Job Market Analysis: Business Intelligence Roles
In this comprehensive job market analysis focused on business intelligence roles, I leveraged web scraping and data mining techniques using Python and Selenium to extract 2,149 relevant jobs with 1,554 unique titles across 748 locations from indeed.co.uk/.ch. The analysis provided key insights into the most sought-after skills, such as “dashboarding” and commonly required software proficiencies, including Excel, SQL, Python, Azure, and Power BI. I wrote a script to filter the 1,554 unique job titles into 16 job categories to streamline the sorting process. Visualisations, such as heatmaps and TF-IDF approaches, highlighted distinct skillsets for each role and allowed a comparison of median salaries and job postings. The analysis delivered valuable information on key competencies, salary trends, and regional distribution for business intelligence roles. This will be an ongoing project to explore evolving job demands and emerging technologies like machine learning and AI.
In this UK housing prices analysis, publicly available housing data from the Office of National Statistics (https://www.ons.gov.uk/) was explored to investigate changes in prices across regions over a 20-year period. Employing ETL, data analysis, and data visualisation techniques with Python (Pandas, Numpy, Matplotlib, Seaborn) and Power BI, the project transformed data types, converted dates into a year column, and merged datasets based on “County” and “Year” columns. Visualisations revealed a shift towards higher housing prices over time, with increased variability in 2021 data. Population density changes were examined, including a potential influence of the HS2 high-speed rail project. Linear regression analysis indicated a weak correlation between population density and average house prices. The analysis provided valuable insights into housing trends and influencing factors, enhancing the understanding of housing dynamics across different UK regions.
In this customer spending analysis project, the aim was to explore patterns and relationships among international customers using a publicly available dataset sourced from Kaggle. Utilising data analysis and visualisation techniques with Python (Pandas, Numpy, Matplotlib, Seaborn), the study examined eight columns of data without any null values. Key insights were obtained through Seaborn heatmaps, revealing intriguing spending patterns based on gender and education levels. Additionally, the analysis explored income and spending variations across education and age groups. The project also involved categorising countries into continents using the pycountry package. While some unexpected results were observed, such as females with a PhD earning more than males on average, it was noted that the dataset might have been generated for practice purposes rather than real-world insights. Nonetheless, the project provided valuable experience in working with diverse datasets and employing Python tools for meaningful visualisations.
Ormiston Analytics 