Datavisualisering

Detta är en Kandidat-uppsats från Lunds universitet/Högskoleingenjörsutbildning i datateknik

Sammanfattning: This work was carried out for the company Verifyter AB in Lund. Their main customers are microchip manufacturers. Verifyter provides a tool called PinDown that does automatic debugging on their customers' test runs. The tool creates a large amount of data with information about these test runs. The company wants to be able to visualize this data for their customers in order for their customers to be able to get better insight in how their projects are developing. This work broadly brings up what visualization is and why it is needed. We have then slected the best visualization techniques for the different kind of data that Verifyter wants to visualize, which is a line chart for sequential data and a filterable list for other types of data. To implement these charts different libraries were investigated to find the most suitable library for the visualizations that was to be impelemented. For this work D3.js in conjunction with NVD3.js was the best library to visualize the line chart while List.js was used to create the filterable list. This resulted in two different visualizations, a line chart and a list. The line chart visualizes a range of different data series and how they change over time. Some of these data series are for example project cost, project quality and project status. The list that was implemented shows how often files are occurring in failed tests. This list became very long and in order to make it easier for the user functions for filtering, sorting and seach were implemented. The implemented list as well as the chart meet the requirements and desires that Verifyter had on the product.

  HÄR KAN DU HÄMTA UPPSATSEN I FULLTEXT. (följ länken till nästa sida)