Chris Pollett > Students > Shenoy

    Print View



    [CS 297 Proposal]

    [Improve Q/A Patch]

    [Literature Review]

    [Hindi Part of Speech Tagger]

    [Refactor Code and support Localization]

    [CS 297 Report-PDF]

    [CS 298 Proposal]

    [Del 1: Triplet Extractor]

    [Del 2: Lexicon in Database]

    [Del 3: Best Answer at the Top]

    [Del 4: Add Named Entities]

    [CS 298 Report-PDF]

    [CS 298 Presentation-PDF]

Project Blog

Dec 05, 2017

  • Submitted a patch for Deliverable 5 and Reviewed the CS298 Report

Nov 28, 2017

  • Reviewed changes made for Deliverable 5

Nov 14, 2017

  • Discussed and started working on Deliverable 5

Nov 07, 2017

  • Reviewed patch for showing best answer at the top
  • Continued work on creating the index for Hindi, used steps mentioned in a video where Professor created an index for Canadian websites

Oct 17, 2017

  • Reviewed patch for triplet extraction and storing lexicon in database
  • Discussed how to test the changes after integration, how setup an index for Hindi specific urls

Oct 17, 2017

  • Submitted patch for triplet extraction and storing lexicon in database

Sep 05, 2017

  • Discussed progress on Deliverable 3 and 4, how to implement the code and extract the best answer from the possible answers

Aug 29, 2017

  • Discussed the plan of action for the semester and how to start with Deliverable 4 which is to present the best answer at the top

Aug 29, 2017

  • Discussed the plan of action for the semester

April 25, 2017

  • Worked on getting Question Answer System working using Unit Test cases. Tried to set up crawl, a limiting factor being how much index could be stored on local copy of Yioop

April 18, 2017

  • Worked on tweaking the triplet extractor code and testing it with the unit test cases

April 11, 2017

  • Worked on optimizing the data base insertion and retrieval of the lexicon faced issues in understanding Sqlite3 functionality as database insertion was failing

April 04, 2017

  • Noticed issues in the Suffix tree generation as not information was being sent to the tokenizer. Worked with Professor to write unit test case for Suffix Tree and found that although not completely correct it was returning some information which could be used by tokenizer.
  • Worked on storing the lexicon in the database

March 28, 2017

  • No Meeting Spring Break

March 21, 2017

  • Worked on trying to debug as to why the page description was null. As a workaround hardcoded a paragraph as the description and test Deliverable 1.

March 14, 2017

  • During the week from Mar 07 to March 13,created a patch for Deliverable 1 and worked on the changes required for Deliverable 2
  • Noticed some issues in parser while testing, the page description was returned was empty. Discussed this with Professor

March 07, 2017

  • During the week from Feb 28 to March 07, worked on the changes required for Deliverable 2
  • Discussed the table structure for the lexicon table and constraints on that table

Feb 28, 2017

  • Continued working in integration of the triplet extractor code with the Yioop code base
  • Asked the professor about about code changes required to change the Yioop database
  • Professor explained which files need to modified to implement the database change

Feb 21, 2017

  • During the week Feb 14 to Feb 21, implemented a CFG for Hindi, tried to integrate it with the Yioop code base but faced some compilation errors
  • Professor helped remove a compilation error in the code
  • Discussed how Yioop would distinguish between webpages for different languages

Feb 14, 2017

  • Discussed how to go about implementing a CFG for Hindi
  • Discussed about how the lexicon would be stored in the database.

Feb 07, 2017

  • Discussed the deliverables and project proposal for CS 298

Dec 08, 2016

  • Discussed with Professor the contents to include in the report and code refactoring deliverable

Nov 29, 2016

  • Demoed the Hindi Part of Speech Tagger and discussed code refactoring of code related to Question Answering System

Nov 22, 2016

  • Discussed with Professor the triplet extraction mechanism and Brill Tagger code in Yioop, for English

Nov 15, 2016

  • Professor explained on how the Question Answering mechanism works in Yioop

Nov 8, 2016

  • Discussed the paper on Prashnottar: Hindi Question Answer System
  • Discussed on difficulties in getting a Hindi lexicon in text format, continue working on the tokenizer for Hindi

Nov 1, 2016

Oct 25, 2016

  • Discussed changes needed to the Question Answer Patch uploaded on Mantis
  • Make localization changes to the patch
  • Work on tokenizer for Hindi and French

Oct 18, 2016

  • No Meeting

Oct 11, 2016

  • Presented the paper about gathering knowledge for a QA system
  • Work on code in the patch and align it with the coding guidelines
  • Read the paper: 'Multilingual Question Answering through Intermediate Translation: LCC's PowerAnswer' at QA@CLEF 2007

Oct 4, 2016

Sep 27, 2016

  • Discussed understanding of Conditionally maximal strings and suffix trees
  • Discussed what needs to be done to get existing Question Answer System patch to work with latest version of Yioop

Sep 20, 2016

  • Show understanding of what is conditionally maximal and try to understand Ukkonen's Algorithm
  • Get git to work on windows without Powershell

Sep 13, 2016

  • Continue reading about Question Answer System and come up with more concrete understanding of the same
  • Read and understand the Yioop Search Engine Ranking Mechanism
  • Read about the Coding Guidelines for Yioop
  • Understand how patches can be applied in Yioop. Create a dummy patch, apply it on Yioop.

Sep 6, 2016

  • Professor Pollett reviewed the CS 297 project proposal
  • Discussed the initial steps on how to understand the existing patch. We need to get the patch to work with the latest version of Yioop
  • Start reading about Question Answer System