PATS2
You are not logged in
Time stamp: 12:54:31-23/9/2020

[Login]

2012 | 2013 | 2014 | 2015 | 2016 | 2017 | 2018 | 2019 | 2020
[PDF/Print]

Projects from 2020

StudentTitleSupervisorModeratorModule
Alice EdwardsIdentifying the best machine learning model for web based attack.Amir JavedLiam Turner1920-CM3203
In the project, you will be required to create a vulnerable web application (OWASP Broken Web Applications Project) and carry out various attacks on the web app and intercept the network traffic to the server. Based on the traffic you will intercept you will build a machine classifier to detect attacks.[more]
Amna Mohd A HameedA machine learning approach for forensic memory investigationMichael DaleyLuis Espinosa-Anke1920-CM3203
"The aim of the project to create a tool based on machine learning for forensic memory investigation. Where a machine learning model will be developed to automate assistance in identifying areas in memory dump for forensic analysis. "[more]
Amy ShieldsA personal diary appAlia I AbdelmotyStuart M Allen1920-CM3203
A lot of people may be interested in a simple way of keeping a diary of events in their day- whom they met, they talked to, where they went, how they felt. Some of this information can be extracted automatically from their location traces on their phones and from their activity on social web sites. You will build an app that will automatically track and record some of these personal information and present it to the user in the form of a browsable diary. You will be able to build upon/make u...[more]
Arron SehmbyDeep Learning Guitar TuningsDave MarshallPaul L Rosin1920-CM3203
Guitars can be tuned to a a variety of tunings. This is especially common in certain genres of music, notably folk music and heavy metal, for example. The basic idea here is to take recordings on music played in known guitar tunings. I have database of known songs in know tunings by a variety of artists and have the associated recordings. A deep learning system can then be trained with input samples (many short snippets from every recording) and trained to classify the tunings. New music can t...[more]
Artem ProtasavytskySerious games to teach children statisticsDr Daniel J. FinneganYuhua Li1920-CM3203
This project aims to develop a serious game for teaching young children (primary school level) basic statistics. The project involves the conception, design, development, and evaluation of a game to teach young children summary statistics e.g. mean, mode, median etc. You should contact me to discuss the project further with me as details need to be determined soon after selecting the project.[more]
Bedoor BahassanMalware Detection from IoT Devices Using Machine Learning AlgorithmsAmir JavedDr Daniel J. Finnegan1920-CM3203
This project, will look at the Network Traffic from IoT devices, pre-process the data, and implement Supervised Machine Learning models to Classify Malicious and Benign Network Traffic. After that, the evaluation process will take place to illustrate the results and to decide the best Machine Learning Algorithm.[more]
Benjamin RantEducation Recommendation ToolPadraig CorcoranYuhua Li1920-CM3203
One of the greatest challenges when learning a new subject is having the background knowledge necessary to learn that subject. For example, before learning about data structures one must first know about computer programming. However knowing what knowledge is necessary before learning a new subject is challenging. In this project you will develop a recommendation tool which recommends subjects to study before one begins to study a new subject.[more]
Benjamin ThorntonIOT Data Trading User Study PlatformCharith PereraPadraig Corcoran1920-CM3203
This project aims to develop a platform to gather with data and analyse data to understand how consumers may be willing to trade their personal data for different types of rewards in the future. This project requires strong web app development skills[more]
Brendan LambMusical Harmonization via Machine LearningDave MarshallMatthias Treder1920-CM3203
I would like to work on a project that combines my passions of music and Computer Science. Given an input of a MIDI file, I would like to create a machine learning algorithm to locate chords and a key-center within the given music, and then to substitute those chords with different, more interesting choices. The goal of this is to create a 'JazzBot' which could create complex harmonies from a given melody.[more]
Callum HaineSentiment analysisJose Camacho ColladosKirill Sidorov1920-CM3203
Automatically analyzing the sentiment has been a long-standing task in Natural Language Processing (NLP). For example, given a movie release, we may be interested in knowing whether the public has liked it or not. This project will consist of developing a sentiment analysis classifier which can be robust across different domains (e.g. social media or movie reviews). The main language of study would be English, although other languages are also possible. Good Python programming skills is required...[more]
