Filter by keyword, subject, or both. Updates live as new model answers are added to our portal.
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 →
Data Mining / Data Science
Customer Service Escalation Risk Analytics Using Data Mining and Machine Learning
This advanced Data Mining assessment applies the Knowledge Discovery in Databases (KDD) process to a realistic customer-service risk scenario. Acting as a Data Scientist, students analyse a historical Customer Service Escalation Risk dataset to help an organisation identify early indicators of dissatisfaction and operational bottlenecks that may lead to serious or legal customer escalations. The resulting analysis is intended to support strategic decisions relating to staffing, employee training, customer-journey improvement and escalation prevention. CMP-7023B_Assessement_2 (2) The dataset incorporates customer demographics, account characteristics, communication channels, issue categories, operational measures such as waiting times, transfers and SLA breaches, behavioural indicators including sentiment and response delays, and commercial variables such as monthly fees and contract value. The target variable, escalation_level, contains four categories: No escalation, Minor escalation, Serious escalation and Legal escalation. CMP-7023B_Assessement_2 (2) Students begin with data exploration and visualisation, producing appropriate descriptive statistics and identifying patterns, distributions and potential data-quality concerns. They then perform data cleansing, transformation, feature engineering and preprocessing. Variables that may introduce leakage or unreliable predictions because of their meaning, timing or quality must be critically assessed and justified. CMP-7023B_Assessement_2 (2) The supervised-learning stage requires students to develop, tune and compare predictive models using techniques such as k-nearest neighbours, Decision Trees, Support Vector Machines, ensemble methods or neural networks. Appropriate multiclass evaluation metrics must be used, alongside interpretation of influential variables and model behaviour. CMP-7023B_Assessement_2 (2) The assessment also includes unsupervised learning, requiring comparison of clustering methods such as K-Means and hierarchical clustering after removal of the target variable. Students may apply encoding, normalisation and dimensionality-reduction methods such as PCA or t-SNE and must interpret how the resulting clusters relate to escalation behaviour. CMP-7023B_Assessement_2 (2) Overall, the project assesses independent analytical judgement, modelling justification, comparative evaluation and clear communication of actionable findings for both technical and executive audiences. CMP-7023B_Assessement_2 (2) Overview word count: approximately 340 words. AI-use note: AI tools may only assist with small, specific code snippets. Any AI-generated code must be clearly acknowledged and cited, while the analysis, coding decisions, interpretation and final evaluation must remain the student's own work. CMP-7023B_Assessement_2 (2)
Read Model Answer →
Data Mining
Data Mining – Classification, Model Optimisation and Evaluation
This individual Data Mining assignment requires students to apply the techniques covered in the module using the WEKA data mining platform. The assessment is worth 40% and focuses on practical application of machine learning and data mining methods, requiring students to configure algorithms, analyse datasets, optimise model parameters, evaluate classification performance and explain their technical choices and results. The assignment also assesses the ability to critically evaluate different algorithms and models of data mining. The assignment includes several tasks covering different stages of the data mining process. Students are required to work with supplied datasets and use appropriate preprocessing and classification techniques. The datasets include a balanced screenshot dataset containing processed screenshots classified into categories such as "Okay" and "Bad", where the objective is to train a model capable of identifying inappropriate content. The data has been processed using PCA to provide a smaller four-dimensional representation while protecting privacy and reducing the size of the data. Another supplied dataset concerns furniture reviews, containing positive ("pos") and negative ("neg") written feedback, with the objective of building a model that can determine whether a furniture review belongs to either class. The assessment evaluates students' ability to understand and describe the datasets, including the number of instances, number of columns, data types and relevant statistical information. For text-based data, students must apply appropriate vectorisation and describe the resulting dataset characteristics. Students must also consider class imbalance and apply an appropriate method where necessary, explaining how the chosen approach affects the distribution of instances between the classes. A significant component of the assignment involves classification algorithms and parameter optimisation. The assessment requires students to work with algorithms including Naive Bayes, LibSVM and J48. Students must investigate appropriate parameters and perform parameter searches or fine grid searches to identify suitable configurations. They must explain the selected parameters, their impact on the model and the reasoning behind the chosen values. Model performance must be evaluated using appropriate validation techniques, including cross-validation. Students are required to compare the algorithms using results such as overall accuracy and confusion matrices. The assignment expects students to identify an appropriate or best-performing algorithm in the context of the dataset and to provide a clear explanation of the comparison rather than simply reporting numerical results. The rubric places emphasis on accurate configuration, clear explanation of parameter choices, dataset analysis, class-balance treatment, parameter optimisation, cross-validation and critical comparison of algorithm strengths and weaknesses. High-quality work should explain both the technical process and the implications of the results, with results presented clearly through appropriate tables, confusion matrices and graphical outputs where required. The submission must be a single PDF document containing the report and must not exceed 10 pages. Students are instructed to include their student ID at the beginning of the report but not their name or other identifying details so that marking remains anonymous. Screenshots are specifically required to demonstrate use of the student's ID number as the random seed; other WEKA results should be presented in the student's own tables or result formats. The brief also states that no research beyond the material covered in the module is required and therefore no citations or reference list are required. The assignment explicitly prohibits the use of Generative AI tools for creating content and prohibits using GenAI tools or proofreading services for proofreading. Students are expected to complete the practical work themselves and explain their own technical choices and results.
