Digital BookDocument visualization

Each line represents a whole paragraph.
Red signifies that there is at least one occurrence of the pattern (n-gram) in the paragraph.
The "time line" represents the exact number of occurrences.
It should be possible to "turn off" chapters and timelines one by one.
Overlaying a line should show an overview of the whole paragraph.
Select patterns that have an overall frequency > min_freq or length > min_length. This is intended to remove "noisy" patterns.
Group the patterns by length or frequency (alphabetic order is not useful as text search is available and it is not relevant for long patterns). Order by frequency or length within each category.
Filter patterns using the search box, will remove patterns that do not contain the word. Also, right-click on a pattern should only keep longer patterns that contain this pattern.
Use a meaningful subset of trends to filter patterns (no need to display the bottom-left widget, instead focus on the trend definition, analysis and retrieval).
Drag and drop patterns on the top window of overview assigns a color in the legend and display the time series.
Drag and drop a pattern on the header of a section in the middle overview window displays the vertical time series. If pattern is dropped on the background, display the vertical time series in every sections.
When a line is clicked, the right view displays current document and surrounding documents. Every documents is colored with selected patterns.
A text view can be duplicated so that its actual text and coloring can be compared with another view.Retrieved from "https://wiki.cs.umd.edu/cmsc734/index.php?title=Mockup" Pasted from <https://wiki.cs.umd.edu/cmsc734/index.php?title=Mockup>

Whole Interface

Workflow :