Callum HughesIdiom search engineIrena SpasicMartin Caminada1920-CM3203
Idioms are phrases (i.e. groups of words) whose meaning may not be deducible from those of the individual words (e.g. "bury the hatchet" = "end a quarrel or conflict and become friendly"). One difficulty in finding idioms in text is the fact that they can vary, so a simple string search is not appropriate. For example, searching for "bury the hatchet" would miss this idiom in the following two sentences: Christmas looks to be a time for burying the hatchet or exhuming it for re-examination. Fr...[more]
Cassie MageeTackling low numbers of females in IT.Matthew J W MorganPadraig Corcoran1920-CM3203
This project aims to explore the reasons why women do not study IT and build a system to tackle what I believe is one of the main causes. The goal of the project is to provide an online platform that can be used as a tool to test the user’s abilities against the core skill that is required for the role of a computer scientist. As a result, a technical minimum viable product was completed where the user can submit their answers in a form and receive their results. The project also includes an i...[more]
Chun Kuan LimImplementation of a data privacy protection method for transaction dataJianhua ShaoNeetesh Saxena1920-CM3203
As an increased amount of transaction data being gathered and stored, how to protect the private information contained within such data sets becomes an important issue. This project aims to implement one of the exisiting algorithms for anonymizing transaction data. The system is expected to be developed in Python.[more]
Cyrus DobbsExploration of AI effectiveness within the game of CastleFrank C LangbeinDavid J Humphreys1920-CM3203
The aim of this project: • Implement a user interface and game engine to run this game o I will produce a GUI using JavaFX as an interface for human players to interact with the game. The game engine will also have an interface for AI players to interact with. • Add artificial intelligence o Information Set Monte-Carlo Tree Search (ISMCTS) player. • Add networking capabilities to the game o An authoritative...[more]
Daniel HoworthDesigning an Interactive Web Application to Enhance Weight LossNervo Verdezoto DiasMatthew Moloughney1920-CM3203
Obesity is a major health problem facing the UK, with 28.7% of adults in England identified as obese in 2019. In this day and age, an increasing number of people are turning to apps to help them with their health problems, including obesity. There are a plethora of health and fitness apps on the market designed to help users lose weight. This project will review a range of weight loss apps currently available in order to identify areas for improvement. This will then inform the design and deve...[more]
Dervla O'BrienOrdering Pizza using Internet of ThingsCharith PereraMatthias Treder1920-CM3203
Children worldwide have poor diets and poor understanding of nutrition, confining food into dichotomous “healthy” and “not-healthy” groups, rather than grasping deeper, more realistic concepts of variety and balance. This project explores how an interactive food ordering system could be designed to help enable children learn to make healthier dietary choices for themselves. The paper extends a previous study on the PizzaBox system and explores how it can be applied to an audience of 9-1...[more]
Elise BaileyComparing Interpretability Metrics for Diagnostic Classification of Neuropsychiatric Disorders Using Clinical MRI DataMatthias TrederStuart M Allen1920-CM3203
This project will be jointly supervised with the University of Stellenbosch, South Africa. They collected a brain imaging dataset involving Parkinson's, PTSD, and schizophrenia patients. The data have been preprocessed and parameters describing the brain structure have been extracted. This is a data science project, that is, you will use Sci-kit learn and statistics in Python to analyse, visualise and interpret the data. Excellent Python skills are vital, prior knowledge of machine learning is ...[more]
Ellie RobinsRecognising and disambiguating place names in text documentsChris B JonesSurya Thottam Valappil1920-CM3203
In order to index documents with regard to geographic space it is necessary to recognise and geocode (i.e attach coordinates to) place names in the textual content. This information retrieval project is concerned with developing machine learning methods to perform this task. A gazetteer consisting of a list of place names will be used to assist in the process. The task can be challenging because of the difficulty of distinguishing genuine place names from other terms, such as the names of peop...[more]
Fani NonchevaUsing sentiment analysis to improve emoji predictions in tweetsLuis Espinosa-AnkeMatthias Treder1920-CM3203
