Re: cse494 project t Task 2

[Sreelakshmi <slakshmi@asu.edu>]

If you are done with task 1, I am assuming you are able to get top 10 
documnets using vector space ranking given a query.
So now you get top 50 documnets instead of  10 and use these to apply LSI.

There is no source file that can be extended to do this in the code 
provided to you. You have to write your own program to do this.
- call the vector space model to get top 50 documnets
- Make the term-document matrix for this 50 documents
- use SVD method in the JAMA package ( for this you need to inlude the 
JAMA package in import statement)
- use LSI to rank the documents according to the query

> I'm still working on project. on Task 2 i don't what
> source files should i start implementing. Second
> question is how i can generate the top 50 docs using
> Vector Space Model for a givin query. Any help should
> be appreciated
>
>