Filter by keyword, subject, or both. Updates live as new model answers are added to our portal.
Advanced Research Topics
Generative Adversarial Networks – Building and Applying GANs to Real-World Problems
This individual assessment focuses on building and understanding Generative Adversarial Networks (GANs) and applying them to real-world problems across different domains. The assignment is based on the GAN topics covered in Unit 5 and requires students to implement GAN models, analyse their behaviour, generate synthetic data and critically evaluate the quality of the generated outputs. The assessment consists of two main parts: building and understanding GANs from scratch using synthetic two-dimensional data, followed by applying GANs to three real-world application areas involving medicine, cybersecurity and creative artificial intelligence. Part 1 focuses on developing a practical understanding of GANs through implementation using PyTorch and synthetic 2D data. Students must first reproduce the sine-wave GAN demonstrated in the tutorial. They then create and model a new two-dimensional distribution, selecting from a 2D spiral, a mixture of Gaussians, or a noisy parametric curve defined as y = sin(2x) + 0.3cos(5x) + epsilon. Students must modify the GAN architecture, for example by changing the activation function or layer depth, and visually compare the original data distribution with the generated samples. This section is intended to reinforce understanding of the generator, discriminator, GAN training process and the effect of architectural choices on generated data. Part 2 addresses three real-world GAN applications. The first application concerns optical coherence tomography (OCT) retinal images using the OCTMNIST subset of MedMNIST. Students must explore the dataset, examine class distributions and sample images, build a ConvNet-based Deep Convolutional GAN (DCGAN) using PyTorch or TensorFlow/Keras, train the model and track generator and discriminator losses. Generated retinal images must be compared with real images both visually and quantitatively, using a performance measure such as the Fréchet Inception Distance (FID). An optional extension involves implementing a conditional GAN and demonstrating generation of images for specific classes. The second application focuses on cybersecurity using the preprocessed CICIDS2017 dataset containing DDoS and benign network traffic. Students must combine the relevant data files, explore the dataset and understand its features and class balance. A GAN must be developed to generate synthetic feature vectors rather than images. The model should be trained using benign and DDoS attack data, training loss curves should be monitored, and PCA or t-SNE should be used to visualise and compare real and generated feature distributions. The quality of the synthetic data should then be evaluated. An optional extension involves expanding the analysis to the full CICIDS2017 dataset and examining generalisation across different attack types. The third application focuses on Creative AI using the QuickDraw birthday cake category. Students must explore the birthday cake sketch dataset and implement a ConvNet-based GAN (DCGAN) to generate realistic birthday cake sketches. Generated sketches should be evaluated visually and quantitatively, including comparison with real examples and an appropriate metric such as FID. An optional extension involves generating samples from additional QuickDraw categories and discussing how model performance changes across different classes and levels of sketch complexity. The submission consists of both code and a written report. The code accounts for 60% of the assessment and must complete the required modelling tasks, present generated samples, compare generated and real data, use appropriate functions and include clear annotations so another user can understand the implementation. The report accounts for 40% and must be 6–8 pages. It should explain the analytical steps undertaken, justify the selected approaches, provide brief descriptions of the models used, interpret the results and include appropriate figures, evaluation metrics and academic references. The report should critically discuss why particular network architectures were selected rather than simply providing textbook definitions.
