MultirDevelopmentTimetable

Development Timetable

Green means DONE
Blue means IN PROGRESS
Red means TODO

10/28 - 11/1

1. Develop extensible Database Interface for Preprocessed Corpus Abstraction
2. Load Preprocessed Corpus Information into Database

11/4 - 11/8

1. Run Distant Supervision (Argument Identification and Relation Matching)
2. Run Feature Generation
3. Convert to Multir Input Format
4. Train Multir Model
5. Test Multir Model and Validate Performance Results

11/11 - 11/15

1. Convert Batch-mode preprocessing code to in-memory preprocessing
2. Resolve discrepancies in NER, POS tags, and dependency parses
3. Develop sentential extractor

11/18 - 11/22

1. Process KBP Corpus with Newer Models

11/25 - 11/27

1. Process KBP Corpus with Newer Models

12/2 - 12/6

1. Implement New MultirExtractor with New Models

12/9 - 12/14

1. Evaluate Sentential Extraction
2. Debug ADEPT implementation of Multir

12/16 - 12/20

1. Debug ADEPT implementation of Multir
2. Annotate Distant Supervision Errors for NER Argument Identification

12/30 - 1/3

1. Generate new Wikification Data from UIUC 2013 Wikifier in batch
2. Run UIUC 2013 Wikifier as Server for faster test time Argument Identification

1/6 - 1/10

1. Train a New Multir Model that uses old Wikification Data for Argument Identification

1/13 - 1/17

1. Implement Coref Data Into Multir
2. Implemented new manual evaluation scheme

1/21 - 1/24

1. Implemented Various Generalized Feature Generators
2. Chunked the Corpus with Hadoop

1/27 - 2/1

1. Evaluated Feature Generation
2. Generate new Wikification Data from UIUC 2013 Wikifier in batch

2/3 - 2/7

1. Generate new Wikification Data from UIUC 2013 Wikifier in batch

Longer Term

1. Move Corpus Representation to a SOLR Database as an infrastructure for Information Omnivore
2. Develop Web App