Timex3-Tagger - Documentation

Building a TIMEX3 Tag Identifier Model for Hindi

Info: This project was undertaken during the Advanced Summer School on NLP at IIIT-Hyderabad from 21st June to 7th July, 2018.

The project contains the data, scripts and the final results of the project.

File Details

DATA

Actual Log Files: 
Rule Based Log Files: 
ILTIMEX2012/BIO_TaggedM: 
ILTIMEX2012/Final_Tagged_Preprocessed: 
ILTIMEX2012/Manual_Tagged:
ILTIMEX2012/Plain_Text:

RESULTS

test_results_crf.txt : 

SCRIPTS

F1ScoreCalc.py: 
POS_tagging.py:
concatenate_articles.py:
data_script2.py:
extras/PivotTokens.txt:
extras/daysOfWeek.txt:
extras/festivals.txt:
extras/MonthOfYear.txt:
extras/timeLogs.txt:
pos:

How to Run

BIO, PIV, DAC Tagger:

datascript2.py

Parts of Speech Tagger:

POS_tagging.py

Concatenating Individual Document Feature sets (300 nos.) into a Single Document:

concatenate_articles.py

Rule Based Tagger:

Calculate F1 Score of Rule Based Model:

F1ScoreCalc.py