This project is focused around using sentiment analysis to improve the task of emoji prediction in tweets. Given a string of text, the machine learning model must learn to find patterns that would yield different predictions as to what emoji would follow. Generally, people use different emojis to express similar emotions, so to improve emoji predictions in tweets would be to observe the differences in the way people express themselves via text and how it affects the emojis they use. To keep the ...[more]
Flynn EllisAutomatic Generation of Pixel ArtYukun LaiFrank C Langbein1920-CM3203
Pixel art is a form digital art where characters are represented using a small number of pixels. It was popular for video gaming, and still useful nowadays for various applications ranging from icons to animation. However, traditional approaches of pixel art generation are very time-consuming and require expertise. Recent work considers automatic generation of pixel art. The aim of the project is to develop a prototype system that achieves automatic generation of pixel art by implementing some r...[more]
Fraser OrrSecure Smart Grid InfrastructureNeetesh SaxenaHantao Liu1920-CM3203
The aim of this project is to design and create prototype for a cyber-secure communication network that ensures the system level security under a scenario where several devices (clients) are connected to a server and exchange critical information. The security solution will demonstrate (through a prototype) the working of network access control, scalable key management, and end-to-end cryptographic message authentication.[more]
George CloseMajel: Voice Assistant Style Interface For Command Line TerminalDave MarshallYipeng Qin1920-CM3203
A program that allows users to speak command line inputs aloud and have them executed by the system. This would be useful in particular for file system navigation; often user friendly directory and file names with spaces are difficult to access in command line. The project will also in part be an attempt to introduce a 'natural language' way of interacting with the terminal; the input of the speech "Execute My Script" is much more intuitive and descriptive of what the user wants to happen compar...[more]
George KennyProcedural Voxel Planet GenerationFrank C LangbeinNatasha Edwards1920-CM3203
In this project, I intend to create an efficient voxel rendering system using a level of detail system and culling. I also intend to implement one of the algorithms listed above to smooth the voxel terrain. Then I wish to create a voxel sphere and combine different levels of noise to give the sphere a terrain. Throughout the project, it will be important to try and maximise performance so it can be rendered in real-time after generation. The main issues I wish to address are: ●Performance (Mai...[more]
Ieuan GriffithsLocalising a device in a building using known access pointsStuart M AllenMatthew J W Morgan1920-CM3203
GPS navigation has become a key part of modern life, with many using mobile map apps to find local buildings and shops. The map apps struggle significantly indoors however, and accuracy is a large issue. Even if buildings have the internal structure mapped out, these apps would struggle to identify the users location accurately due to traditional GPS having an accuracy of ~10m. By identifying access points in a building and marking their location on a map, we should be able to crudely identify ...[more]
Isaac RapleyEvaluating serious games as an alternative to traditional teaching methodsDr Daniel J. FinneganHelen R Phillips1920-CM3203
My project aims to study the validity and effectiveness of serious games as a method of teaching basic statistics to primary school children. The project will involve developing a 2D visual novel using Unity and artwork made up of both stock assets and custom assets created by myself, some of which have already been created. Within the game players will take on the role of a detective, using the skills they are taught to solve problems and deduce the identity of criminals. The game will teach pl...[more]
Iwan MunroLinear Programming When There is No Feasible SolutionRichard BoothJose Camacho Collados1920-CM3203
Many real-life industrial problems can be expressed as linear programs (LPs) (sometimes with huge numbers of decision variables and constraints). It is usually assumed that there is at least one feasible solution, but sometimes due to manual error in formulating the problem there might be no feasible solution. In this case what is needed is some way to automatically pinpoint these errors (e.g. finding some minimal set of infeasible constraints) so the industrial partner can correct them, or even...[more]
Jack FurbyExplainable Artificial Intelligence for Multi-Sensor Information FusionAlun D PreeceBailin Deng1920-CM3203
Recent years have seen rapid advances in the field of artificial intelligence (AI), mainly due to breakthroughs in machine learning (ML) using deep neural networks. These approaches learn powerful classification and prediction models from large volumes of data, but have a limitation that the resulting models are usually "black boxes" - it's hard to determine how they work, and even harder to generate explanations ("why?") for a user! Many AI / ML applications involve information fusion: the ge...[more]
