Scribd
Upload
62 views

CSL361 Programming Assignment 2:: March 9, 2014

This document outlines programming assignments on latent semantic indexing and principal component analysis. For latent semantic indexing, students are asked to implement LSI in Matlab using SVD and QR, and demonstrate information retrieval and clustering. For principal component analysis, students are asked to develop a face recognition code in Matlab using PCA on a database of faces, experimenting with different rank approximations and reporting their findings.

Uploaded by

Jam Singh
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
62 views

CSL361 Programming Assignment 2:: March 9, 2014

This document outlines programming assignments on latent semantic indexing and principal component analysis. For latent semantic indexing, students are asked to implement LSI in Matlab using SVD and QR, and demonstrate information retrieval and clustering. For principal component analysis, students are asked to develop a face recognition code in Matlab using PCA on a database of faces, experimenting with different rank approximations and reporting their findings.

Uploaded by

Jam Singh
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 2

CSL361 Programming assignment 2:

March 9, 2014
Latent semantic indexing:
1. Study the articles on Latent Semantic Indexing available at
http://poorvi.cse.iitd.ernet.in/suban/csl361/LSI.
2. Make a list of common CSE and Mathematics text books that you
may have studied in your core and elective courses and a glossary
of standard CSE terms to generate a term-document matrix sim-
ilar to the examples in the articles.
3. Implement LSI in Matlab using both SVD and QR and demon-
strate information retrieval. Verify whether SVD gives better clus-
tering and noise reduction though the matrix approximation using
QR looks more similar to the original matrix. Also try to gen-
erate two dimensional plots (dierent cross-sections) of the terms
and documents with the queries to verify whether you indeed ob-
tain meaningful clusterings. Be warned that you may have to
tweak the weights a bit.
4. Generate a report explaining why the scheme works (if it works
at all).
Principal component analysis:
1. Study
(a) The PCA based face recognition paper by Turk and Pentland.
A local copy is available at
http://poorvi.cse.iitd.ernet.in/suban/csl361/PCA/jcn.pdf
1
(b) The Face recognition Demo page at
http://vismod.media.mit.edu/vismod/demos/facerec/
2. Show that in the projected space, the eigenfaces represent the
principal components, i.e., the directions corresponding to maxi-
mum variances in the training data.
3. Obtain a database of faces of people in the class. Obtain about
10 images of each person with normalized lighting and camera
settings. Please mount your camera on a tripod and make peo-
ple stand so that the faces are registered in a common frame.
Crop the faces interactively using any stand software (such as
http://www.gimp.org). Use 5 images of each person for building
the PCA database, and the remaining for testing.
4. Develop a piece of code (Matlab) for face recognition and experi-
ment. In particular, experiment with various low-rank approxima-
tions (5,10,20,30,...) of the covariance matrix and nd out where
do you obtain the best results. Give reasons for what you observe.
Display the average face and the eigenfaces. Also try to generate
some interesting new faces by taking dierent linear combinations
of the eigenfaces (e.g. Person1+Person2/2).
5. Write a report elaborating your ndings.
2

You might also like