Understanding the MALLET lab

Screen Shot 2014-10-29 at 10.51.34 PM

1) Death: night poor happened de…

Death:

night poor happened death met terrible father heard mother dead creature sister died wild dreadful

Murder:

hand body found round blood dead head lay close shot revolver moment blow fell carried

Law:

police case inspector found law hands force evidence official arrest charge quietly court arrested constable

Relationship:

wife woman husband life left love knew child married heart made give lived loved happy

Time:

morning night hour work clock train past half time station late early breakfast morrow quarter

Finance:

set money business hundred week ten work began company earth position pay pounds thousand paid

Message:

word short letters words means make message single listened earth men expected true american game

Note:

paper note read hand book letter handed pocket written wrote writing put attention drew reading

Family:

father young wife son girl family returned boy child married mother poor died lived daughter

Expressions:

eyes face looked dark thin figure features tall voice lips expression pale turned spoke drawn

After using MALLET, I realized that it was a very easy tool to use. The interface was extremely user friendly and easy to work with. The word cloud above is comprised of my results from topic modeling. My most used words were child, death, and dead. I chose more morbid topics when choosing my ten to post.

Before coming across these 10 topics, i searched through 2000 iterations and chose three out of fifty for my first 3 topics. I then searched through 1000 iterations, and after, i searched through another 1000.

This word cloud helped to emphasize the occurrences of certain words in certain topics.

I chose topics based on my interest in the words that appeared. I always find myself fascinated with morbidity so of course I chose the topics associated with death etc.

In order to understand the underlying topics in Sherlock Holmes stories, topic modeling can be used to help understand what is going on at any given time.

The topics I worked on were very similar to eachother, so I wasnt very surprised to see that certain words were more prevalent than others, especially words having to do with death, etc.

Topic modeling is an incredible tool when trying to understand many works at once. The idea of distant reading is a brilliant way of analyzing thousands, or even millions of texts at once. Distant reading and topic modeling are both tools that I know I will use in my future life.

The reoccurring words that appear help to identify the theme of the story/stories you are reading, which is incredibly useful.

Advertisements

Leave a Reply

Please log in using one of these methods to post your comment:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s