Complexity-Prediction-SemEval2021

Repository for model training to predict lexical complexity prediction from https://sites.google.com/view/lcpsharedtask2021 for my final year thesis project.

Experiments Overview

This project use BERT, XLNet, RoBERTa, Word2Vec, and Fasttext to extract content from the text and token. Also, 33 word features fed to the model to increase performance