Our Progress Log

From Humanitarian FOSS Summer Institute

Week 2

• Learned about Support Vector Machines(SVMs)
• Worked with a Tutorial on SVMs written by Dr. Morelli to get comfortable working with lib-svm
• Had conference call with Nicolas from InSTEDD to get directions on what to do along with necessary materials

Week 3

• Wrote code necessary to test Reuters Data:
  • Parser to parse sgm code containing articles
  • Dictionary class to find list of all words and frequencies of each
  • InvFreq class to find inverse frequencies of each word
  • Vectorizer class to create vectors (using above classes) for each article
• Started testing Reuters data:
  • Training lib-svm with 1000 articles
  • Testing lib-svm with 1000 different articles
  • Analyzing test results and finding ways to refine code and tests

Week 4

• Finished debugging code for creating vectors from Reuters
• wrote ResultsAnalyzer class for obtaining sensitivity and specificity from libsvm results on Reuters
• Conducted LIBSVM-Reuters experiments to find the sensitivity and specificity of LIBSVM against the Reuters database

Week 5

• Put results and graphs from Reuters data onto Wiki (see Week 4)
• Obtained Pro-Med data to parse and test with lib-svm
• Wrote Parser and made minor changes to code for Pro-Med data
• Created vectors for Pro-Med data and tested the data with lib-svm
• Started looking at Android Tutorials and POSIT

Week 6

• Conducted LIBSVM-Promed experiments to find the sensitivity and specificity of LIBSVM against the Promed database.
• Learned about Naive Bayes Classifiers and talked about final project

Week 7

• Started working on writing a Bayes Classifier
• Wrote Bayes Parser and preliminary Classifier
• Met and talked with new consultant, Chris Walti, about final deliverable

Week 8

• Finished the Bayes Classifier
• Ran tests to find True Positives Vs. False Positives for the Classifier with different biases
• Javadoc-umented all code

Week 9

• Had conference calls and figured out what to do for our final project
  • Put all code together into one user-friendly application that would both parse and classify documents
  • Based off of an application written by Professor Morelli to cryptography

Week 10

• Wrote application and found a name for it: ALPACA Light Parsing And Classifying Application
• Presented our application and packaged it up