Loading presentation...

Present Remotely

Send the link below via email or IM

Copy

Present to your audience

Start remote presentation

  • Invited audience members will follow you as you navigate and present
  • People invited to a presentation do not need a Prezi account
  • This link expires 10 minutes after you close the presentation
  • A maximum of 30 users can follow your presentation
  • Learn more about this feature in our knowledge base article

Do you really want to delete this prezi?

Neither you, nor the coeditors you shared it with will be able to recover it again.

DeleteCancel

Make your likes visible on Facebook?

Connect your Facebook account to Prezi and let your likes appear on your timeline.
You can change this under Settings & Account at any time.

No, thanks

Large-Scale Bayesian Logistic Regression G14

Presentation
by

warut leerasuntudkul

on 20 October 2011

Comments (0)

Please log in to add your comment.

Report abuse

Transcript of Large-Scale Bayesian Logistic Regression G14

Group 14 Warut Leerasuntudkul Qin Nathan CHEN Mohammad Asadi Karen Mansukhani Large-Scale Bayesian Logistic Regression
for Text Categorization Aims and main contributions Documents are represented as vectors of numeric values

Pure logistic regression has drawbacks.

Hence a Bayesian approach to logistic approach was applied.

Lasso Logistic Regression & Ridge Logistic Regression Structure Introduction
Use of supervised learning in language processing
Basics Bayesian approach to logistic regression
Fitting algorithm
Data sets and methods
Results
Future work Steps 1. Feature Selection 2. Text Categorization Algorithms 3. Testing Algorithms Used 1. Laplace Priors & Lasso Logistic Regression

2. Gaussian Priors & Ridge Logistic Regression

3. CLG Algorithm for Ridge Logistic Regression Testing 10-fold cross-validation
Calculation of F score Highlights Feature
Selection Feature Quality Measures Chi Square Test Binormal Separation Pearson Correlation Coefficient Threshold Default & Tuned Datasets Modapte
RCV1-v2
Ohsumed
WebKB
Newsgroups Highlights(cont') Drawbacks Limitation of Lasso Logistic Regression was not mentioned

Some alternative methods were not applied such as Meinshausen’s “relaxed lasso”

More classifications should be used and compared such as perceptrons. Summary - Lasso Logistic regression provides state of the art text categorization and produces sparse and efficient models

- Applications:
Personalised text classification
Sorting email
Spam filtering
Prediction of adverse drug events. Thank You
Full transcript