Skip navigation.
Home

Visualizer

The visualizer program, built on the BioInfer basic classes framework, provides a graphical interface for viewing the corpus sentences and annotation types to help understand the different annotation types and how they come together to capture the information stated in the sentences.

The interface of the visualizer is shown below:

Class documentation

Detailed class documentation of the extractor in Epydoc format can be found from this address by selecting "visualize" from the Module Heirarchy.

Usage

The visualizer program can be run on Windows platforms with Python installed by double-clicking on the visualize.py program, and on UNIX platforms with

python visualize.py

After starting the visualizer program, the user interface shown above appears, with all of the views empty. The corpus data can be loaded into the visualizer by clicking on the "Open" button (top left) and selecting the corpus XML file in the "Open file" dialog that is opened.

When the corpus data is loaded into the visualizer, the user interface appears as shown above. It is divided into the following elements:

  • The topmost frame shows a tokenized form of the currently selected sentence with the dependencies in the selected syntactic annotation shown as arcs above the tokens. In this frame the selected entites are highlighted as blue rectangles appearing below the text they are connected to, and selected relationships similarly appear as green rectangles.
  • The middle-left frame shows a hierarchical view of the relationships annotated in the current sentence. The relationships and their arguments can be selected for highlighting by either clicking on them with the mouse or choosing one with the arrow keys and pressing enter to select. Relationships can be collapsed and uncollapsed in the hierarchical view with the left and right arrows, respectively.
  • The middle-right frame shows the entites annotated in the current sentence. Entities can be selected similarly as in the relationship frame.
  • The bottom frame contains a sentence selector. The current sentence can be changed by selecting another sentence in this frame either by clicking with the mouse or using the arrow keys and enter.

In all views, pressing shift when selecting an entity or relationship selects that entity or relationship without unselecting the previously selected entities and relationships. Selected entities and relationships are shown highlighted in all views.

The toolbar on the top of the screen contains buttons for the following functions:

Open
Open a new data file.
Help
Bring up a help screen
Quit
Quit the visualizer program
Go to
Jump to a sentence by entering a sentence number
Linkage type
Select the linkage type shown in the dependency frame

The linkage type selector can be used to view, for example, the basic ("raw") linkage containing the untyped dependencies as they were originally annotated, or different typed linkages where noun premodifiers have been expanded either serially or in parallel.