This repo provides the code for our paper submitted to ECML-PKDD-2024: "MESS: Coarse-grained Modular Two-way Dialogue Entity Linking Framework"
conda create --name biel python=3.8
conda activate biel
pip install -r requirements.txt
conda install -c pytorch faiss-gpu cudatoolkit=11.0
These dataset (except BLINK data) are a pre-processed version of Phong Le and Ivan Titov (2018) data availabe here. BLINK data taken from here.
- BLINK train (9,000,000 lines, 11GiB)
- BLINK dev (10,000 lines, 13MiB)
- AIDA-YAGO2 train (18,448 lines, 56MiB)
- AIDA-YAGO2 dev (4,791 lines, 15MiB)
- ACE2004 (257 lines, 850KiB)
- AQUAINT (727 lines, 2.0MiB)
- AIDA-YAGO2 (4,485 lines, 14MiB)
- MSNBC (656 lines, 1.9MiB)
- WNED-CWEB (11,154 lines, 38MiB)
- WNED-WIKI (6,821 lines, 19MiB)
- WIKI-ABSTRACTS (6,221,563 lines, 5.1GiB)