Category: Computer Science

Here is this weeks ariticle: https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2594754Overview

These are 450-500-word “essays” on the reading for that week. The point of the critiques, as the name suggests, is to write your opinion on the reading of the week. Please do not summarize the article. Thats not the point. You have to take a position and say something about the article.

Due (most Mondays)

The critiques are due by 11am on Brightspace. Writing clarifies thought and will help you be better prepared to discuss the topic of the week.

Grading Rubric

The weekly critiques will be graded on the following rubric:

5 points (max points) Your critique takes a position on the article, supports it well, and is well written

4 points Your position on the article is clearly stated and supported

3 points – Your position on the article is NOT clearly supported

2 points Neither your position is clear nor is it well supported

1 point Your just summarize the article/paper OR repeat the contents of the paper without analysis

0 point You do not submit your critique or if your file does not open

All critiques must use examples in their critiques. If you are using more than one examples in your critique, only one can come from the material you are currently critiquing. All external examples used have to be clearly cited. You will lose :

• (-3) point if external examples are not fully cited (including name of article, names of authors, publication venue, year of publication)
• (-2) point if no examples are used in a critique
• (-1) point if more than one examples are from the material being critiqued

Obviously, your score will have a lower bound of ZERO (0). For example, so if you wrote a poor critique that would have received a 2/5 and you also did not cite your external examples properly you will have 3 pointed deduced from 2/5 and get a zero.

First critique due Monday, February 9 by 12pm.

Using of ChatGPT for writing critiques

You may experiment with ChatGPT for your critiques. We will talk about ChatGPT in this class in the middle of the semester. However, should you want to take the help of ChatGPT to write your critique please follow the following guidelines:

1) For any part of the critique that you used ChatGPT, you need to provide a footnote (see example below) in your submission describing how you used ChatGPT or another AI assistant and whether it was helpful or not.

2) Please be alert to the fact that ChatGPT might seem to produce authoritative text, there is nothing at the backend of this technology (other than you) to fact check the results.

Example footnote text (please use something like this):

This paragraph was generated by ChatGPT in response to the prompt <give the exact prompt you used >. I used the text verbatim as it sufficiently explained <describe the topic for which you were generating text>. I used ChatGPT here because <give your reasons here>. I found ChatGPT to be <describe your experience of using ChatGPT here>

Be honest, you will not be penalized for the reasons you give here. But if you forget to include the footnote after using ChatGPT (and basically claim that you wrote everything) and we find out. You will be fail the course immediately.

Example of a (good) and (not so good) critique

Here is an example of what a good critique (5 points) and not so good critique (1 point) looks like.

I would recommend that you read the following paper first to get an idea of what the critiques are saying.

by Eugene Spafford

Here are the two example critiques.[Critique_Example.pdf]. It includes my annotations and notes. Please read carefully.

A. Purpose of the Thesis

• The development of an innovative interview simulator specifically designed for software developer candidates.
• The simulator will be implemented as an Android application and will leverage Large Language Models (LLMs) for the dynamic generation of interview questions, adaptive difficulty adjustment, and the provision of personalised feedback.

The main objective is to investigate whether the use of a mobile-based interview simulator can improve candidates performance in realistic mock technical interviews and increase their confidence towards the interview process.

B. Research Question

Can an interview simulator for software developers, implemented as an Android application and based on LLMs, improve candidates performance and confidence in real interview scenarios?

Secondary research questions:

• How do users perceive the realism of the simulator?
• Which categories of questions are considered more helpful (technical or behavioural)?

C. Methodology

• Design and development of the Android application.
• Experimental procedure involving participants.
• Data analysis (statistical pre-/post-comparison, qualitative analysis).

D. Expected Contribution

• Improved understanding of the impact of LLM-based interview simulators.
• Provision of a preparation tool for students facing technical interviews.
• Investigation of the feasibility of implementing such tools on mobile devices.

E. Deliverables

• A functional Android application.
• Documentation and an architecture report.
• A pre-/post-experimental study.
• The Masters thesis.