Key learnings from recent BDTI Skills Studio workshops

From geospatial analytics to dashboards for visualisation, each session provided hands-on experience with open data and open-source tools to solve real-world challenges and answer pertinent questions with data. Whether you missed the live sessions or want a recap, here are the key learnings from each webinar and why you won’t want to miss the upcoming ones.

1. Introduction to geospatial analytics

This session provided participants with a strong foundation in Geographic Information Systems (GIS), and explained how geospatial data is crucial in solving modern challenges such as climate change, urban planning, and public health.

What participants learned:

- The importance of GIS in understanding spatial relationships: Participants learned how GIS helps visualise and analyse data linked to geographical locations. They explored how to map patterns and trends, such as identifying public health risks or planning smarter urban developments.

- Geospatial data types: The session covered the two main types of geospatial data—vector and raster. Participants saw how vector data (points, lines, and polygons) is ideal for mapping precise locations like roads or city boundaries, while raster data is helpful for representing continuous data like elevation or temperature.

- Using the Geospatial Analytics Extension for KNIME: The hands-on portion introduced the Geospatial Analytics Extension, where participants learned to process and visualise geospatial data with no coding required. By working with crime data and population data from Eurostat, they mapped crime rates across European cities, identifying high-risk areas and trends over time.

Catch up on this session: Get the slides and watch the recording here.

2. Introduction to statistics

In this webinar, participants delved into the world of statistics, learning both the theory and practical applications of statistical analysis for making data-driven decisions.

What participants learned:

- Fundamentals of statistics: This included a deep dive into core concepts like descriptive statistics (mean, median, mode) and variability (range, standard deviation). These concepts were used to summarise large datasets, such as crime rates in various EU countries, helping participants understand how to describe and interpret data distributions.

- Probability and correlation: Participants were introduced to key probability concepts and calculations, learning how to assess the likelihood of different events occurring. They also explored how to measure relationships between variables, such as crime rates and employment survey data using correlation techniques.

- Hands-on with KNIME: Using KNIME’s low-code platform, participants ran statistical analyses without needing to write complex code. They learned how to build workflows that detect outliers, calculate correlation matrices, and test hypotheses, making statistical analysis.

Catch up on this session: Get the slides and watch the recording here.

3. Introduction to graph analytics

This session introduced participants to graph theory—a powerful way of representing and analysing relationships within complex datasets. Graph analytics is particularly useful for analysing networks, such as social relationships, transport systems, or city infrastructures.

What participants learned:

- Understanding graphs: Participants learned the fundamentals of graph theory, including the key concepts of nodes (vertices) and edges (connections). These structures were used to represent real-world networks, such as city-to-city transport routes or social connections between individuals.

- Why graph analytics is important: The session covered how graph analytics can reveal patterns that aren’t easily visible through traditional analysis methods. For example, participants explored how graph algorithms can predict new connections or uncover hidden patterns in transport data or social networks.

- Practical use case: The session featured a hands-on demo using KNIME to model a trip-planning scenario across Europe. Participants used graph analytics to optimise travel routes between cities, taking into account visa data and city distances to plan the most efficient routes.

Catch up on this session: Get the slides and watch the recording here.

4. Dashboards for data visualisation

In the final session, participants learned how to create interactive dashboards that transform raw data into clear, actionable insights. Dashboards are a crucial tool for communicating data trends and supporting decision-making in real time.

What participants learned:

- Automating workflows with Python: Python was used to automate data management and processing tasks. Participants wrote scripts to pull data into PostgreSQL, clean and transform it, and then make it available for dashboarding and geospatial visualisation. This integration of Python, PostgreSQL, and dashboard tools allowed for seamless data flow from back-end processing to front-end visualisation.

- Creating dashboards with Metabase: Metabase, an open-source dashboarding tool, was used to build real-time visualisations. Participants learned how to connect Metabase to PostgreSQL databases, allowing them to pull in live data and create dynamic charts and tables. They built dashboards that displayed traffic accidents in Madrid, and used interactive filters to explore different trends in the data.

- Geospatial visualisation with QGIS: Using QGIS, a leading open-source GIS tool, participants visualised geographic data on maps. They learned how to map accident hotspots in Madrid, overlaid with radar locations, to see where accidents occurred most frequently and what traffic enforcement measures might be needed.

Catch up on this session: Get the slides and watch the recording here.

Ready to take your skills to the next level?

Our upcoming webinars continue to build on these essential data skills. Whether you’re a beginner looking to get started or a seasoned professional wanting to refine your expertise, register today to secure your spot!

Upcoming workshops:

16 OCT 2024: Communicating complex datasets: Integrating real-time data for urban insights

06 NOV 2024: Harnessing climate data: Classification and predictive analytics for tourism

20 NOV 2024: Predictive modelling and real-time analysis: Real-time forecasting and monitoring of bicycle use