you need to use JAMA combined with Lucene, using the vectors that are
builded by lucene to compute SVD with JAMA
http://math.nist.gov/javanumerics/jama/
Best
jose
On 3/28/07, Mark Stiner <[EMAIL PROTECTED]> wrote:
Hi,
I have a research project where I want to implement LSI technique. The
scenario is something as follows.
Search
the news sites for the locally event based news. Cluster the similar
news items together. For example hurricane in New York city.
We want to apply basic LSI as follows
-Key word extraction
-Filter using stop list
-Stemming
-Option: Synonym detection
- Frequency Matrix
- SVD Decomposition
-Cluster related News items
The input data will be from the web based news sites such as Yahoo, google
etc.
How can we use Lucene to achieve this. Please provide me the steps.
Thanks.
Faikeeyes
____________________________________________________________________________________
Now that's room service! Choose from over 150,000 hotels
in 45,000 destinations on Yahoo! Travel to find your fit.
http://farechase.yahoo.com/promo-generic-14795097
--
José Ramón Pérez Agüera
Dept. de Ingeniería del Software e Inteligencia Artificial
Despacho 411 tlf. 913947599
Facultad de Informática
Universidad Complutense de Madrid