Jack WilliamsA personal diary appAlia I AbdelmotyPhilipp Reinecke1920-CM3203
A lot of people may be interested in a simple way of keeping a diary of events in their day- whom they met, they talked to, where they went, how they felt. Some of this information can be extracted automatically from their location traces on their phones and from their activity on social web sites. You will build an app that will automatically track and record some of these personal information and present it to the user in the form of a browsable diary. You will be able to build upon/make u...[more]
Jacob WinkworthAI to detect early signs of Alzheimer’s DiseaseXianfang SunAlia I Abdelmoty1920-CM3203
Despite nearly 50 million people worldwide have Alzheimer's disease or another form of dementia, no cure exists for Alzheimer’s disease. Promising drugs have emerged in recent years that can help stem the condition’s progression. However, these treatments must be administered early in the course of the disease in order to do any good. This race against the clock has posed a serious challenge on how to diagnose the condition earlier. In the face of this challenge, we seek to develop a new dee...[more]
Jorge Correa MerlinoCyber Incident-Response Smart Grid ToolNeetesh SaxenaMartin J Chorley1920-CM3203
The aim of the project is to design and create a tool that could help operators to respond to the cyber-attacks in order to mitigate their impact onto the smart grid. The work involves finding the indicators of compromise, log analysis, and exploring other footprints to understand the nature of cyber-attacks and then develop capabilities to respond to the attacks. The working prototype will demonstrate the working of the project.[more]
Joseph DinnReconstruction of Shredded ImagesPaul L RosinJose Camacho Collados1920-CM3203
Reconstruction of Shredded Images[more]
Joseph HarrisArtificial Intelligence / Machine Learning for Understanding Misinformation in Social MediaAlun D PreeceNeetesh Saxena1920-CM3203
Modern artificial intelligence (AI) has made huge advances due to machine learning on large amounts of data, usually collected from the internet. Current AI systems excel at making predictions or classifications for unseen inputs, based on previously-seen examples. Many researchers, companies and agencies are applying these sorts of data-driven predictive AI to the processing of social media data, e.g., to determine what content to push into a user's feed, or to work out "what's happening". Th...[more]
Joseph WilsonAugmented Reality Card GameBailin DengPhilipp Reinecke1920-CM3203
A card game using Augmented reality. Idea referenced by Dr Daniel Finnegan. The base of the game is to integrate a form of augmented reality within a card game. This can be referenced by the Japanese manga YuGiOh. It should be user-friendly and allow an easy viewing window where the object interacts and the augmentation occurs.[more]
Karolina HornikovaDelivery of Patient Centred Care - System ModellingMartin CaminadaGeorge Theodorakopoulos1920-CM3203
NHS Wales, the Cwm Taf Health Authority, is a child agency of the National Health Service in the United Kingdom. As healthcare services are vital features of our society today, it is important to be able to utilise their large inflow of data effectively. Previously NHS Wales would use patient ‘age’ data as their only way of categorisation; currently they have expanded to categorise patients based on previous health conditions and current ongoing conditions. From current concerns raised by t...[more]
Katie Barrett-PowellA Dashboard Interface for Policing Large EventsAlun D PreeceIrena Spasic1920-CM3203
Cities often host large events. In recent years, for example, Cardiff has played host to diverse events including the 2014 international NATO Summit and the 2016 UEFA Champions League Final, as well as numerous smaller events requiring extensive police and security operations. Policing such events requires very good situational awareness (SA): an ability to keep track of what is happening in real-time. Effective SA involves making use of data collected in real-time from multiple sources - for ex...[more]
Kayley ThomasVirtual Reality AppYukun LaiSurya Thottam Valappil1920-CM3203
With the availability of low-cost virtual reality solutions, including mobile phone based approaches, virtual reality (VR) has become much more accessible to users. The aim of the project is to develop a virtual reality app that provides immersive user experience for navigating in the virtual 3D world. The project can be based on e.g. Google VR SDK. [more]
Kieran ParnellWeb Apps: Using Bots to Expand the Functionality of DiscordFrank C LangbeinJing Wu1920-CM3203
