BSc/MSc Dissertation Projects
 
Here is a list of BSc dissertation and MSc summer term dissertation project ideas that require some further thoughts and discussions. Some of these are more research oriented (focusing on building a test bed for scientific investigations, evaluated via experimentation) and others are more software engineering oriented (focusing on software development, evaluated via software and user tests). All projects are open to modification and negotiation, if you think you have a good idea to make them more interesting.
 
 
Research Oriented Project Ideas
 
  • Applying machine-learning and data visualisation methods for analysing JASSS publications
  • Prototyping ideas for SimPy/AgentPy/Mesa extentions (good programming skills in Python required)
  • Testing different ways of knowledge representation in the Hybrid Climate Assessment Model (see here for more)
  • Testing the usefulness of AI decision modelling through "utility theory" and "fuzzy logic" (see here for more)
    • Application example: Optimising software development team compositions
  • Energy-efficient data forwarding in mobile wireless sensor networks
    • Mixing behavioural and software agent modelling + optimisation
    • Application example: Simulating the use of ad-hoc sensors to track elephant movement
  • Benchmarking work efficiency for software maintenance
  • Smart transportation/smart roads/smart cities
    • Using data science methods to improve the "Smarter Travel - Nottingham" platform (see here for frontend)
    • Novel modelling/simulation approaches for exploring smart transportation/smart roads/smart cities ideas (e.g. modelling knowledge as an abstract agent)
  • Exploring opportunities for simulating manual assembly lines more effectively with the help of artificial intelligence (see here for more)
  • Generating synthetic populations for simulation models
 
Software Engineering Oriented Project Ideas
 
  • Developing a 2D simulation game for teaching Systems Thinking (good software development skills required) (see here for more)
  • Creating a web presence (frontend + backend) for the Reporting Standard for Rigour and Transparency for Data Use in Agent-Based Modelling (good web development skills required) [paper link]
  • Maintaining and extending the Hybrid Climate Assessment Model (see here for more)
    • Generating synthetic populations for simulation models
  • Maintaining and extending the UoN ABM Toolkit
    • Add editor for state machine diagrams (drawing; error check; translation into code)
    • Add process flow elements to allow hybrid DES/ABS modelling
    • Add an optimisation engine (e.g. HeuristicLab)
  • Developing an uncommon (card)game with virtual (AI driven) players which can help the elderly to have a greater awareness of their own energy consumption (previous implementations include an adapted version of Yaniv)
  • Improving "code2uml", a tool for generating UML class diagrams from java .class/.jar files (see here for more)
  • Creating a multi-agent system basketball emulator in Java/Python (inspired by Buckland (2005) Chapter 4)
  • Providing additional statistics for Google Scholar profiles
  • Smart transportation/smart roads/smart cities
    • Adding more functionality to the "Smarter Travel - Nottingham" platform (see here for frontend)
 
Back to Top

 
Full Year MSc Dissertation Projects
 
Project 1: IDE for an Agent-Based Modelling Library
The task is to create a component base drag-and-drop IDE for developing state machine diagrams. These then need to be translated into a script that can be read by an agent-based simulation library other dissertation students have developed in the past. The library is written in Java. There is also some maintenance work on the library to do. If you want to take on this project, you need to be proficient in Software Engineering and Java. There is an alternative project for Python programmers, so if you are interested in the topic let me know , and we can discuss the Python variant. To find out more about agent-based modelling and simulation and the related state machine diagrams, have a look at https://arxiv.org/pdf/0803.3905/.
 
{taken} Project 2: Improving the 'Smarter Travel - Nottingham' Platform
In collaboration with Nottingham City Council we have developed the 'Smarter Travel - Nottingham' platform (http://51.140.251.121/), a website to provide smart travel information. We want to add some new widgets and update the ones that are not working anymore. The project uses VertX (https://vertx.io/) for the backend, a toolkit for building responsive, resilient, and scalable JVM applications, and Bootstrap 4 for the frontend. If you want to take on this project, you need to be proficient in Java, and have some experience with web development. Knowledge in data science (predictive modelling) would be a bonus.
 
Project 3: Translating Causal Loop Diagrams into System Dynamics Diagrams
Causal loop diagrams are qualitative tools used to illustrate and understand the relationships and feedback loops within complex systems without requiring numerical data. System dynamics diagrams, on the other hand, often incorporate quantitative elements, modelling system behaviour over time using stocks, flows, and differential equations to simulate and analyze system dynamics. In this project we want to investigate, if we can use AI tools to translate qualitative diagrams into quantitative ones. If time allows, we could also look at using generative AI for creating causal loop diagrams in the first place. This projects requires someone proficient in Applied Data Science.
 
{taken} Project 4: From Qualitative to Quantitative Data with the Help of AI
Social systems can be described as complex network of people, groups, and institutions that interact with each other according to established norms, values, and behaviours; encompasses families, communities, governments, and even social media platforms. One method to explore the complex dynamics of such systems is to use Agent-Based Modelling and Simulation (ABMS). The data that is needed to support ABMS comes in two formats: qualitative and quantitative. While quantitative data can be used straight out of the box, qualitative data needs to undergo some procedures to be transformed into quantitative data that can be used for ABMS. In this project we are interested in optimising the process of getting from qualitative to quantitative data in the context of ABMS. Your task will be to explore different Machine Learning and Generative AI techniques for doing this job in a semi-automated way. This projects requires someone proficient in Applied Data Science. More information about ABM can be found here.
 
Back to Top

 
Advice for Dissertation Students
 
  • Technical Skills for Thriving in Graduate Studies (YouTube Video Series) [link]
  • David Taylor - Scientific Writing and Reading Skills (YouTube Channel) [link]
    • How to Write a Literature Review in 30 Minutes or Less [link]
    • How to Read a Journal Article in 10 Minutes or Less [link]
    • How to Write a Research Proposal - Undergrad, Masters, Doctoral [link]
    • Giving an Academic Talk (targeted at conference presentations, but but most of this also applies to dissertation presentations) [link]
  • Miscellaneous
    • Academia Stackexchange: Is there a convention on how to write pseudocode in scientific papers? [link]
  • Internal Moodle links (only accessble for students signed up for the module)
    • BSc Project 22/23: Moodle Page [link]
    • BSc Project 22/23: Project Handbook [link]
    • Enhanced MSc Project 22/23: Moodle Page [link]
    • Enhanced MSc Project 22/23: Student Handbook [link]
    • Enhanced MSc Project 22/23: Supervisor Handbook [link]
    • MSc Project 22/23: Moodle Page [link]
 
Back to Top

This site uses cookies to anonymously measure how people use it!