Skip to content

Latest commit

 

History

History
14 lines (12 loc) · 1011 Bytes

README.md

File metadata and controls

14 lines (12 loc) · 1011 Bytes

Time Expression Detection Using Soft Patterns

Sources

To understand the Soft Patterns model please refer to the following paper:

"SoPa: Bridging CNNs, RNNs, and Weighted Finite-State Machines" by Roy Schwartz, Sam Thomson and Noah A. Smith, ACL 2018

Here is the link to their GitHub

Data Sources

The data used for this project can be find here

The data added manually is in the files data/Guardian_time.txt and data/Financial_tine.txt

To create the datasets download the AQUAINT training data and run the notebook "parse_time.ipynb"

Embeddings

For the project we used glove-300 as pretrained embedding

To get the embeddings files: run the function "read_embeddings" from sopa-master/data.py