Discord has recently taken over services like Skype, TeamSpeak and Messenger. Discord has become a platform to collaborate, communicate and create. Although Discord is great, I wish to create a Discord Bot to extend the functionality of Discord; exploring the possibilities and limitations of bots within Discord. My aim would be to create a working Discord Bot that brings extra functionality to Discord. Examples of Functions that could be added by a Bot: - Automated Moderation - Activity L...[more]
Lewis HemmingGeneration of Facial CartoonsPaul L RosinYukun Lai1920-CM3203
The aim of this project is to take an input image of a frontal face and generate a cartoon stylised version. Many possible approaches are possible, see e.g. Data-driven face cartoon stylization, Zhang et al., 2014. Image Based Hair Segmentation Algorithm for the Application of Automatic Facial Caricature Synthesis, Shen et al., 2014. At least in the first instance, this project will take a simple approach, in which facial features (eyes, mouth, etc) are matched to a library containing multip...[more]
Matthew Trimbysystem development for placement modules (post start of placement)Martin CaminadaSteven Arthur1920-CM3203
For the two placement modules (undergraduate and postgraduate), the placement team is facing increased student numbers and associated increased workload. Because of this, we are interested in the development of a system that can serve as a platform and provide support to many of the tasks involved in running the placement module. Basically, in the same way as PATS is providing support to the Final Year Projects, the placement team is interested in a system that provides support to the placement ...[more]
Miles BuddenBibTeX Web DatabaseFrank C LangbeinMartin Caminada1920-CM3203
BibTeX is a system to keep track of references and include them in TeX/LaTeX documents. It keeps the information about the references in a single file. This is problematic when more than one person wants to edit the file and keep entries synchronised. The idea of this project is to create a web interface to edit a bibtex database, where individual entries are stored in a NoSQL database. An interface to LaTeX editing systems (to create the bibtex database for a specific document), and searching a...[more]
Muhammad AhmedRemote Trap Activation Detection using IoTCharith PereraHantao Liu1920-CM3203
This is a real-world project. Your task is to develop a IoT based sensor node that can be deployed in a remote location to detect trap activation (in the wild in Malaysia) https://www.cardiff.ac.uk/danau-girang-field-centre. Students who complete the project successfully may be funded by the Cardiff University to travel to DGFC in Malaysia for 3 weeks (Subject to the quality of the project completion). Last year we sent 2 students to DGFC.[more]
Neofytos NeokleousIdentifying diagnostic features of Parkinson’s disease using classical machine learning.Matthias TrederDr Daniela Tsaneva1920-CM3203
This project will be jointly supervised with the University of Stellenbosch, South Africa. They collected a brain imaging dataset involving Parkinson's, PTSD, and schizophrenia patients. The data have been preprocessed and parameters describing the brain structure have been extracted. This is a data science project, that is, you will use Sci-kit learn and statistics in Python to analyse, visualise and interpret the data. Excellent Python skills are vital, prior knowledge of machine learning is ...[more]
Niall CurtisAlgorithmic ‘Idea’ clustering via natural language processingAlun D PreeceSteven Arthur1920-CM3203
During my industry year, the company I worked for produced web software for other companies in order to enable them to conduct challenge-based innovation, through crowdsourcing ideas from their employees. In many cases, a single challenge might amass a large number of idea contributions from staff, which is a problem for a small staff team to process manually. This project aims to address part of this problem through creating a system that uses advanced techniques (such as natural language proce...[more]
Oliver GunnellDetecting network based attacks in Industrial Control SystemsEirini S AnthiNeetesh Saxena1920-CM3203
The student will use publicly available datasets from industrial control systems such the Secure Water Treatment (SWaT) Dataset, to build a lightweight intrusion detection system tailored for ICS protocols. These datasets contain not only benign traffic but also malicious, which can be used to build (machine learning based) models that distinguish between the different types of attacks. Other parameters for detecting attacks, such as PLC data, can also be taken into consideration.[more]
Oliver Storey-YoungCreating an online adblockerPadraig CorcoranFederico Liberatore1920-CM3203
When browsing the internet, users are regularly bombarded with adverts that will hinder and alter their browsing experience. Adverts are used as a revenue technique for many free and paid websites. Over time the market for adverts has evolved and now an ever-raging war between website hosts and adblockers exist, with each trying to combat the other. In recent years adverts have been used as a strong part of political campaigns – being used to spread propaganda across the internet using smart t...[more]
