Filter by keyword, subject, or both. Updates live as new model answers are added to our portal.
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 →
Machine Learning / Artificial Intelligence and Data Science
1,000 words
End-to-End Machine Learning Model Development, Tuning and Evaluation
This Level 7 Machine Learning and Intelligent Agents assessment requires students to develop and document an end-to-end machine-learning solution, covering the complete workflow from data preparation through model training, tuning, testing and evaluation. Students select an appropriate dataset or scenario, formulate a research question and determine whether the problem is most appropriately addressed through supervised learning, unsupervised learning or reinforcement learning. Suitable machine-learning techniques must then be implemented to create a model that can be systematically trained and tested. The assignment requires students to follow a structured machine-learning development process and document the complete development journey. The report should explain the selected scenario, data collection or dataset, Exploratory Data Analysis (EDA), rationale for selecting particular machine-learning methods, model training, fine-tuning and evaluation. Model performance must be assessed using appropriate established metrics, with relevant published research used to justify methodological decisions and support the interpretation of results. The technical implementation should demonstrate the ability to identify the performance of machine-learning algorithms, implement machine-learning approaches using one or more object-oriented programming languages, and determine which algorithms are most appropriate for a particular analytical brief. These requirements directly correspond to the module learning outcomes relating to machine-learning performance, implementation and algorithm selection. Students are advised to document their work within a Jupyter Notebook, combining Markdown explanations with executable code. The notebook may be submitted directly or converted to PDF. Alternatively, students may prepare the 1,000-word report in Microsoft Word, provided that the Python code is included within the submitted document. Assessment is divided into three principal areas: Introduction (20 marks), Machine Learning Process (40 marks), and Evaluation of Model Performance (40 marks). Higher-level work is expected to demonstrate strong understanding of machine-learning concepts, a functioning and thoroughly tested implementation, appropriate selection of algorithms and critical evaluation of the developed solution. Overall, the assessment integrates research-question formulation, data exploration, algorithm selection, programming, model optimisation and evidence-based evaluation within a reproducible machine-learning workflow. All academic sources must be presented using Harvard referencing.
Read Model Answer →
Data Science / Artificial Intelligence and Machine Learning
2,500 words
Predicting ADHD Diagnosis Using Machine Learning and Explainable Data Science
This Data Science assessment requires students to develop a comprehensive analytical solution to a real-world healthcare prediction problem using the WiDS Datathon 2025 Health Outcomes Prediction Dataset. The dataset contains socio-demographic information, diagnostic variables and functional MRI data relating to children and adolescents, with the principal objective of developing predictive models for ADHD diagnosis. The assessment is designed to demonstrate the complete data-science lifecycle, from data preparation and exploratory analysis through predictive modelling, interpretation and evidence-based recommendations. Students begin by exploring the dataset's features, data types and distributions before addressing missing values, outliers and other inconsistencies. Appropriate feature engineering should be undertaken where necessary, followed by Exploratory Data Analysis (EDA) using relevant visualisations to identify relationships, patterns and correlations within the data. Students with limited computational resources may use a representative subset, provided that the sampling method preserves the integrity and distribution of the original dataset and is clearly justified. A major component of the assignment involves developing and comparing at least three classification models. Appropriate techniques may include Logistic Regression, Random Forest, Gradient Boosting and Neural Networks. Model performance should be evaluated using measures including accuracy, precision, recall, F1-score and ROC-AUC, allowing students to identify the strongest-performing model through systematic comparison. The assessment also requires model interpretation and explainability. Students should explain the results of the selected model and may apply techniques such as SHAP or LIME to investigate feature importance and individual predictions. A feature-importance visualisation must be produced, and the most influential variables should inform practical recommendations for healthcare professionals regarding the potential use of predictive modelling in supporting earlier ADHD diagnosis and intervention. Overall, the assignment integrates data cleaning, exploratory analytics, predictive modelling, model comparison, explainable AI and research-informed healthcare recommendations. Students must submit a comprehensive report of no more than 2,500 words, alongside a Jupyter Notebook containing the implementation and outputs. The report must use Harvard referencing, with appropriate academic research integrated into the analysis, recommendations and conclusion.
