Sadia Khan

– Conduct in-depth data profiling to identify structural patterns, anomalies, and quality issues, enabling targeted interventions. – Utilise tools including MS Power BI, Excel, Python (pandas), and SQL to execute data validation, transformation, and cleansing operations. – Led implementation of data quality and reporting improvements, improving data accuracy and reducing reporting errors through structured validation and transformation processes. – Rolled out interactive Power BI dashboards and reporting tools across teams, enabling improved decision-making and self-service analytics adoption.

– Conducted end-to-end analysis on customer churn data to identify drivers of attrition and improve retention strategies. – Cleaned and prepared data by handling missing values, encoding categorical features, and normalising numerical variables. – Performed exploratory data analysis (EDA) using correlation matrices, boxplots, and heatmaps to uncover key churn patterns. – Built a K-Means clustering model to segment customers by churn risk, providing actionable insights into customer behaviour. – Developed interactive Power BI dashboards visualising churn rates by tenure, spend, and contract type for stakeholder presentations. – Key Findings: Frequent support calls, payment delays, and short contracts strongly predicted churn. – Achievements: Improved churn prediction accuracy by identifying top behavioural indicators

– Conducted geospatial and temporal analysis of crime data to identify regional hotspots and temporal trends. – Cleaned and merged datasets from multiple public sources, standardising location formats and enriching with socio-economic indicators. – Performed EDA using heatmaps, time-series plots, and frequency analysis to uncover patterns by crime type, location, and time. – Built K-Means clustering models to detect high-risk zones and ARIMA models to forecast future crime incidents. – Designed interactive Power BI dashboards for local authority use and presented findings to public safety teams. – Key Findings: Theft and anti-social behaviour peaked on weekends; crime rates correlated with socio-economic deprivation. – Achievements: Identified high-risk areas enabling targeted policing recommendations

– Contributed to the development of machine-learning-based diagnostic tools, including hands-on work with data collection, cleaning, annotation, and model evaluation – Improved dataset accuracy by 5% and reduced data cleaning time by 15%. – Organised and delivered workshops and seminars while maintaining research records, managing datasets, and supporting day-to-day administrative functions to assist the research team.

– Conducted comprehensive online research and crafted data-backed reports to inform marketing and operational strategies. Managed data entry, cleaning, and quality assurance processes to maintain database accuracy. – Oversaw regular updates and validation workflows using Excel and cloud-based collaboration tools.

– Delivered undergraduate lectures in Python for data analysis, Excel-based analytics, and database systems, strengthening students' practical data skills. – Developed lab exercises and assessments, tracked student performance data, maintained academic records, and offered data-informed guidance to improve student outcomes.

– Developed data-driven dashboards using tools such as Excel to analyse fertiliser production, sales trends, and financial KPIs, providing actionable insights into company growth, profit margins, and loss patterns. – Participated in IT and networking department briefings and training sessions, gaining hands-on technical knowledge. – Supported analysis of data centre operations, gaining exposure to infrastructure systems and performance metrics.