Read Model Answer →
Principles of Data Science
2,000 words
Principles of Data Science – Data Analysis Portfolio
This portfolio assignment for the Principles of Data Science module at Coventry University requires students to analyse the Global Life-Work Balance Index 2025 dataset using statistical and data science techniques in R. The dataset ranks 60 countries according to life-work balance using factors including statutory annual leave, paid maternity leave, sick leave, healthcare, public safety, public happiness, LGBTQ inclusivity and average working hours per employee. The assignment has a 2,000-word equivalent limit, excluding the reference list and output. The portfolio consists of two main tasks. Task 1 is a group task involving multivariate data analysis. Students must use R to perform Principal Component Analysis (PCA) and Cluster Analysis on the dataset. For PCA, students analyse quantitative variables, produce and interpret relevant visualisations such as screeplots, biplots and loadings plots, and investigate the effects of Region and Healthcare System. The PCA analysis also requires comparison of the overall dataset with countries from Europe. The cluster analysis component requires students to cluster both countries and variables using different distance metrics and hierarchical clustering methods. Students compare methods such as Manhattan and Euclidean distances and single linkage and Ward’s method, present comparisons in compact tables, and interpret relevant dendrograms. They must then compare the conclusions obtained from PCA and Cluster Analysis, identifying common insights and apparent conflicts and discussing the extent to which the results are explainable rather than simply interpretable. Task 2 is an individual task focusing on Exploratory Data Analysis and Linear Models. Students create a scatter matrix using ggpairs(), investigate strongly correlated variables, and identify quantitative variables that may help predict Region for European and Asian countries. They then develop and critically assess linear regression models for predicting Score, including models based on employment variables and broader quantitative predictors. Model comparison and selection use concepts including AIC, while diagnostic plots are used to identify countries requiring further investigation. The individual task also requires students to use European Life-Work Balance Index 2023 data to make predictions for European countries not included in the 2025 dataset and to construct a Residuals versus Fitted Values plot. Finally, students must combine the conclusions from their individual linear modelling work with the PCA and Cluster Analysis findings to identify specific discoveries about the variables and countries in the dataset. R code, output and relevant plots must be included directly within the reports. The assignment encourages use of the R tidyverse and requires appropriate referencing of sources. The brief specifies APA-style referencing for the individual and group work. It also states that generative AI may be used for inspiration but not for generating answers or analysing the datasets, and any permitted AI use must be acknowledged and documented.
Read Model Answer →
Data Mining / Data Science
Customer Service Escalation Risk Analytics Using Data Mining and Machine Learning
This advanced Data Mining assessment applies the Knowledge Discovery in Databases (KDD) process to a real-world customer-service analytics scenario. Acting as a Data Scientist, students analyse a historical Customer Service Escalation Risk dataset to help an organisation reduce serious and legal customer escalations by identifying early signs of dissatisfaction, service bottlenecks and operational risk. The findings are intended to support business decisions relating to staffing, employee training, customer-journey improvement and escalation prevention. The dataset contains information covering customer demographics, account characteristics, communication channels, issue categories, operational measures such as wait times, transfers and SLA breaches, behavioural indicators including sentiment and response delays, and commercial variables such as monthly fees and contract value. The target variable contains four escalation outcomes: No escalation, Minor escalation, Serious escalation and Legal escalation. The first stage requires data exploration, visualisation and summary, including examination of variable distributions, dataset structure, descriptive characteristics and potential data-quality issues. Students then perform appropriate data cleaning, transformation, feature engineering and preprocessing. Particular attention must be given to variables that could introduce prediction leakage because of their meaning, timing or reliability. The supervised-learning component requires development and tuning of predictive models using suitable techniques such as k-nearest neighbours, Decision Trees, Support Vector Machines, ensembles or neural networks. Models must be evaluated using appropriate multiclass metrics and compared systematically, with interpretation of influential features and model behaviour. The assessment also requires unsupervised learning. After removing the escalation target, students apply and compare clustering approaches such as K-Means and hierarchical clustering. Appropriate preprocessing, encoding, normalisation or dimensionality reduction may be used, with visualisations such as PCA, t-SNE or scatterplots used to explore cluster structure and its relationship with escalation behaviour. Overall, the project assesses the student's ability to independently design a coherent KDD workflow, justify analytical decisions, compare alternative modelling approaches and communicate actionable findings to both technical and executive audiences. Overview word count: approximately 350 words. AI-use note: the brief permits AI tools only to assist with small, specific code snippets. Any AI-generated code must be clearly acknowledged and cited, while the submitted coding, analysis, interpretation and decision-making must remain the student's own work.
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 →