the context of this module, so it is not enough to use a neural network for the project to be in scope (but you can use neural networks as part of the analysis, for instance after extracting features from some text, you can then use a neural network with those features as input)
Objectives
Ø Independent learning
Ø Expand on what was presented in lectures
Ø Find and read relevant papers
Ø Learn and practice how to work on a network or text analysis problem
Ø Learn and practice how to present your results scientifically in a report
Key questions
Ø What is the problem you are studying?
Ø Why is it interesting?
Ø What is your proposed solution/approach?
Ø What are your results?
Deliverables: pitch deck project presentation
Ø A pre-recorded presentation outlining your mini-project proposal
Ø No more than 5 minutes!
Ø You can use any software you like to record it (Zoom, Teams, …)
Ø To submit, upload a PDF containing a (private) link to the presentation on Teams, Youtube, OneDrive, …
Ø What is the research question and why is it interesting?
Ø How has this question been studied in the literature?
Ø How will you try to solve/study it?
Ø What data will you use?
Ø What methods will you use and how?
Ø Think about slides design: is there too much text? Are the figures clearly legible?
Ø Think about your style. are you speaking too fast or too slow?
Ø Think about the structure: does the presentation have a narrative?
Deliverables: mini-project report and code
Ø 3 page report: figures are included in the 3 page limit, references are not included so can go on a fourth page
Ø The code used for the project (Jupyter notebook/Python script. & PDF version)
Ø The data (or a link to it)
Ø ALL IN A ZIPPED/COMPRESSED FOLDER
Ø Think about how you write: is it clear, concise and understandable? Remember that we don’t know what you are thinking so you need to write it
Ø Think about figures: do they support what you wrote? Are they legible and high resolution?
Ø Think about the structure: is it clear? Does it follow a structure? Did you provide all the relevant details?
Ø Problem statement
Ø Related work
Ø Your solution
Ø Results
Report
Problem statement
Clearly define what is the problem you were studying. Explain the problem and its importance. Include a mathematical formulation if possible. Keep in mind that the description should be readable by anyone without specialized knowledge.
Related work
Mention any relevant papers and state what problem they are solving and the general approach. This should be really short for each paper, just a couple of sentences. You do not need to explain their works in detail. Explain how your works relates to these. Is it better in some way? Does it solve a different problem from all of them? Why is your work important compared to these?
Your solution
Describe in details your idea for solving the problem. Show why it is different from previous work where possible. Explain why the problem is challenging, explain what your ideas are and justify them. Time may not permit you do everything, in which case explain what you are presenting here and what can be done later.
Results
Describe your results: algorithms, proofs, general arguments and analysis. Give plots, tables that show your results. Discuss them. You need to explain the results and their importance to us. We were not there during the work, so we cannot understand things that you do not explain. Discuss what else can be done on this topic etc.
Report structure – a suggestion
Ø Introduction: context and motivation
Ø Description of methods and data
Ø Discussion about experiments
Ø Conclusions
The code
Ø A Jupyter notebook (ideally) or a Python script.
Ø Readable and well commented
Ø Provide both the code and a PDF version of the notebook (if using)
The data set
Ø If possible, submit the data in the zipped submission folder
Ø Otherwise, a link to it in a document you submit