Enhancing NPF with state-of-the-art data visualisation techniques

Data visualization is a crucial area of study that enables the transformation of raw information into actionable knowledge. Currently, although researchers face no challenges in data collection thanks to the availability of various automation tools, there is a lack of progress in the development of data visualization tools. Consequently, we have created npf-web-extension as an accessible and interactive solution for visualizing data. During our research, we explored different architectures while adhering to our vision of offering a seamless and intuitive data visualization tool. Our solution had to be both modern and maintainable, as we aimed to create a tool that could be extended and serve as a foundation for future research and development. Given these circumstances, several design choices were made. On one hand, we built the entire software using React and Typescript to meet quality standards. On the other hand, we wrapped the tool with a Python package, enabling users to quickly get started with it. As a result, this combination yields software that is not only easy to install but also integrates smoothly into existing workflows, distinguishing it from other available solutions. Currently, our solution supports four different chart types: line chart, bar chart, box plot, and pie chart. It also allows the partitioning of complex data into subsets to create multiple charts from a single dataset. To validate our software, we conducted a comprehensive test suite to assess our code, and we also gathered feedback on the interface. This master thesis highlights some of the main difficulties when developing a data visualisation tool. Overall, users highly appreciated the interactivity and simplicity of the interface, while their suggestions for missing features were taken into account for our roadmap. Consequently, our solution strikes an excellent balance between ease of use and functionality.