Paige PesiganEvaluating the robustness of a lattice-based cryptosystemEirini S AnthiFrank C Langbein1920-CM3203
Quantum computing is expected to be mainstream at some point in the not-too-distant future. However, current public-key cryptographic methods can be broken by such devices with the use of Shor's algorithm, meaning that data encrypted today is in danger of being decrypted later by these more-powerful computers. Lattice-based cryptography is said to be safe against both classical computers and, according to IBM, quantum computers that are fault-tolerant and have computing power even in the million...[more]
Patrick NoyauHyperpartisan News DetectionLuis Espinosa-AnkeDr Daniel J. Finnegan1920-CM3203
This is a Natural Language Processing project which addresses hyperpartisanism [1]. Formally, the task is defined as: Given a news article text, decide whether it follows a hyperpartisan argumentation, i.e., whether it exhibits blind, prejudiced, or unreasoning allegiance to one party, faction, cause, or person. The project will use the official data from a recent competition [2]. A strong deliverable should consist of a text categorization system written in Python that works, i.e., produces o...[more]
Peter GhawiIoT Security (multiple projects available)Philipp ReineckeRichard Booth1920-CM3203
General idea: Try to successfully compromise an IoT device.[more]
Reema AbaoudAI/ML Performance Evaluation (multiple projects possible)Philipp ReineckeMatthew J W Morgan1920-CM3203
Android malware growth has been increasing dramatically along with increasing the diversity and complicity of their developing techniques. One of the main technique that is used to detect malware is machine learning methods. This is why there is a crucial demand to evaluate machine learning algorithms performance in detecting malware. Hence,the purpose of this work is to evaluate and examine the machine learning algorithms performance in classifying malware and benign files. Three different cla...[more]
Robert TaylorHIV Neuroimaging Data Analysis and Supervised Machine LearningMatthias TrederVíctor Gutiérrez Basulto1920-CM3203
HIV is one of the deadliest infections affecting almost 40 million people in the world. It's estimated that 20.4% of South African Women have been infected. As HIV causes neuroinflammation and neuronal death, it should be detectable in neuroimaging data. A neuroimaging dataset has been provided by Stellenbosch University, South Africa with 124 participants involved. This project will use data analysis and machine learning techniques in the python programming language to help understand the brain...[more]
Ross SingletonAbusive language detection against immigrants and womenLuis Espinosa-AnkeDavid J Humphreys1920-CM3203
In this proposal a student will explore computational approaches for modeling and detecting biased and abusive language in social media. There are two broad topics: (1) Offensive language detection, which occurs when individuals take advantage of the perceived anonymity of computer-mediated communication, and engage in behavior that many of them would not consider in real life [1]; and (2) Hate Speech detection. Hate Speech is commonly defined as any communication that disparages a person or a g...[more]
Samuel Meehansystem development for placement modules (pre start of placement)Martin CaminadaJing Wu1920-CM3203
For the two placement modules (undergraduate and postgraduate), the placement team is facing increased student numbers and associated increased workload. Because of this, we are interested in the development of a system that can serve as a platform and provide support to many of the tasks involved in running the placement module. Basically, in the same way as PATS is providing support to the Final Year Projects, the placement team is interested in a system that provides support to the placement ...[more]
Samuel WincottGenerating Differentially Private Datasets Using Deep Learning (in collaboration with: Office for National Statistics)George TheodorakopoulosYukun Lai1920-CM3203
Government organisations, businesses, academia, members of the public and other decision-making bodies require access to a wide variety of administrative and survey data to make informed and accurate decisions. However, the collecting bodies are often unable to share sensitive data without risking breaking the confidentiality and consent checks required to obtain this data. Therefore, researchers have proposed many methods for generating synthetic data to replace the raw data for the purposes o...[more]
Sarah LoughranCurriculum mapping databaseHelen R PhillipsStefano Zappala1920-CM3203
