Skip to content

Create Combined Dataframe

This class combines two datasets: self-reported emotion data (like arousal and valence ratings) and movement data (from a smartwatch's accelerometer). It matches these two types of data based on the time they were recorded and saves the combined result in a new file. This is helpful for analyzing how emotions relate to movement patterns.

Step-by-Step Explanation:

  1. Input Data:
    • The class works with two datasets:
      • Self-Reports: A table where participants record their emotions (like arousal or valence) at specific times.
      • Movement Data: A table of movement information (accelerometer data), recorded at different times.
    • These datasets must be provided as inputs when using this class.
  2. Label Selection:
    • You can choose which emotions (labels) to include, like arousal and valence. If you don’t specify, it defaults to using these two labels.
  3. Filtering the Data:
    • It removes any rows in the self-report data where the labels are missing or invalid.
    • Ensures that the labels you want to analyze exist in the dataset.
  4. Matching Movement and Emotion Data:
    • For each self-reported emotion entry, the class looks for movement data recorded in a specific time range around that entry (called a "time window"). This links movement patterns with the emotions reported at a similar time.
  5. Combining the Data:
    • It creates a new table (or DataFrame) where each row includes:
      • Participant Information: Like their ID.
      • Emotion Labels: For example, arousal and valence ratings.
      • Movement Data: Accelerometer values (x, y, z) recorded at specific times.
    • Each row connects a participant's emotion and movement data based on the matching time window.
  6. Group Identifier:
    • A unique "group ID" is added to each combined entry to indicate which participant and self-report it belongs to.
  7. Saving the Results:
    • The combined data is saved as a file. The file name includes details like the size of the time window and which labels were used. This makes it easy to identify later.
  8. Friendly Messages:
    • The class shows messages at different stages (e.g., “Processing label: arousal” or “Combined dataframe exported successfully”), so you know what’s happening.