I 256: Applied Natural Language Processing

Course Information

 

Final Projects

 

Due dates:

 
Dec 1 and 3: Class Presentations (see schedule below)
Thursday  Dec 10: Final Project Write-up Due
 

Write-up

Due by end of day Thursday  Dec 10 -- NOTE: this is a hard deadline!

The write-up should be maximum 6 pages long and contain the following sections:

  1. Abstract
  2. Introduction
  3. Related work
  4. Data and Features
  5. Models and Results
  6. Conclusions
  7. References

The goal of the write-up is two-fold; explain the problem and your approach and give enough of the important information to allow  some else  to reproduce your experiments.

I hope you read some related work on the topic of your project; if you did, please include that in the write up, explaining similarity and differences.

In the results session, if appropriate, remember to include some baselines (chance and/or the accuracy of always choosing the most frequent class).

If you decide to make the data available, put the link.

If you need extra space for details/visualization of results, references, tables etc. you can add an appendix (but the paper should still readable/understandable in six pages)

(NOT required, but if you want to be really NLP-professional, you can use the Formatting Guidelines of the ACL proceedings and use the the style files -- Word and LaTeX-- that can be found here http://www.naaclhlt2009.org; this conference is a great starting point for related work hunt... Have a look at some of the papers to get a sense of how NLP work is typically described )

Send your final project description to Barbara  and Gopal as

LastNameOfstudent1_LastNameOfstudent2_project_final(.doc|.pdf|.html|.whatever)

with "i256 final project write-up" in the subject line

 

Class Presentations

Send the presentation  to Barbara and Gopal  as

LastNameOfstudent1_LastNameOfstudent2_project_final_presentation(.ppt|.pdf|.whatever)

with "i256 final project presentation" in the subject line

Due by 12 pm on the morning of your presentation (see schedule)

You'll have 10 minutes for your presentation  

We'll want to hear about:

  1. General ideas/introduction
  2. Data and Features
  3. Models and Results

 

Project on the course site

I would like to make your projects available from the course website (unless you have some objections to this, in which case let me know: it's an opt-out).

If I don't hear from you, I'll post the write-ups. If you would like to have the ppt or a specially prepared html file instead, let me know and send that along on  Dec 10.

 

Presentations schedule

Tue Dec 1

  1. Dispute Finder Michael Armbrust and Beth Trushkowsky

  2. Semantic Tremors Dan Byler

  3. Author Gender Analysis Chao-Yue Lai

  4. Noise / Fun Classifier Erin Knight, Ryan Greenberg

  5. Court opinion classifier Longhao Wang and Noah Kersey

  6. Twitter spam and trends Yo-Shang Cheng

  7. NLP-based Blog/Course Classification/Clustering Kentaro Suzuki and Hyunwoo Park

Thu Dec 3

  1. Emotional classification of twitter messages Connor Riley

  2. President speeches Sean Marimpietri

  3. Summarizing Public Opinion Sarah Van Wart and Annette Greiner,

  4. Extracting Character Relationships from Stories Krishna Janakiraman

  5. Doctor’s notes Abrahm Coffman, Nat Wharton

  6. Twitter Stock Value Predictor Ashley Kayler

  7. Poetry remixer Rachelle Annechino