Expose

This page is dedicated at all MSc. students at the University of Liechtenstein who are currently working on a thesis with a strong focus in Computer Science, and who also have some interest in pursuing a career in research or academia (i.e., they aim to go for a PhD).

Goal of the Expose

In principle, the Expose (at UniLi) has one goal: show that the student will be able to make a good thesis. To this purpose, the Expose should demonstrate that the student:

Structure of the Expose

You can use the guidelines of UniLi for the generic structure of an Expose, and/or adhere to the LaTeX template (found at the bottom of this post, or directly on Overleaf). At a high-level, there are four “crucial” sections that allow to determine whether a student will (or will not) be able to do a good job:

Considerations on the Expose

It is impossible to know with so much advance how the thesis will turn out. Hence, while you should “do your best” to write a good introduction and to formulate the specific problem tackled by your thesis, chances are that things will go much different from expected during the thesis.

However, it is unlikely that the background / related work will change: unless you change your thesis completely, your topic will remain the same.

Hence, to maximize the use of your time, I recommend to spend a lot of effort in writing a good Background/Related Work section! Doing this has the following benefits:

Writing a good Section 2

The most common mistake that students (at all levels) make when writing Section 2 of (any) scientific paper is that they simply “summarize prior work”, sometimes by providing redundant information. This is not what Section 2 is for!

We can distinguish two parts of Section 2:

You should use Section 2 to your advantage: instead of merely “explaining what is (already) known”, you should use what is known to point out “what is not (yet) known” – which is what your thesis will make “known”!

What “not to do”

Here are some poor attempts at writing the Background part of Section 2:

Here are some poor attempts at writing the Related Work part of Section 2:

What “could be done”

It is recommended to enhance the “Background” section with schemas or illustrations [a]I recommend using Draw.io. Make them yourself – but if they are similar to those made by prior work do cite the original source.

For the “Related Work”, you can make a Table [b]A good example of this is Table 4 of my SpacePhish paper. that lists all the closest works, and explicitly states what they did “well” and what they did “not-so-well” in the context of the goal of your thesis. The intention is showing the reader that the state-of-the-art does not allow to answer your “research question” (because not a single work did “well” everything that is “needed to answer your research question”).

Of course, the Table mentioned above should not include every paper that has been written up until today! Focus on the most relevant ones. In general, it is good practice to have from 10 to 30 papers in such a table.

Generic Tips

The following are some generic tips that apply for “any” kind of scientific document (and an Exposé is definitely a precursor of one such document).

Fancy words

Try to create “tension”. Remember that people tend to be more attracted by “problems” than by “positive” (or “neutral”) statements.

Below is a list of words that naturally “create tension”, which you can use to build your argument and induce the reader into believing that “there is a problem, which your thesis will solve”:

Anomaly, inconsistency, but, however, nevertheless, problem, contrast, on the contrary, instead, while/although, surprisingly, unfortunately, unclear, confusing, vague, flawed, incomplete, unrealistic, poor, struggle, over/under-whelming, (over)exaggeration, simplistic, over/under-estimation, unless, limited, overlook, neglect.

Additional resources