Skip to content

Melvillian/TextVisualizer

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

58 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Visualizing PDF Sentiment

How to Run TextVisualizer

Assuming you have an update to date version of maven installed on your machine, while in the top level directory run

mvn package

Note that it took 2 minutes on my Macbook 2009 to complete the build. Once the build has finished, run the following line to start TextVisualizer

java -Xmx1G -cp target/textvisualizer-TextVisualizer.jar  edu.finalproj.TextVisualizer

Place the PDF you would like to visualize in the "pdfs" directory, and then load the PDF by entering pdfs/name_of_pdf. Choose which type of visualization method you would like. Note that the Positive vs. Negative method takes around 10 seconds to run, while the Sentence Parsing method takes up to a minute to run.

What is TextVisualizer?

TextVisualizer is a Java applet that allows a user to select a PDF on their computer and visualize its contents. Thus far TextVisualizer performs two types of visualizations:

  • Positive vs Negative Sentiment

    Using blue for positive and red for negative, each page in the PDF is mapped to a column in the visualizer. A page's sentiment is determined by summing the sentiment values for every phrase in a sentence according to the values in SentiWordNet.txt. Normally TextVisualizer will calculate a sentiment value for the page as a whole. If "Use Normalized Sentiment Values" is checked, then every 4 words within a page will receive its own coloring, allowing the user to see a text's sentiment at a much higher granularity.

  • Sentence Structure Parsing

    For each sentence in a page (by default the 2nd page) TextVisualizer paints the parse tree of the sentence using the parts of speech groups given in colored font. The idea behind this visualization method is to let the user identify patterns in a text's syntactic structure.

About

Mari Bennett and Alex Melville's final project for Natural Language Processing. TextVisualizer analyzes a body of text and visualizes the results in a Java Applet

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors

Languages