• This repository contains the source code of my master's thesis. The thesis report is available at https://drive.google.com/file/d/15_Vi6RJ0Zy_m-Yln9d4LnBISkHVfXWW5/view .
• The details of the first three parts of the thesis are available in the Deep-BGT repository: https://github.com/deep-bgt/Deep-BGT.
• This repo contains the details of the representation learning methods used for VMWE identification. I compare character-level CNNs and character-level BiLSTM networks. Also, I analyze two different input schemes to represent morphological information using BiLSTM networks.
• I use the PARSEME VMWE Corpora Edition 1.1, the fastText word embeddings and the bigappy-unicrossy tagging scheme.
• The corpora cover 19 languages: Bulgarian (BG), German (DE), Greek (EL), English (EN), Spanish (ES), Basque (EU), Farsi (FA), French (FR), Hebrew (HE), Hindu (HI), Crotian (HR), Hungarian (HU), Italian (IT), Lithuanian (LT), Polish (PL), Portuguese (PT), Romanian (RO), Slovenian (SL), and Turkish (TR).

Setup with virtual environment (Python 3):

python3 -m venv vmwe_venv
source vmwe_venv/bin/activate
vmwe_venv/bin/pip3 install -r requirements.txt

      .
      ├── data
      |   ├── corpora
      |       ├── sharedtask-data-master
      |           └── 1.1
      ├── results
      |   ├── 01
      |       ├── ES
      |           └── 1
      ├── src
      └── Runner.py

python Runner.py -l BG -t gappy-crossy -cp data/sharedtask-data-master/1.1 -ep data/embeddings -op results -model 05 -exp 1