Curriculum mapping database. The majority of degree programmes, have learning outcomes at module,, year and programme level, however mapping these and keeping track is a challenge. The objective of this project is to development a database for input and retrieval in relation to curriculum mapping. The database would be expected to store, how assessments map to the learning outcomes and potentially also map to the accreditation criteria of the associated subject professional body.[more]
Sean SavitzEnd-to-End Open Source Reconfigurable IoT Stack: Best Practices for Large Scale DeploymentsCharith PereraNatasha Edwards1920-CM3203
In this project, you will add new features to an IoT edge network we have developed last year. Currently, Eclipse Kura and Eclipse Kapua used to capture data through sensors and cameras. This is a 100 Raspberry Pi Edge network. Currently, we use TensorFlow to perform deep early at the edge Your job is to improve this infrastructure by arriving data visualisation and semantic data modelling techniques.[more]
Sebastian SquirePredicting Driver Insurance ClaimYuhua LiAlun D Preece1920-CM3203
A more accurate claim prediction will allow insurance companies to further tailor their prices, and hopefully make auto insurance coverage more accessible to more drivers. This project will build a model that predicts the probability that a driver will initiate an auto insurance claim in the next year. You are free to choose to use dataset(s) you find yourself. A potential dataset can be https://www.kaggle.com/c/porto-seguro-safe-driver-prediction/data. This project requires you to have good kn...[more]
Sin Ching CheungArtificial Intelligence and Human Factors for Multimedia ApplicationsHantao LiuIrena Spasic1920-CM3203
Do you know where you are looking at while watching online videos? Do you know where the computers think you are looking at? Modelling visual saliency - predicting where human eyes pay attention to in visual content - has been a very active research area over the past few years in both academia and industry. This project aims to investigate saliency modelling in emerging applications in image and vision computing, such as computer rendered images, high-dynamic-range (HDR) imaging, medical imagin...[more]
Spyros LontosLearning to Detect and Recognise Road SignsYukun LaiMichael Daley1920-CM3203
To improve driving experience, computer vision techniques are widely used for providing assistance. The project aims to develop a prototype that implements some machine learning technique to detect and recognise road signs from real world camera footage. Such technology is also an essential component for autonomous driving.[more]
Stanislav KataevDetecting fraud from customer transactionsYuhua LiXianfang Sun1920-CM3203
This project aims to develop a method that detects fraudulent e-commerce transactions so that customers are protected from financial fraud. It will apply suitable machine learning techniques to deal with distinct characteristics of e-commerce transaction data which is highly imbalanced. You may get dataset(s) for your project, but a potential dataset that can be used for this project can be downloaded from https://www.kaggle.com/c/ieee-fraud-detection/overview. This project requires you to have...[more]
Talis HarrisAugmented Reality development Project: The Stone, the Mermaid, and the PainterFrank C LangbeinMartin Caminada1920-CM3203
ABSTRACT: The following Final Year Project provides a cutting-edge Augmented Reality solution to tourism challenges faced by The Hilton Historic Trust in a remote highland village in Scotland. It includes 4 AR experiences linking to key artefacts surrounding the village. The experiences are of an infographic nature, containing facts, media, links, and additional features. The results of this project have the potential to answer whether AR is viable in remote locations to bring historic informa...[more]
Worawat Kimparaautomatic generation of Word documents for the Year in Industry programmeMartin CaminadaNatasha Edwards1920-CM3203
For its Year in Industry programme, the School of Computer Science and Informatics uses a document called the Skills Framework for the Information Age (SFIA). This is an industry standard that contains a wide variety of IT-related professional skills (such as "Software Development" or "Testing"). For each of these skills, different levels are identified, at which a professional can operate. In general, the higher levels indicate that a person is doing work of greater complexity, with more influe...[more]
Yibin XuEnhancing the accuracy of rigid registration using a general and adaptive robust loss functionBailin DengMatthew J W Morgan1920-CM3203
Rigid registration of two geometric data sets is essential in many applications, including robot navigation, surface reconstruction and shape matching. Most commonly, variants of the Iterative Closet Point (ICP) algorithm are employed for this task. These methods alternate between closest point computations to establish correspondences between two data sets, and solving for the optimal transformation that brings these correspondences into alignment. However, a major difficulty for this approach ...[more]