Read Model Answer →
Data Science
1,000 words
Individual Project – Image Segmentation
This individual project focuses on image segmentation using a subset of the ADE20K dataset. The assessment requires students to investigate viable image segmentation models, analyse and visualise the provided data, develop appropriate data preprocessing procedures, train a segmentation model and critically evaluate the resulting performance. The assignment is worth 25% of the module assessment and is designed to assess students’ ability to apply research methods to a data science problem, communicate analytical findings effectively and select appropriate methods while understanding their advantages and limitations. Students are required to work with the provided ADE20K dataset and produce a model capable of semantic segmentation for four specified object classes: person, car, book and airplane. The model must identify which regions of an unlabelled image correspond to these four classes. The assignment specifically requires semantic segmentation rather than simply identifying whether an object is present. Treating the task as binary segmentation, where all objects are considered a single foreground class, will result in a significant penalty. The project requires students to investigate different approaches and models, perform exploratory data analysis, develop suitable preprocessing procedures and customise their selected model or models to improve performance. Students may use more than one model, although at least one model must be partially or fully trained by the student. Where multiple models are used, comparison with pretrained models is encouraged. Potential approaches include segmentation architectures such as U-Net and Mask R-CNN, with appropriate model selection justified through relevant literature and experimental evidence. The report should contain at least four core sections: an Introduction incorporating a literature review, Data Description and Exploratory Data Analysis, Methodology, and Results and Discussion. The literature review should cite at least three relevant research papers. The results section should include evaluation of three test images using the developed model and comparison with relevant published literature. Students are expected to provide a critical analysis of their work rather than simply reporting numerical results, explaining the reasons for observed outcomes and considering how modelling and implementation choices affected performance. The assessment also requires a documented Google Colab notebook containing the implemented steps used to train and evaluate the model. The notebook should be accessible to markers and should demonstrate the preprocessing, model development, training and evaluation process. The report must be between 700 and 1,000 words, excluding references and code, and should include code in text format as an appendix rather than screenshots. Figures and tables should have clear captions explaining what they present and their source. Assessment criteria include literature review and context, dataset description and exploratory data analysis, implementation quality, quality of analysis and critical discussion, and report quality and structure. Strong work is expected to demonstrate appropriate model selection, effective preprocessing and augmentation, suitable evaluation metrics such as Intersection over Union (IoU) and Dice score, meaningful visualisation and a critical interpretation of results and limitations.
Read Model Answer →
Advanced Artificial Intelligence
CW1: Applying advanced AI methods for analysing text documents
This coursework is an individual practical assessment for the Advanced Artificial Intelligence module, focusing on the application of advanced artificial intelligence, natural language processing (NLP), natural language understanding (NLU), and neural network techniques to analyse text documents. The coursework uses a provided social-media dataset containing more than 89,000 posts associated with 947 news headlines. Each social media post is treated as an individual text document and is labelled as either real or fake according to the relationship between the linked news headline, its ground-truth status, and majority annotator agreement. The assessment consists of two main tasks. The first task focuses on the identification of fake text documents. Students are required to perform descriptive data analysis and NLP preprocessing before transforming text into numerical feature representations suitable for neural network classification. Students must design and implement both a Multi-Layer Perceptron (MLP) network and a deep learning neural network to classify documents as real or fake. The assessment requires justification of the selected architectures, including inputs, layers, neurons, activation functions and outputs. Students must also experiment with hyperparameters, evaluate model performance using appropriate metrics, compare different models, select suitable models, and save trained models for later demonstration. The second task focuses on topic discovery using natural language understanding techniques. Students are required to apply text-processing methods such as tokenisation, stop-word removal, lemmatisation or stemming and experiment with at least two different text representation strategies. Possible approaches include Bag of Words, TF-IDF, Latent Dirichlet Allocation, word vectors and word embeddings. The discovered topics must be analysed and interpreted in relation to the document content, associated news headlines, and class labels. The assessment is supported by a practical bench demonstration and a maximum of seven PowerPoint slides covering the design, model improvement process, performance evaluation and discussion of results. Students must submit their own Python code and presentation through Blackboard and demonstrate their saved models without retraining them during the demonstration. The marking scheme allocates 45% to fake text document identification, 35% to topic discovery using NLU, and 20% to the structure, organisation, professionalism and question-and-answer performance of the demonstration.
