The Event Data Quality Dashboard (EDQD)

About:

The Event Data Quality Dashboard (EDQD) is a tool designed to help users evaluate the quality of event logs used in process mining and other data-driven analysis. Event logs are crucial for understanding processes within organizations, but the quality of these logs can significantly impact the accuracy and reliability of any insights derived from them.

Key Features of EDQD:

• Assessment of Data Quality Dimensions: EDQD evaluates event logs across five main dimensions of data quality:
  • Completeness: Ensures that all necessary data is recorded without omissions, such as missing values or events.
  • Accuracy: Checks for errors like duplicate events, incorrect timestamps, or traces that might belong to a different process.
  • Interpretability: Assesses whether the data is easy to understand and interpret, checking for issues like coarse timestamps or unclear activity names.
  • Relevancy: Identifies and flags events that are irrelevant or noise, which could skew the analysis.
  • Validity: Verifies that the data conforms to expected formats and types, ensuring logical consistency.
• Visual Representation: The dashboard provides a visual summary of the data quality assessment, with color-coded boxes representing each quality dimension. Green indicates high quality, yellow indicates some issues, and red indicates significant problems.
• Detailed Analysis: Users can click on each quality dimension box to view detailed information about the specific issues found, allowing them to understand and address the data quality problems effectively.
• User-Friendly Interface: The dashboard is designed to be intuitive, making it easy for users, even those without a technical background, to assess and improve the quality of their event logs.

In essence, the Event Data Quality Dashboard is a powerful tool for organizations looking to ensure the reliability of their process mining and data analysis by providing a clear, actionable overview of the quality of their event logs.

Running the code with Docker

Make a new directory and download the code to it. Navigate to the directory and run:

docker build -t edqd .

to build the image. After that you run:

docker run -p 5001:80 edqd

to expose the port 5001 for the container.

Now you can open a browser of your choice and access the dashboard with the url:

http://127.0.0.1:5001/