Read Model Answer →
Machine Learning / Data Science
Linear Regression and Stability of the Moore–Penrose Pseudoinverse Using Python
This machine learning practical and assessment activity develops an understanding of linear regression, Ordinary Least Squares and the Moore–Penrose pseudoinverse using Python. The work progresses from generating synthetic regression datasets to implementing regression algorithms manually, applying established machine learning libraries, analysing real datasets and evaluating the stability of estimated regression coefficients. The laboratory component begins with the generation of synthetic linear regression data using NumPy, including explanatory variables, random noise and an outcome variable. Students then implement simple linear regression without relying on machine learning libraries, using the least-squares solution to estimate the intercept and slope. The resulting observations and fitted regression line are visualised using Matplotlib. The work is subsequently extended to multiple linear regression, where several independent variables are used and coefficients are first calculated manually before the same problem is solved using Scikit-learn. The laboratory also introduces application of regression to the Scikit-learn Diabetes dataset, including feature and target standardisation, model fitting, prediction, correlation analysis and interpretation of regression coefficients. It also highlights the importance of residual analysis when assessing whether a linear model is appropriate. The associated weekly challenge focuses on the stability of linear regression solutions estimated using the Moore–Penrose pseudoinverse. Using a house-price dataset containing variables such as property size, number of bedrooms, distance from the city centre and property age, students construct the design matrix, standardise features and the response variable, and calculate regression coefficients using the pseudoinverse. Students then investigate model robustness by repeatedly fitting the regression model to random subsamples of different sizes and analysing the mean and standard deviation of each coefficient. Tables, boxplots or error-bar visualisations can be used to compare coefficient variability. The final discussion considers which variables are most influential, which coefficients are most stable, how sample size affects stability and whether coefficient interpretation remains reliable across different samples. The final work is submitted as a single PDF exported from Jupyter Notebook or Google Colab, combining documented Python code, experimental results, plots and written interpretation in a professionally organised notebook. Overview word count: approximately 360 wor
Read Model Answer →
Machine Learning / Artificial Intelligence and Data Science
1,011 words
End-to-End Machine Learning Model Development, Testing and Evaluation
This Level 7 Machine Learning and Intelligent Agents assignment requires students to develop and document an end-to-end machine-learning solution, covering the complete process from data preparation through model training, tuning, testing and evaluation. Students independently select a suitable dataset or scenario, formulate an appropriate research question and determine whether the problem should be addressed using supervised learning, unsupervised learning or reinforcement learning. Appropriate machine-learning algorithms must then be implemented to create a model capable of being trained and objectively tested. The assessment encourages the use of a structured machine-learning development methodology. Students are expected to explain the selected scenario and data source, perform suitable data preparation and Exploratory Data Analysis (EDA), and provide a reasoned justification for the machine-learning methods selected. The development process should demonstrate how the chosen algorithms are trained and fine-tuned before their performance is evaluated using established and relevant metrics. Published academic research should be incorporated to justify methodological choices and support the interpretation of results. The technical work is normally documented within a Jupyter Notebook, combining Markdown explanations with executable code cells. Alternatively, the report may be produced in Microsoft Word provided that the Python implementation is included. The assignment therefore assesses both conceptual understanding and practical programming competence. Students must demonstrate an ability to identify the performance of machine-learning algorithms, implement machine-learning techniques using an object-oriented programming language, and evaluate which algorithms are appropriate for a particular analytical brief. Assessment places particular emphasis on three areas: the Introduction, the Machine Learning Process, and the Evaluation of Model Performance. The marking criteria reward strong understanding of machine-learning concepts, a functioning and thoroughly tested implementation, appropriate selection of algorithms, and critical evaluation of the final solution. At the highest achievement level, implementations are expected to work without exception, satisfy the required functionality, demonstrate comprehensive testing and extend beyond the basic requirements. Overall, the assignment combines research-question formulation, data analysis, algorithm selection, machine-learning implementation, model optimisation and evidence-based evaluation within a reproducible technical workflow. All academic sources and supporting material must be presented using Harvard referencing.
