These files represent the results of training and running the summarizer over documents.
To view the main library's source code see summarize
This script will train the SVR over the document in train/doc.txt
over the summary train/sum.txt
The document used to train the model
The summary used to train the model
Files that store the trained machine
The script that reads a document and generates a summary. Usage : python generate.py folderName
.
In the rest of the folers, doc.txt
is the document amd sum.txt
is the generated summary.
The description of the documents is as follows
An article reviewing the Neuxs 5 phone. It consists of technical but non-scientific comments. Summary seems accurate
The introduction from Lowe's SIFT paper.The text is techinical, Summary seems accurate.
A review of the movie 300. Article is non technical, but is mostly descriptive.Summary is accurate.
A news article regarding the lost malaysian airline.The text mostly describes events. Summary is relevant, but not as good as others.
An article about Sachin Tendulkar. The text non descriptve, and tends more towards a story-like organization. Summary seems to list out some important events from the article.