Read Model Answer →
Artificial Intelligence / Natural Language Processing / Deep Learning
Applying Advanced AI Methods for Analysing Text Documents
This Advanced Artificial Intelligence coursework requires students to implement and evaluate natural language processing, natural language understanding and neural-network techniques for analysing text documents. The assessment uses a supplied social-media dataset containing more than 89,000 posts linked to 947 news headlines, with each post labelled as real or fake through a combination of headline ground truth and majority-vote annotation. Each social-media post is treated as an individual text document for classification and topic-analysis purposes. CMP_6059B_7059B_2025_26_CW1-pre… The first major task focuses on identifying fake text documents. Students must preprocess the text using appropriate NLP techniques, transform documents into numerical feature representations and experiment with alternative preprocessing approaches to determine which performs best. A separate unseen test set must be reserved to evaluate generalisation, while the remaining data is used for training both shallow and deep neural-network classifiers. Students are expected to explain and justify how the data is split. CMP_6059B_7059B_2025_26_CW1-pre… Students first design a multi-layer perceptron (MLP) capable of predicting whether documents are real or fake. The architecture must be justified in terms of input dimensions, number of layers, neuron counts, activation functions and outputs. A second deep-learning neural network must then be developed for the same classification problem, with justification of the chosen network structure, layer types, activations and other configuration decisions. CMP_6059B_7059B_2025_26_CW1-pre… For both networks, students train baseline models and select three hyperparameters considered most important for improving performance. These hyperparameters must be tuned systematically, with visuals prepared to show the experimentation process and resulting performance changes. Appropriate evaluation metrics are then used to compare the trained models. The strongest MLP and deep-learning models must be saved so that they can be loaded and tested on unseen data during the final demonstration without retraining. CMP_6059B_7059B_2025_26_CW1-pre… The second task focuses on topic discovery using NLP and NLU techniques. Students perform syntactic preprocessing such as tokenisation, stop-word removal and lemmatisation or stemming, and experiment with at least two different text-representation approaches. Suggested methods include Bag of Words, TF-IDF, LDA, word vectors and word embeddings. Students must interpret the discovered topics and explain how those topics relate to document content, linked news headlines and class labels. The best topic-discovery model or models must also be saved for live analysis during the demonstration. CMP_6059B_7059B_2025_26_CW1-pre… The assessment is completed through a bench demonstration, supported by a maximum of seven PowerPoint slides. The slides should document the system design, model-improvement process, performance evaluation and discussion of results for both fake-document classification and topic discovery. Students also submit a ZIP file containing only their Python source files. The demonstration lasts up to 15 minutes, consisting of approximately 10 minutes for presentation and technical demonstration followed by 5 minutes for questions and transitions. CMP_6059B_7059B_2025_26_CW1-pre… The marking scheme allocates 45% to fake-document identification, including descriptive analysis, preprocessing, MLP design and deep-learning design; 35% to topic discovery, including preprocessing, model development and interpretation; and 20% to the structure, organisation, professionalism and Q&A quality of the demonstration. CMP_6059B_7059B_2025_26_CW1-pre… Important for the public Reference Library: the brief explicitly states that the use of Large Language Models or generative AI to produce any part of the submission is strictly prohibited, including code, data processing, testing, writing or PowerPoint content. Therefore, this entry should remain only a high-level public description of the assessment and should not be presented as material intended for direct student submission. CMP_6059B_7059B_2025_26_CW1-pre…
Read Model Answer →
Data Science / Artificial Intelligence / Generative Modelling
Generative Modelling Case Study: GANs for Medical Imaging, Cybersecurity and Creative AI
This Generative Modelling Case Study requires students to design, implement and evaluate Generative Adversarial Networks (GANs) across a range of synthetic and real-world applications. The coursework develops both theoretical understanding and practical deep-learning skills, with emphasis on building models, evaluating generated data and critically interpreting model performance. The assessment addresses research understanding, originality, future development and scholarly communication in data science. Generative modelling case study… Part 1 – Building and Understanding GANs from Scratch focuses on fundamental GAN concepts using synthetic two-dimensional data. Students first reproduce a sine-wave GAN from the tutorial and then create a second synthetic distribution using either a 2D spiral, a mixture of Gaussians or a noisy parametric curve. They must modify aspects of the GAN architecture, such as activation functions or network depth, and visually compare generated samples with the original data distribution. Generative modelling case study… Part 2 – Real-World GAN Applications extends the work across three application domains. The first application uses the BloodMNIST subset of MedMNIST to train a DCGAN that generates synthetic blood-cell microscope images. Students explore the dataset, analyse