Read Model Answer →
Data Science / Artificial Intelligence and Machine Learning
2,500 words
Predicting ADHD Diagnosis Using Machine Learning and Explainable Data Science
This Data Science assignment focuses on developing a comprehensive analytical solution to a real-world healthcare prediction problem. Using the WiDS Datathon 2025 Health Outcomes Prediction Dataset, students are required to analyse complex and high-dimensional healthcare data containing socio-demographic information, diagnostic variables and functional MRI data relating to children and adolescents. The principal predictive objective is to determine ADHD diagnosis from the available features. Students may use a representative subset of the dataset where computational resources are limited, provided that the sampling approach maintains the integrity and distribution of the original data and is appropriately justified. The assessment requires a complete data-science workflow beginning with data understanding and preprocessing. Students investigate the dataset's features, data types and distributions before addressing missing values, outliers and inconsistencies. Appropriate feature engineering should then be undertaken where it can improve the predictive capability of the models. Exploratory Data Analysis is used to identify important patterns, relationships and correlations, supported by relevant visualisations that communicate meaningful insights. A major component of the work involves the development and comparison of at least three classification models for predicting ADHD diagnosis. Suitable approaches may include Logistic Regression, Random Forest, Gradient Boosting and Neural Networks. Models are evaluated using performance measures including accuracy, precision, recall, F1-score and ROC-AUC, after which the most effective model is selected based on the evidence obtained. The assessment also places substantial emphasis on model interpretation and explainability. Students must interpret the selected model and may use approaches such as SHAP or LIME to explain feature importance and individual predictions. A feature-importance visualisation is required, and the most influential variables should inform practical recommendations. The final section translates analytical findings into recommendations for healthcare professionals, considering how predictive modelling could assist early ADHD diagnosis and intervention. Research literature must be integrated into the recommendations and conclusion. The assessment therefore combines preprocessing, exploratory analysis, predictive modelling, explainable AI and evidence-based healthcare decision-making within a single applied data-science project. The required report is a maximum of 2,500 words, with code, supplementary charts and tables permitted in appendices. A Jupyter Notebook containing the implementation and outputs is also required. Harvard referencing must be used throughout.
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 →
Data Science
1,000 words
Data Investigation Pipeline — Exploratory Analysis and Statistical Evaluation of a Chosen Dataset
This assessment simulates the opening stages of a real data investigation. Students choose their own research question and dataset, then build the full pipeline from raw data through preparation, exploration and statistical testing to visualisation — and, where the question supports it, simple modelling or forecasting. Either Python or R is acceptable; the statistical route through R typically expects an explicit hypothesis rather than a purely exploratory question. The work is structured around an established process methodology such as CRISP-DM, and the development journey is documented alongside the code rather than reported after the fact. The usual submission format is a single notebook combining markdown and code cells, so the written report and the analysis sit in one artefact, though a word-processed document containing the code is normally also accepted. The written element is short — around a thousand words — which makes selection the hardest part of the task. It must cover the scenario, the data collection, the exploratory analysis, the reasoning behind the choice of statistical tests, their results, and the visualisations. Students routinely spend that budget describing what they did and leave nothing for why. The mark distribution makes the priority explicit. Framing the problem and the data source carries the smallest share. Preparation and exploratory analysis, and evaluation of the results in context, carry the bulk in roughly equal measure. That final component is where most marks are lost: it asks for an honest assessment of accuracy, limitations and usefulness. A notebook that produces clean output and then claims more than the data supports scores below one that reports a modest result and explains precisely why it is modest. Established metrics should be used for the statistical tests, and published research cited where it informs the background or interprets the findings. Note that assessments of this type increasingly include a live demonstration in which the student explains their own project to verify authorship, so every line of the submitted work needs to be something the student can talk through unprompted. Our support on assessments of this type is guidance-based. Typical areas of help include: explaining how to scope a research question so the analysis fits the word limit, clarifying which statistical test suits which data type and why, reviewing whether a chosen visualisation communicates what it claims, showing how to write an honest limitations section, checking Harvard referencing, and reviewing a student's own draft notebook against the published marking criteria.
Read Model Answer →