Download the code from github To clone repository, use HTTPS link: https://github.com/George540/comp472_mini_project_3.git

This project experiments with word embeddings to solve a Synonym test automatically and compare it with a human standard. Other model comparisons are made, more specifically, models with different corpora and embedding sizes. Different metrics and results are graphed to better understand and analyze the fundamentals of word embedding

The proper functioning of this project requires several installations.

• Python 3.8.3 or equivalent compatibility (using pip or conda)
• pyplot (python library)
• numpy (python library)
• pandas (python library)
• gensin (python library)

1. Download the code from github
2. Open the project in your IDE
3. Run the file MP3.py . You will have to wait until all models are processed. Once complete, you will see a graph.

Note: the console will have a loading bar for each model as well

Pull requests will only be permitted by COMP 472 TAs as this is a private project.

Team Oranges

Ian Lopez

Columbus Feller

George Mavroeidis