class distributions, train the model, monitor generator and discriminator losses and compare real and generated images using both visual inspection and quantitative measures such as the Fréchet Inception Distance (FID). An optional extension involves implementing a class-conditioned GAN capable of generating images from specific categories. Generative modelling case study… The second application addresses cybersecurity using the CICIDS 2017 intrusion-detection dataset. Students construct a GAN that generates synthetic network-traffic feature vectors rather than images. The model is trained using benign and denial-of-service traffic, and generated samples are compared with real traffic using dimensionality-reduction techniques such as PCA or t-SNE. Students must evaluate how closely the synthetic traffic reflects the distribution of genuine network data. An extension allows analysis of the full CICIDS dataset and evaluation across different attack types. Generative modelling case study… The third application explores Creative AI using the Google QuickDraw pizza category. Students implement another DCGAN to generate artificial pizza sketches, track training behaviour across epochs, and compare generated sketches with genuine examples using both visual inspection and quantitative metrics such as FID. Extension work may examine additional QuickDraw categories and investigate how model performance changes with class and sketch complexity. Generative modelling case study… Submission consists of a 6–8 page report together with working code. The report should explain the analysis undertaken, justify modelling decisions, describe the network architectures, interpret the results and incorporate suitable figures, evaluation metrics and references. The accompanying code must reproduce the figures, models and numerical results reported and must execute successfully when tested. Generative modelling case study… The marking scheme places 60% of the marks on code and 40% on the report. Within the coding component, 40 marks relate to completing the GAN modelling tasks and 20 marks assess code quality, modularity and annotation. The report is assessed on discussion and interpretation of the analysis, justification of architectural choices, results presentation, document quality, figures and appropriate academic references. Generative modelling case study… Important for the public Reference Library: the brief explicitly states that students must not use generative AI to write the report, and the rubric states that AI-generated report text can result in zero marks for the whole assignment. Therefore, use this entry only as a high-level public description of the assessment and do not present generated report content as something students can submit directly. Generative modelling case study… Generative modelling case study…
Read Model Answer →
Artificial Intelligence / Machine Vision / Computer Vision
Artificial Intelligence and Machine Vision: Neural Network-Based Image Processing Application
This postgraduate Artificial Intelligence and Machine Vision coursework requires students to design, implement and critically evaluate a neural network-based image-processing application addressing a real-world problem. Students may select an application such as medical imaging, plant or fruit classification, skin cancer detection, object detection or image segmentation and must demonstrate an appropriate end-to-end machine vision workflow. CN7023 Coursework T2 25-26 (1) The project begins with a clear definition of the selected real-world problem, the objectives of the proposed solution and its potential practical impact. Students are expected to demonstrate creativity in selecting and designing their approach, explain the neural-network or image-processing methods adopted and justify why the selected techniques are suitable for the chosen dataset and application. CN7023 Coursework T2 25-26 (1) A substantial component focuses on simulation and implementation. Students describe the dataset, including its source, size, classes and representative images, before explaining how image data were encoded and preprocessed for use within the chosen neural network. The report must then document the selected network architecture, learning algorithm and procedures used for training, validation and testing. CN7023 Coursework T2 25-26 (1) Model performance must be communicated using quantitative and visual evidence. Required outputs include test-set accuracy, accuracy curves across training, validation and testing, and a confusion matrix supported by appropriate explanation. Students must also critically analyse the results, identify factors affecting model performance and discuss alternative methods or simulation changes that could improve the solution. CN7023 Coursework T2 25-26 (1) The coursework permits several technical routes, including combining image processing with artificial neural networks, deep learning or computer vision, or focusing on one of these approaches independently. Development may be completed using MATLAB or Python. The wider module covers artificial neural networks, CNNs, digital image processing, image restoration, compression, segmentation, classification and ethical, legal, privacy and social issues associated with AI systems. CN7023 Coursework T2 25-26 (1) Module handbook 2526-B (1) Overview word count: approximately 335 words. Important note: the coursework cover page labels the assignment as “Individual Assignment 100%,” but the module handbook clarifies that the coursework report itself contributes 50% of the module, with the remaining marks allocated to MATLAB course completion (20%), lab tasks (15%) and presentation (15%). For the Reference Library, I would use the handbook’s 50% report weighting if you need to record the assessment contribution.
Read Model Answer →
Multimodal Sentiment Intelligence Platform for Dynamic Market Insights
This assignment presents the development of a Multimodal Sentiment Intelligence Platform for Dynamic Market Insights. The project addresses the need for real-time market sentiment analysis by combining text and visual data through an AI-driven multimodal approach. Traditional sentiment analysis methods may have limitations when dealing with diverse data modalities, and this project aims to address this gap by integrating multiple artificial intelligence and deep learning techniques. The primary objective is to develop an AI-driven platform capable of performing real-time sentiment analysis and supporting market trend prediction. The proposed system also aims to generate business intelligence insights that can support applications in marketing, finance, and customer service. The project incorporates deep learning, large language models, and multimodal fusion techniques to improve sentiment understanding. For text-based sentiment analysis, the assignment identifies BERT and GPT-2 for sentiment classification. For visual sentiment analysis, YOLOv8 is used for object detection while DeepFace is incorporated for facial emotion recognition. Feature-level and decision-level fusion strategies are applied to combine information from different modalities and improve the overall sentiment analysis process. Retrieval-Augmented Generation (RAG) is also incorporated to provide context-aware sentiment insights. The proposed platform uses Amazon reviews and IMDb reviews as text datasets. For image or video-based multimodal data, the assignment references the CMU-MOSEI dataset and an Amazon-Reddit merged reviews dataset. Overall, the work focuses on combining natural language processing, computer vision, deep learning, large language models, multimodal fusion, and retrieval-augmented generation to create a platform capable of producing dynamic sentiment and market insights.
Read Model Answer →
Machine Learning and Deep Learning
2,000 words
Development and Evaluation of Deep Learning Models for Healthcare Classification
This individual technical assessment focuses on the design, development, analysis and evaluation of a deep learning solution for a healthcare-related classification problem. Students select one of two provided scenarios: Polycystic Ovary Syndrome (PCOS) detection using ultrasound images or heartbeat classification using electrocardiogram (ECG) signals. The objective is to develop an appropriate deep learning approach and demonstrate critical understanding of the complete machine learning workflow, from initial data exploration through to model evaluation and reflection. Students may either design and train a deep learning model from scratch or customise and fine-tune an existing pre-trained architecture. The complete work is presented through a single Jupyter Notebook integrating Python code, technical discussion, results and visualisations. The notebook must clearly define the selected healthcare problem, explain its significance, justify methodological and architectural choices, and critically evaluate the resulting solution. The first stage involves exploratory data analysis and preprocessing, including investigation of class distributions, data imbalance and relevant patterns. Students prepare the data through techniques such as normalisation, augmentation, train-validation-test splitting and appropriate handling of class imbalance. This is followed by model design, training, validation and hyperparameter tuning, with the architecture selected according to the characteristics of the data and classification task. Model performance must then be evaluated using appropriate classification measures, including precision, recall, F1-score, ROC curves and area under the curve (AUC). The developed model should also be compared against suitable benchmark approaches, which may include traditional machine learning algorithms or alternative deep learning architectures. This comparison should identify the relative strengths and limitations of the proposed solution. The final component requires clear visual presentation and critical reflection on the complete modelling process, including limitations, challenges and opportunities for improvement. Importantly, grading prioritises methodological rigour, analytical depth and critical evaluation rather than simply achieving the highest predictive accuracy. Overview word count: approximately 330 words. AI restriction: this brief only permits automated AI tools for spelling and grammar checking. It explicitly prohibits tools such as ChatGPT, Gemini or Copilot from authoring assessment text or code; any permitted AI use must also be acknowledged.
Read Model Answer →
Data Science / Deep Learning
3,000 words
Advanced Research Topics (7PAM2016) — Building GANs from Scratch and Applying Them to Medical Imaging, Network Traffic and Sketch Generation
This Masters-level assessment asks for a complete generative adversarial network study, delivered as an annotated code submission carrying sixty per cent of the marks and a six-to-eight page technical report carrying the remaining forty. The work spans four separate GAN implementations, moving from a controlled synthetic setting into three contrasting real-world application domains. Part one builds a GAN from scratch in PyTorch on synthetic two-dimensional data. The tutorial sine-wave generator is reproduced first as a baseline, then a new distribution is modelled — a noisy parametric curve of the form y = sin(2x) + 0.3cos(5x) with an additive noise term — before the architecture itself is varied. Activation functions and layer depth are altered systematically and the resulting sample distributions plotted against the originals, so the effect of each architectural choice on convergence and sample fidelity can be seen rather than asserted. Part two applies the same principles at scale across three domains. The medical strand trains a DCGAN on the OCTMNIST subset of MedMNIST, generating synthetic optical coherence tomography retinal images, tracking generator and discriminator losses across training, and evaluating output both visually and quantitatively using Fréchet Inception Distance. A conditional GAN extension conditions the generator on class label so that images for a chosen retinal pathology can be produced on demand. The cybersecurity strand shifts from images to feature vectors, using preprocessed CICIDS 2017 network intrusion data. Benign and DoS traffic is combined and explored for class balance, a GAN is built to synthesise tabular feature vectors rather than pixels, and real against generated distributions are compared through PCA and t-SNE projections, with a discussion of how well the model generalises across attack types. The creative strand trains a DCGAN on the QuickDraw 'birthday cake' sketch category, tracking visual outputs epoch by epoch and benchmarking generated sketches against real ones, with an extension covering additional categories of differing sketch complexity. The accompanying report explains the analysis steps and the reasoning behind each architectural decision rather than restating textbook definitions of the method. It gives brief descriptions of the models used, presents generated samples and loss curves as figures, interprets the evaluation metrics, and reflects honestly on failure modes — training instability, mode collapse, and the visible flaws in synthetic output that determine whether such data is fit for downstream use. The code is written as reusable functions, commented for a reader other than its author, and reproduces every figure and numerical value quoted in the report.
Read Model Answer →
Artificial Intelligence
2,000 words
End-to-End Applied AI Development — Comparative Machine Learning and Neural Network Modelling on a Public Dataset
This assessment runs a complete applied AI development cycle end to end: problem definition, dataset selection, preprocessing, model building, optimisation, evaluation and critical reflection. Students identify a real-world problem themselves, formulate a research question from it, and source a suitable dataset from a recognised public repository such as UCI, Kaggle, Data.gov or OpenML. Dataset choice carries more weight than students expect. It must be genuinely suitable for supervised learning, complex enough to make preprocessing and feature engineering meaningful, and — critically — structured so that a traditional machine learning approach and a deep learning approach can be sensibly compared on it. A dataset too small or too clean makes the neural network component pointless; one too large or too noisy makes the whole pipeline unfinishable within the page limit. The source must be referenced and the choice explicitly justified against the research problem. The modelling requirement is fixed: at least two supervised machine learning models, plus one artificial neural network built in a mainstream deep learning framework, all trained and tested. The comparison between them is the analytical core of the work. Reporting that the neural network scored higher is not an answer; explaining why, in terms of the data's structure and each model's inductive assumptions, is. Marks are distributed across problem framing, the traditional models, the deep learning model, evaluation and critical analysis including responsible AI considerations, and academic communication. That responsible AI component is easy to overlook and is not decorative — it asks what the model's limitations mean for anyone who might rely on it. Presentation requirements are specific. The report is page-limited rather than purely word-limited, and every plot must be described in the text while also being legible enough to communicate on its own — a common failure is dense default library output pasted in without axis labels or scale. The implementation is documented in a notebook combining markdown and code cells so the development process is visible, not just the final result, and submissions typically include the cleaned dataset alongside the code. The strongest submissions treat the notebook and the report as one argument. Weaker ones produce a working notebook and then write a report that describes it, rather than a report that uses it as evidence. Our support on assessments of this type is guidance-based. Typical areas of help include: advising on whether a candidate dataset can actually support the required model comparison, explaining how to justify preprocessing decisions, clarifying which evaluation metrics suit which problem type and why accuracy alone is often misleading, showing how to structure a critical limitations and responsible AI discussion, checking Harvard referencing, and reviewing a student's own draft against the published marking criteria.
Read Model Answer →